Slashdot Mirror


MS Says Windows 7 Will Run DirectX 10 On the CPU

arcticstoat writes "In what could be seen as an easy answer to the Vista-capable debacle, Microsoft has introduced a 'fully conformant software rasterizer' called WARP (Windows Advanced Rasterization Platform) 10, which does away with the need for a dedicated hardware 3D accelerator altogether. Microsoft says that WARP 10 will support all the features and precision requirements of Direct3D 10 and 10.1, as well as up to 8x multi-sampled anti-aliasing, anisotropic filtering and all optional texture formats. The minimum CPU spec needed is just 800MHz, and it doesn't even need MMX or SSE, although it will work much quicker on multi-core CPUs with SSE 4.1. Of course, software rendering on a single desktop CPU isn't going to be able to compete with decent dedicated 3D graphics cards when it comes to high-end games, but Microsoft has released some interesting benchmarks that show the system to be quicker than Intel's current integrated DirectX 10 graphics. Running Crysis at 800 x 600 with the lowest quality settings, an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics."

503 comments

  1. Oh boy. by Anonymous Coward · · Score: 2, Insightful

    So we can play things at 7fps with ultra low settings. Whoopee.

    Seriously, buy a goddamn graphics card.

    1. Re:Oh boy. by White+Flame · · Score: 5, Funny

      Seriously, buy a goddamn graphics card.

      I did, but then I only got 5fps. :-P

    2. Re:Oh boy. by jadedoto · · Score: 5, Funny

      But what if I want to play Crysis on my EeePC during that boring office meeting!?

    3. Re:Oh boy. by Mad+Merlin · · Score: 5, Funny

      But what if I want to play Crysis on my EeePC during that boring office meeting!?

      Your 8 core Core i7 EeePC?

    4. Re:Oh boy. by Anonymous Coward · · Score: 1, Funny

      Remote Desktop.

    5. Re:Oh boy. by davester666 · · Score: 1

      Didn't Microsoft redefine "works well" so that Vista could ship on product's with Intel's Integrated Graphics?

      And if it doesn't need any special graphics chips or vector instructions, why can't they make it work on my 486 box?

      --
      Sleep your way to a whiter smile...date a dentist!
    6. Re:Oh boy. by arkane1234 · · Score: 1

      For the same reason that Windows xp isn't terribly good on a 386. Sppppeeeeeeeeeeeed.
      The only thing that works good in todays world on a 486 is Minix, Linux, BSD, Windows 95, and DOS.

      --
      -- This space for lease, low setup fee, inquire within!
    7. Re:Oh boy. by arkane1234 · · Score: 1

      Amen. For Windows, probably the best idea.

      --
      -- This space for lease, low setup fee, inquire within!
    8. Re:Oh boy. by Joce640k · · Score: 4, Funny

      Eight cores at 3GHz beat one core at 400MHz!!!

      Film at eleven.

      --
      No sig today...
    9. Re:Oh boy. by RonnyJ · · Score: 1

      I think this is worth mentioning:

      Microsoft also points out that the technology could be useful for emulators and virtual environments that are attempting to display advanced 3D graphics.

      It may only get 7fps in Crysis, but I suspect that would be certainly more than enough for a virtual machine running an OS and/or games from a while back.

    10. Re:Oh boy. by Joce640k · · Score: 4, Interesting

      Just use the Intel ray tracer...

      http://video.google.com/videoplay?docid=7079133482718383307

      Much more impressive than "DX10 rasterizer".

      --
      No sig today...
    11. Re:Oh boy. by loonycyborg · · Score: 1

      s/Windows 95/Windows 3.1

    12. Re:Oh boy. by lyml · · Score: 2, Insightful

      I had windows 95 on my 486

    13. Re:Oh boy. by neumayr · · Score: 1

      I had it running on my 386 back then (was way behind with my computer technology..)
      Luckily, I discovered Slackware not long after that experience..

      --
      Truth arises more readily from error than from confusion. -Francis Bacon
    14. Re:Oh boy. by neumayr · · Score: 2, Interesting

      Looks interesting, yes.
      But what's this talk about a 3D internet in the Details box? How does this technology enable something like a 3D internet, when people's been distributing information in 2D forever and computer interfaces are build around that?

      --
      Truth arises more readily from error than from confusion. -Francis Bacon
    15. Re:Oh boy. by leuk_he · · Score: 4, Funny

      Nope, correct question would be:

      your eeepc with 800x600 resoultion?

    16. Re:Oh boy. by SupremoMan · · Score: 2, Interesting

      Mod parent up +1 nostalgia. I also had win95 on 486. It did the job, though I still ran most games in DOS mode. I was able to use win95 to compress part of my hard drive for added storage. Something I wouldn't be able to do without it.

    17. Re:Oh boy. by mike_sucks · · Score: 5, Funny

      Yes, the battery life is awesoNOCARRIER

      --
      -- "So, what's the deal with Auntie Gerschwitz et all?"
    18. Re:Oh boy. by Zelaron · · Score: 1

      Does this imply someone is finally going to purchase a Tesla now?

    19. Re:Oh boy. by smallfries · · Score: 1

      One core?

      --
      Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
    20. Re:Oh boy. by jonaskoelker · · Score: 4, Funny

      At one frame per meeting, you're at least better off than people who play Quake over email ;)

    21. Re:Oh boy. by Bert64 · · Score: 1

      When windows 95 came out a 486DX4/100 with 64mb ram was quite high end and ran comparatively well.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    22. Re:Oh boy. by RedK · · Score: 5, Informative

      Windows 95 didn't invent disk compression. Stacker and Doublespace were some products that did the same thing for DOS.

      --
      "Not to mention all the idiots who use words like boxen."
      Anonymous Coward on Monday August 04, @06:49PM
    23. Re:Oh boy. by Anonymous Coward · · Score: 5, Funny

      To: "John"
      Subject: Re: Quake

      Boom headshot!

    24. Re:Oh boy. by Dachannien · · Score: 5, Funny

      Much better frame rate than Halo over e-mail, where half the e-mails are spent teabagging each other.

    25. Re:Oh boy. by pm_rat_poison · · Score: 1

      Still, that didn't change the fact that it was Windows 95

    26. Re:Oh boy. by VagaStorm · · Score: 1

      "an eight-core Core i7 system managed an average frame rate of 7.36fps" seriously, I think you will get better performance bying a quadcore system with a decent graphics card.

    27. Re:Oh boy. by Spy+der+Mann · · Score: 3, Insightful

      I had been always confused by software advertising (especially Microsoft's). When they say it (i.e. a new version of Windows) would run faster than previous versions, I thought: "Hey! This will work great on my old computer!" - until I saw that the product requirements included the next generation of CPUs. WTF?

      Granted, it may be for some of the new CPU instructions that eliminated latency, but still, I felt kinda deceived.

    28. Re:Oh boy. by aliquis · · Score: 1

      octocore vs integrated graphics seems like a totally balanced system to me.

      And yes, a 100 dollar graphics card would totally smoke this.

      Funny how they speak about an 800 MHz to, Crysis at 0.1 fps on P3 anyone!?!

    29. Re:Oh boy. by aliquis · · Score: 1

      You fail if you buy a graphics card which is actually slower than Intels integrated graphics :D

      Turn in your geekcard.

    30. Re:Oh boy. by aliquis · · Score: 1

      ... unless you bought it for nostalgically reason and it runs on ISA, in that case it's all ok.

    31. Re:Oh boy. by aliquis · · Score: 1

      I like how much the complexity of the software increase. Computers still get around twice as fast in 18 months but solving many of the same old tasks take just as long time now as they did 20 years ago...

      Progress!

      To make the car analogy I guess it would be something like having a 200 hp engine but waste 90% on AC to get same old performance :D

    32. Re:Oh boy. by aliquis · · Score: 1

      Yeah, a DX 66 MHz was awesome, could even play 128 kbps mp3s in Winamp!

      C&C ran smooth to, supersonic on an 100 MHz one :D

    33. Re:Oh boy. by aliquis · · Score: 1

      7 fps isn't enough, 64 core would be something, but then what if he want to connect it to a real screen and raise the quality? 512 or 1024 core ... Microsoft really got something here.

      Play Crysis in 1680x1050 all high on future 1024 core machine! (Or get a decent graphic card today!)

    34. Re:Oh boy. by aliquis · · Score: 1

      Nah, makes much more sense to instead of coding for the cpu code for a virtual gpu running on the cpu!! That will show them who needs to upgrade and who doesn't!

      Why use raytracing when you can take advantage of the deaccelerated accelerated graphics!?

    35. Re:Oh boy. by aliquis · · Score: 1

      But then those games doesn't require DX 10 and would run just fine without that crap to. And they would run fastest if they just had a threaded software renderer from the beginning =P

      Quake on octocore anyone?

    36. Re:Oh boy. by aliquis · · Score: 1

      But obviously they believe that things like 8 core will be standard by the time Windows 7 is out, and they are probably correct. But then once again integrated graphics will also improve, and a shitty graphics card for 20 bucks will still own it, and a decent graphics card will totally smoke it.

    37. Re:Oh boy. by Gr8Apes · · Score: 1

      Should have spent the 2 minutes downloading InfoZip's Zip/Unzip and used that for your compression needs.

      And to add to the nostalgia, I was running OS/2 on my 64MB RAM 486 DX66. Along with the two 1 GB SCSI drives, that machine was a monster back in 91. :) Windows 95 shipped with "SmartDrive" (AKA - DumbDrive") which had this minor bug in it that would wipe the CMOS of the EISA motherboard in this machine.

      Ahh yes, good times....

      --
      The cesspool just got a check and balance.
    38. Re:Oh boy. by Midnight+Thunder · · Score: 1

      Seriously, buy a goddamn graphics card.

      And make sure it is not an "intel graphics card", which ends up being not much better than using your CPU - at least in my experience.

      --
      Jumpstart the tartan drive.
    39. Re:Oh boy. by DurendalMac · · Score: 1

      You'd get better performance buying a single core system from three years ago with a graphics card to match. This is laughable.

    40. Re:Oh boy. by modmans2ndcoming · · Score: 2, Insightful

      Perhaps if you would try some cognition before you type, this is more about running the interface graphics than it is about gaming graphics.

    41. Re:Oh boy. by pizzach · · Score: 1

      A better description would be it makes new hardware run faster. But if you are seeing that screen to begin with, you are most likely installing it on old hardware. It's a bit like a kick to the sack if you think about it.

      --
      Once you start despising the jerks, you become one.
    42. Re:Oh boy. by Kagura · · Score: 4, Funny

      Next thing you'll be telling me is Windows didn't invent the concept of OS.

    43. Re:Oh boy. by Anonymous Coward · · Score: 0

      Why, oh, why did I google the word "teabagging"...

    44. Re:Oh boy. by Belial6 · · Score: 4, Informative

      And in fact, DOS 6 also had disk compression. You could even see that they had illegally copied Stackers code because they forgot to take out Stackers copyright notice.

    45. Re:Oh boy. by Anonymous Coward · · Score: 0

      Yeah, Winamp was great but my LLama's ass was getting pretty sore so I had to change platforms. As unfortunately, the Linux client leaves a little to be desired.

    46. Re:Oh boy. by supernova_hq · · Score: 0, Flamebait

      1) Bob: Hey Joe, I bet I can drive faster than you
      2) Joe: Ok, I'll do 5 laps, then you do 5 laps
      * 5 laps latter *
      3) Bob: WTF? I can't drive that piece of shit!!!

    47. Re:Oh boy. by supernova_hq · · Score: 1

      8 core i7 EeePC...
      * blinks *
      * thinks *
      * ponders *
      * blinks *
      Well, I know what I'm building this Christmas!

    48. Re:Oh boy. by supernova_hq · · Score: 4, Insightful

      In every quake (except 4), all hit areas are equal.
      Please hand over your geek card.

    49. Re:Oh boy. by oakgrove · · Score: 4, Informative
      I would prescribe a healthy dose of Arch Linux for this problem you're having. I have an old Toshiba laptop laying around here that I had given up for dead. 600MHz Celeron, 192MB RAM, 12GB HDD. It came with Windows 2000 and was tolerable, I suppose. Only problem is, I don't know anything about Windows and none of my command line-fu worked on it so, off it went. I tried Ubuntu first which was horrible. Even with a lightweight window manager like IceWM and most of the unneeded services like bluetooth, etc. turned off, it bumped against 100 MB RAM doing nothing. Load Firefox with a couple of tabs (don't care for Opera and Konqueror needs more extensions to be useful for me), and it was over. Swap city. So, to get to the point, I tried Gentoo, and after waiting 7 hours for KDE to compile and then ending up with an error, I then wiped it in disgust.

      Enter Arch Linux. Installed to a CLI in about 10 minutes. Getting the wi-fi working from the cli with wpa_supplicant and the zd1211 firmware for my card was a breeze. Then I proceeded to download and install xorg and icewm. All told, at a cli with wi-fi working it idles at eleven MB. Logged in to icewm it sits at 17. And with firefox running, a grand total of 51 Megabytes. And of course, it's blazing. With Firefox 2, it's at least as fast as my Pentium 4 laptop running Debian with Firefox 3. And, of course everything works in Firefox. Flash 10, etc.

      Although what I've said doesn't speak completely to your point, suffice it to say, depending on your setup, you aren't doomed to a slower computer when running reasonably up to date software.

      --
      The soylentnews experiment has been a dismal failure.
    50. Re:Oh boy. by blincoln · · Score: 1

      But then those games doesn't require DX 10 and would run just fine without that crap to.

      I think you're missing the point.

      Current virtualization software supports 3D hardware either poorly or not at all. A full-featured software renderer might let games and other 3D apps from a few years or more ago run at full speed with the look of 3D acceleration on new hardware in a virtualized OS. That would be really handy for software that isn't compatible with newer versions of Windows.

      --
      "...always new atoms but always doing the same dance, remembering what the dance was yesterday." -Richard Feynman
    51. Re:Oh boy. by CaptainDefragged · · Score: 1

      This makes me want to go to my shed and retrieve the Compaq 486 dx 2/66 for a test run.

      --
      Don't tailgate - the end is near!
    52. Re:Oh boy. by Xenographic · · Score: 1

      > I had been always confused by software advertising (especially Microsoft's). When they say it (i.e. a new version of Windows) would run faster than previous versions, I thought: "Hey! This will work great on my old computer!" - until I saw that the product requirements included the next generation of CPUs. WTF?

      In other news, they make a ton of wild promises in order to generate news and get people to want their new releases. But the system requirements will only get higher. If it runs faster, it's only because the CPU designers have outpaced them.

      Frankly, I'd be more inclined to believe them if they said that I could play Duke Nukem Forever on my new Phantom console running Windows 7 with a WinFS file system ... It's a lot more likely than to suppose that this is anything but marketing nonsense that will get cut from the final version (because it never actually existed, except maybe as a demo they could show journalists).

    53. Re:Oh boy. by RobertM1968 · · Score: 0, Redundant

      So we can play things at 7fps with ultra low settings. Whoopee.

      Seriously, buy a goddamn graphics card.

      Why? When buying an eight core system is so much cheaper (than a new graphics card)? And I get an amazing 2.19fps!!! Oh... wait... nevermind, you're right.

      If this isn't the biggest waste of programming resources, I dont know what is (ok, maybe WinMe and MS Bob - but this is pretty close).

      Oh, wait... by doing this, they can "standardize" all the developers on DX10 since it will "run" on every system, and finally end all the games that are coming out still supporting DX9 (and finally end XP as the system of choice for gaming).

      This isnt about innovation. It's about trying to make it more appealing to game developers to finally drop XP/DX9 support so people are forced to get Windows7 to play the "latest and greatest" games.

      I for one would have preferred they truly work on their performance issues instead of working on something that eats even more system performance just to push their newest OS in the hopes of finally ending the demand for XP. Instead, they foist off some UI tweaks to make things appear to operate faster, and spend their efforts on this instead.

      Bravo Microsoft! I will run right out and buy that 8 core mobo with Intel integrated graphics!

      And really... what gamer out there worth their joystick doesnt go out and get a decent gaming card anyway? This truly isnt marketed or designed for the gamers - but for Microsoft via the gaming companies instead.

    54. Re:Oh boy. by man_of_mr_e · · Score: 1

      It's not really that confusing if you understand the tradeoffs for performance. Often times in order to get better performance, you need more resources (memory, typically, but also taking advantage of new CPU instructions or co-processors). So, it's quite possible that an old version of software will run slower on new hardware than a new piece of software, but the new software will run slower on the older hardware than the old software.

    55. Re:Oh boy. by aliquis · · Score: 1

      Thanks but I will never touch ArchLinux again since they just decided and switch their device system or whatever it and made not working alsa packages and shit like that.
      (Not important and no needs to explain it more, if it's not an issue for you then fine, but it was for me and it's enough for me to not use it.)

      If I ran Linux I'd go with something more stable such as Debian, but personally I would prefer FreeBSD over any Linux dist but have a mac now. And no, 2 GB ram isn't enough for Leopard.. ;D

      Impressive stats you mention, somewhat weird considering it's more or less the same software in the end.

    56. Re:Oh boy. by man_of_mr_e · · Score: 1

      And you can run Vista Aero in a VM..

      Now, if only someone would come up with a similar system for Linux to run Compiz/Beryl in a VM.

    57. Re:Oh boy. by spiderbitendeath · · Score: 1

      I kind of miss my 486 box. IBM PS/2 Model 77 486 DX2 66mhz, 48MB ram, full scsi and MCA, with a XGA/2 graphics card. System was a beast. Had Dr Dos running on that, with Windows 3.11 I eventually skipped over Windows 95 on it and went right to WinNT 4.0. Forced it to run Half-Life when that came out, got a low 19000 ping playing multiplayer over lan.

      --
      Sometimes when I'm working on projects things disappear, I suspect gremlins.
    58. Re:Oh boy. by amnezick · · Score: 0

      yes. unfortunately at 22khz becuase at 44khz it was just too damn slow. I know. I had one. And I burned it playing mechanic warrior. boy what a game..

      --
      mov ax,4c00h
      int 21h
    59. Re:Oh boy. by amnezick · · Score: 0

      flamebait or not but that's what causes the "problems" in the first place: numbers!

      Back in the XP days (please don't): you take a look in the task manager and you see 200-300MB out of 512 or 1024 and you say it sound ok.
      Vista days: same custom software installed and you see 500 out of 512 or 900 out of 1024 and you say oh boy vista is soooo hungry. There's no way that all that ram is filled with cache so when you click firefox it loads in under a second (it does for me).

      Same with ubuntu: i have (in the first 2 minutes since boot) ~1G out of 2G ram occupied. But i know it's preload and readahead trying to keep things smooth for me.

      So those 100 megs are there to keep you from getting old while firefox loads from your 5400rpm slow-as-a-turtle laptop hdd

      --
      mov ax,4c00h
      int 21h
    60. Re:Oh boy. by amnezick · · Score: 0

      unfortunately for you se7en only stands 256 cores

      --
      mov ax,4c00h
      int 21h
    61. Re:Oh boy. by petermgreen · · Score: 1

      I always thought that drivespace was a usefull tool and in more modern times i've been known to use the compression features in NTFS when i've been short on disk space.

      Zips are great for transferring stuff arround but less usefull for resolving shortages of disk space.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    62. Re:Oh boy. by LBt1st · · Score: 1

      As a developer this is great news. Even if it means something will run slow, at least I don't have to worry about support not being there at all.
      This means every Windows7 machine will support DirectX10. It's a step in the right direction.

    63. Re:Oh boy. by oakgrove · · Score: 2, Insightful
      While I understand what you are thinking, here is what's really going on. When Ubuntu was loaded on this thing, actual programs were chewing up that 100 MB I was talking about. And since, it had 192 MB of RAM, there were 92-ish left over for caching. Which, of course, Ubuntu was using. When I would start Firefox, and open a few tabs, the computer would reclaim all of that cache and more thus the swapping.

      With Arch installed, according to free, there are 17 MB being used by actual programs at the desktop with no extra programs being started by me. The other 175 are, of course, being used for cache. With Firefox going, 51 MB of RAM would be used. The rest being available, and being used, for cache. I get how this cache thing works. I understand that you want as much in cache as possible. I'm not using DOS. These are modern operating systems and that's how they work. I'm not saying "ZOMGWTF WHERE'S MAh RAMS!"

      Here is the crux of the matter. All things being equal, you still want the lightest footprint OS as it leaves more left over for cache. Case in point, Arch loads programs faster than Ubuntu on my machine. A lot of that has to do with the fact that there is more RAM left over for it to cache to. As far as Vista and XP are concerned, I'm going to go out on a limb here and say that the caching system/algorithm in Vista is better than the one for XP so although Vista itself may use up 400MB, leaving the other 612 MB in a 1 Gig system for caching, it still starts programs better than what you would get from the 900 MB that would be left over on the same system after a base install of XP. Of course, this leads us to one of the crucial differences between open and closed source OS's. Wouldn't it be great to just use the Vista caching system in XP? You could have the speed advantage of XP with the latency advantage of Vista. Only problem is, nobody except Microsoft is privy to the underlying code necessary to make backporting non-trivial Vista features to XP possible. With FOSS, this isn't the case. I can install a lightweigh fast distro like Arch and still use the latest latency reducing features like preload, for example, of any other version of Linux.

      I hope I wasn't unclear.

      --
      The soylentnews experiment has been a dismal failure.
    64. Re:Oh boy. by tzot · · Score: 0, Flamebait
      Oh, come on. Your first sentence is correct. The second sentence is false, since they bought the software, they didn't illegally copied it.

      If you intended it as a joke: don't mix jokes with serious and correct stuff (i.e. your first sentence), otherwise people will think you're just uninformed, like I did.

      --
      I speak England very best
    65. Re:Oh boy. by RobertM1968 · · Score: 1

      I had been always confused by software advertising (especially Microsoft's). When they say it (i.e. a new version of Windows) would run faster than previous versions, I thought: "Hey! This will work great on my old computer!" - until I saw that the product requirements included the next generation of CPUs. WTF?

      Granted, it may be for some of the new CPU instructions that eliminated latency, but still, I felt kinda deceived.

      Actually, though sometimes they misstate such (ok, maybe not sometimes), often it is the media pandering to them and extracting only portions of their statements.

      Vista on the newest top of the line hardware is indeed faster than XP on the now antiquated hardware it came on in the early 2000's.

      The problem being, Microsoft misleads by not doing an "honest" comparison... XP on state of the art hardware compared to Vista on the same hardware. Or XP on a 2001 box compared to Vista on the same box.

      The general non-Slashdot public simply dont know or dont care, as they cant walk into Walmart and buy a Vista machine built off 2001 equipment - so when it works at relatively comparable speeds, they dont see the big deal.

      It (to them) isn't like trading in their beat up 2001 car for a 2008 model that has twice the horsepower but somehow performs the same as their underpowered 2001 car. They don't make the connection in increase in hardware performance as it relates to how one would expect the operating system to perform.

      And for those on the edge of such perception, Microsoft handily has a list of "enhancements" that they expect such people to believe is the requirement for the unexpected negligible increases in performance.

      And for most computer users, the point is moot... "Will it still run Office and IE as well?" Yes? Then Microsoft has achieved the necessary goal to appease most of the masses. Now, if you are a server admin or a gamer, (or relatively technically savvy) the story changes.

      Sadly for Microsoft, the people who fit in one or more of those categories is slowly increasing, while the "performance enhancements" in new versions of Windows is decreasing - or nonexistant to begin with, while actual OS enhancements dont warrant the increased resource needs - slowly, more people are beginning to understanding that. The only thing keeping such operating system releases viable is the continuous effort on hardware manufacturers to create "better" and faster hardware with more resources available (speed, memory, etc). That's not a bad thing regardless of the OS. The sad (or bad, depending on how you look at it) thing is, each new iteration of their OS means less of those new resources/speed is available to power users as the OS itself consumes more of it.

      Keep in mind, many of Microsoft's performance "innovations" of late can be chalked up to the appearance of a performance increase (with no actual increase in performance), such as their UI tweaks to make menus and such load quicker, while the background work still takes the same time, or them redefining the term "boot" to mean "resume from S3/S4 state" (instead of cold boot), or them moving more and more service startups to after the GUI is on the screen making the machine appear to boot/"boot" faster when in fact the machine still isnt usable for the same amount of time.

      Yes, they have made some improvements in driver and service load times, but the fact is, those dont offset the increasingly large bulkly OS's requirements for total load to ready.

      Thus instead, they often emphasize the small gains a certain subsystem has in loading, while not pointing out that the total number of subsystems that need to load and the sheer size of the OS have both increased meaning that the net gain... well... isnt a gain, but a loss in speed/performance/resources available.

      ...

      And folks, as misleading as that is (and not saying I agree with such methods), that is simply called good marketing - and taking advantage of the fac

    66. Re:Oh boy. by w0mprat · · Score: 1

      Older OSes can't utilise newer cpu instructions and other hardware functions. On the same hardware, the older OS may be slower. Of course code optimised for newer versions 64-bit/SSE/MMX will do much more per clock cycle than old 386 compatible binaries. But there are other reasons - improvements to processor scheduling, threading etc that extract much more power out of the latest and greatest. However without the newer hardware functions available, the more sophisticated kernel will be just an overhead.

      So yes Microsoft(et al) claims are not untrue, but this is all behind the scenes and may be misleading.

      --
      After logging in slashdot still does not take you back to the page you were on. It's been that way for 20 years.
    67. Re:Oh boy. by nschubach · · Score: 2, Interesting

      Yeah, and I also had the "manual" way of doing it as well. We had an old DOS menu system that would run batch files for programs. I created a batch file to extract the program to be run, run it and on exit zip it back up. That was my solution to making the most of that 40MB hard drive. ;)

      --
      Every time I start to have faith in humanity, I ruin it by driving to work between 7 and 8 am.
    68. Re:Oh boy. by Belial6 · · Score: 3, Informative

      At the time that DOS 6.0 came out, and the Stacker copyright was still in the code, MS had NOT licensed anything from Stac Electronics. They did not buy the software. They did illegally copy it. So, no joke, AND not uninformed.

    69. Re:Oh boy. by aliquis · · Score: 1

      Ok, I thought it played at 44 kHz, I'm fairly sure, my Amiga 030 @ 50 MHz had to play them at 22 kHz mono. Maybe it was an 80 or 100 MHz in school then? =P, or your machine was doing something else :D

    70. Re:Oh boy. by Anonymous Coward · · Score: 0

      Tell Bob to drive around the track a couple more times and he'll come back as a paperclip.

    71. Re:Oh boy. by Gr8Apes · · Score: 1

      In today's world of <$100 TB drives, if you have a disk space issue, you're doing something I'm not, and neither is 99.999% of the other folks on here.

      --
      The cesspool just got a check and balance.
    72. Re:Oh boy. by Pathwalker · · Score: 2, Interesting

      There were lots of third party compression utilities before DOS 6.

      I used to use one called diet. It would intercept calls to read from files, and check to see if it had compressed them. If it had, it would unpack them to another location (I used a resizable ramdisk) and redirect the read to the uncompressed copy.

      When the file was closed, it would delete the decompressed copy.

      It would only work on read only files, but it worked pretty well. In the days before disk caching, uncompressing to the ramdisk actually made things faster despite the overhead of the decompression.

    73. Re:Oh boy. by speeDDemon+(nw) · · Score: 1

      I say you still need to hand in your geek card.

      Please point me to what '3d accelerator' that runs on ISA bus? My first Voodoo was a PCI card, even the old S3 Virge (minor 3D acceleration) were PCI.

    74. Re:Oh boy. by bursch-X · · Score: 1

      I used to include HURD in that list. Funny enough Stallman has outpaced DukeNukem Forever, so we might actually see a release very soon...

      --
      There are two rules for success:
      1. Never tell everything you know.
    75. Re:Oh boy. by mysidia · · Score: 1

      Windows 95 didn't invent disk compression. Stacker and Doublespace were some products that did the same thing for DOS.

      DOS 6 was the first DOS-based OS to include the compression feature straight from the OS vendor.

      Stacker and Doublespace were "hacks" in the form of TSR programs.

      I.e. They were providing filesystem facilities in replacement to the normal OS filesystem drivers.

      If there was enough demand for a third-party product to thrive, then surely there was enough demand for the feature that incorporating it in the OS would be an encouragement for customers to upgrade

      And get an official OS-supported compression capability.

      Rather than a third-party tool that applications are more likely to be incompatible with.

      i.e. There were programs that simply weren't compatible with your system if you had a Stacker-compressed drive.

      At least with Windows '95 disk compression, fewer software applications were incompatible.

    76. Re:Oh boy. by aliquis · · Score: 1

      You fail.

      There are plenty of ISA cards which would be slower than integrated graphics, even if they do 3d acceleration or not, which they didn't.

    77. Re:Oh boy. by home-electro.com · · Score: 1

      "but Microsoft has released some interesting benchmarks that show the system to be quicker than Intel's current integrated DirectX 10 graphics. Running Crysis at 800 x 600 with the lowest quality settings, an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics."

      That's some interesting benchmark indeed.

    78. Re:Oh boy. by garphik · · Score: 1

      Thats why Consoles exist. Seriously, since Vista, windows has become really bulky for games.

    79. Re:Oh boy. by Count+Fenring · · Score: 1

      Wow. Nice to know I'm not the only one who had a 100mhz 486 chip. My parents bought the AMD, right when Pentiums were first coming out.

      I've been behind the curve ever since ;-)

      -------

      This comment posted from a 1.4ghz Athlon XP, the fastest computer I've ever owned.

    80. Re:Oh boy. by Count+Fenring · · Score: 1

      Also, emphasis on the "Was."

      I held on to the last 2-series release as long as I could, but man, everything after that was bad.

    81. Re:Oh boy. by symbolset · · Score: 1

      We would be remiss without mentioning the negative effects of using Stacker or Doublespace on an RLL hard drive.

      RLL = Run Length Limited compression. The drive itself is compressed, and reports available space based upon anticipated compressibility of future data.

      STacker - filesystem that compresses the data stored in a way that's not reducible by said hard drive.

      Hard drive crash is inevitable.

      --
      Help stamp out iliturcy.
    82. Re:Oh boy. by aiht · · Score: 1

      Hard drive crash is inevitable.

      I trust this is sarcasm, but in case someone doesn't know any better:
      symbolset*'s post is mixing two concepts: RLL and RLE.
      One is a compression scheme which stores a string of identical bytes as a single byte and a repeat count.
      The other is a family of bit storage or transmission schemes which avoids strings of identical bits to make synchronisation easier.
      I'll leave it as an exercise to the reader to guess which is which, and which one is actually used in hard drives.

      * nice id number, btw!

    83. Re:Oh boy. by hardwarefreak · · Score: 1

      This was also the first large lawsuit MS lost, to the tune of a $121 million jury award:

      http://en.wikipedia.org/wiki/Stac_Electronics

    84. Re:Oh boy. by negRo_slim · · Score: 1

      Yeah, a DX 66 MHz was awesome, could even play 128 kbps mp3s in Winamp!

      and the ftp sites you got them from were fast as hell

      --
      On the Oregon Cost born and raised, On the beach is where I spent most of my days
    85. Re:Oh boy. by tyrione · · Score: 1

      You're correct and it's during a time that Microsoft got away with felony business practices. People acted as if it was okay for Microsoft to steal and now that they are seeing their own fair share of pirating they are crying wolf.

    86. Re:Oh boy. by tyrione · · Score: 1

      He's probably referring to the Core speed of a standard GPU w/o the multiplier, running at 400Mhz.

    87. Re:Oh boy. by amnezick · · Score: 0

      it was doing windows 95 ... or maybe windows 95 was doing my computer.

      (i'm pretty sure the bus and soundcard had more influence back then)

      --
      mov ax,4c00h
      int 21h
    88. Re:Oh boy. by V!NCENT · · Score: 1

      Never believe in something that Microsoft says until they release it. Let's start of by asking ourselves where their Cloud OS has gone. Shouldn't that have been released my now? Oh wait...

      This is ofcourse bullshit: The fastest Quad Core x86-64 CPU doesn't have the power to play Crysis at all and everybody here knows that.

      It may be possible that NT6.1 comes with this but only if you can get 3D Aero to run on a system with Intel integrated graphics so that Micosoft could make Intel a little happier (Please excuse us for making NT6.0 such a resource hog that your integrated GPU can't process 3D Aero)

      --
      Here be signatures
    89. Re:Oh boy. by aliquis · · Score: 1

      I had an Amiga 1200 you insensitive clod .D

      But I guy in my class which we used to play C&C with in school had a small LAN at home where we also played C&C, I think he had a couple of pentiums and the 100 MHz one. I think the school had 133 MHz Pentiums with 32 MB RAM, C&C ran very good on those, and one could play Quake :D

      (The usual 486 we used was 25 or 33 MHz and probably had 16 MB ram, 210 MB harddrives or such and sometimes some morons had stored files on C: so our small C&C install on like 10-30 MB (I think it was possible to get it down towards 10MB eventually, but maybe that was archived) couldn't fit. Solution? Deltree c:\windows

      Year, probably very appreciated by the net admins :D

      Respect for school stuff on said PCs was zero since no-one respected or accepted our gaming :D

      Time of my life :/

    90. Re:Oh boy. by aliquis · · Score: 1

      I think the school had an impressive 2.5 mbps, on the 1000+ students?

      I think the web was pretty fast at first, but got slower with time. 25 kbps = ok, 2.5kbps = not so ok.

      Ah, the days without Flash, AJAX and messed up stylesheets, how I miss you.

    91. Re:Oh boy. by ifrag · · Score: 1

      You might be right about Vanilla Quake. I'd guess there are some mods that used hit areas prior to Q4.

      --
      Fear is the mind killer.
    92. Re:Oh boy. by 0racle · · Score: 1

      mmmmmmmmm Vanilla Quake.

      --
      "I use a Mac because I'm just better than you are."
    93. Re:Oh boy. by bitrot42 · · Score: 1

      Yes, many people are doing something you are not. Try fitting multiple OSes on a laptop, when you have things like Vista sucking untold Gigabytes in the "winsxs" folder.

      I've even seen the "disk space is cheap, so why worry about it" argument on Vista developers' blogs, and it really chaps my hide. Even if I *do* have space for it, I'd rather use that space for something else.

       

      --
      FIXME: Add a sig here
    94. Re:Oh boy. by symbolset · · Score: 1

      Oops. Good catch.

      --
      Help stamp out iliturcy.
    95. Re:Oh boy. by petermgreen · · Score: 1

      Firstly not everyone has the latest and greatest PC, why go to the trouble of convincing your boss to approve an upgrade when you can just tell windows to compress files that haven't been used for a while (it'll take a while so best to set it going friday evening or so)

      Secondly while hard drive space for desktops is cheap things don't look so rosy for laptops for three reasons:
      1: laptop drives have a much higher cost per gig especially as you get to the high end of the range. For example dabs.com want £39.26 for a 500GB desktop drive but £98.88 for a 500GB laptop drive.
      2: any upgrade is a replacement further increasing the effective cost per gig of any upgrade (particually if you coun't the time to copy data across)
      3: noone makes a laptop drive over 500GB (and even those have only appeared on the market recently) this is tiny compared to a desktop where you can easilly have multiple 1.5TB drives

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    96. Re:Oh boy. by Anonymous Coward · · Score: 0

      get an external drive for the cost of a desktop drive plus $20 US for a case. It's not as dire as you make it out to be, unless you want it to reside inside the laptop case.

    97. Re:Oh boy. by Gr8Apes · · Score: 1

      First off, I have at least 5 OSes installed on my laptop. I don't run Vista. There's no reason. So I wouldn't know about it's space hogging ways. So with that out of the way, I run multiple copies of OSes for web development, some in virtual machines as needed, along with a full suite of developer tools and supporting applications. Oh, and I need to run the entire system and at least a single VM to see how things look in various flavors of IE while running in debug mode.

      So what, exactly, are you doing that I'm not other than running a single flavor of an OS?

      And yes, I have servers available also, but it's easier to debug locally for specialty scenarios.

      --
      The cesspool just got a check and balance.
    98. Re:Oh boy. by jonadab · · Score: 1

      Microsoft has included "makes your computer faster", or some variant of that, in the advertising for every new version of their consumer operating systems since, I believe, DOS 5, possibly longer. It has always been, objectively, a flat-out lie. To my knowledge no new version of Windows has ever made any computer faster (than with the previous version on the same hardware) in any way, for any purpose, in the history of computing, ever.

      (Unless, I suppose, you count the case wherein a hardware driver supporting hardware acceleration of some kind was available for the new version and not the old, but that would be fairly unusual; much more often the reverse is true for the first few months after a new OS release until the manufacturer gets its act together, and in some cases the driver for the old hardware *never* appears for the new OS.)

      So far as I can tell, nobody's ever been able to call them on it, perhaps because "faster" is (legally speaking) a somewhat subjective term (kind of like "better") and therefore may fall within the realm of "opinion", and when it comes to opinion advertisements can say pretty well whatever they like.

      But yeah, when Microsoft tells us something will be "faster", discerning users will not as a rule get very excited about this.

      --
      Cut that out, or I will ship you to Norilsk in a box.
    99. Re:Oh boy. by RobertM1968 · · Score: 1

      Someone should actually look at posting times before they decide which posts to mod redundant.

      Just a thought... ;-)

    100. Re:Oh boy. by Anonymous Coward · · Score: 0

      RLL = Run Length Limited compression. The drive itself is compressed

      Wow, symbolset, that's not what RLL coding is at all.

      The point of RLL coding is to guarantee minimum and maximum 0/1 transitions in arbitrary data.

      The minimum is hugely important when doing clock recovery at the receiving/reading end, like when a disk or tape flies underneath a read head: if your non-transitioning run is too long then you may mis-count your bits (for instance you may read binary 00000000000000000000000000000000 as 31 or 33 zero bits if your local oscillator is slightly slow or fast compared to the writer's). Limiting the runs of same-value bits ("run length limiting") gives you a predictable 0/1 or 1/0 transition interval, which you can use to calibrate your reading clock.

      The maximum is also important when dealing with a high transition frequency (like data that is 0101010101010...) which can suffer from noise.

      RLL coding has nothing at all to do with data compression, and everything to do with signal modulation.

      See here for details or try a google search.

      To be generous, the RLL / RLE acronyms are pretty close. You were probably thinking of RLE, which AFAIK has never been implemented in actual hard disk drive units' hardware, because it is unpredictable.

      RLL: makes 0/1 transition density predictable
      RLE: makes data length unpredictable, but probably shorter

      RLE on RLL data, incidentally, does not compress very well. :-)

      RLL on RLE data has the same 0/1 density as RLL on the data fed into the RLE compression in the first place. Which is the point.

    101. Re:Oh boy. by symbolset · · Score: 1

      Yeah, that was sloppy, as I told the last corrector. Thanks for the catch. The two are linked in my mind because I lost all my Huffman RLE source code in an Stacked RLL drive crash.

      Sloppy thinking. I must be getting old. Something good came of it though - I switched to implementing LZW, which is much better and was more educational as well.

      --
      Help stamp out iliturcy.
    102. Re:Oh boy. by Anonymous Coward · · Score: 0

      Wow. I'm sorry if I came across as harsh. I am getting old too, and end up with a GETOFFAMYLAWN reflex when people seem to be spouting off authoritatively about things they appear to have little experience with.

      If it's any consolation, RLL fell out of favour in telecomms compared to 8b/10b and 2-bit-difference variants (64b/66b) in the past decade as single bit times have shortened below low numbers of nanoseconds. On flypast media like disk drives, however, one now sees coding with EPRML or a Reed-Solomon like error correcting code.

      The two codings have diverged substantially because of the differences in sources of error at the physical reader, and are likely to diverge still more as telecomms continues down the path of RZ-DPSK type modulations of infrared light (since a delay line interferometer can be used on the receive side to combine bits in pairs or larger groups for a photodetector that can produce multiple levels of output (rather than just "on" or "off"), while bulk online storage manufacturers get deeper into magnetic couplings (GMR) and even spin-orbit interactions, which increases the reader's sensitivity.

      The optical-only approach deliberately adjusts to the transition times of the receiver (in the optical->electrical sense of receiving) being relatively long but not necessarily constrained to small numbers of output signals, while the greater sensitivity approach aims for producing much smaller differences in the receiver so that it can adjust to transitions faster (transforming small, gently-sloped physical transitions into sharp-edged electrical ones). Unfortunately the optical-only approach is highly constrained by the frequencies existing optical fibres are transparent to. The wavelengths at those frequencies are quite long, and the mean photon->electron interaction time is closely related to the wavelength (by a large factor with current semiconductors' material properties) for physical (quantum electrodynamics) reasons. The transition time on fly-past materials is bounded by grain size primarily, and secondarily by energy considerations in the maximum angular velocity of rotating disks, since (speaking very approximately) the interactions are mainly in the limit of the near field (r << \lambda).

      By comparison, you probably know better than me about how LZW and DEFLATE have been superseded. (Which is also hugely interesting in the delay-state-bandwidth-energy computing relationship).

    103. Re:Oh boy. by symbolset · · Score: 1

      a photodetector that can produce multiple levels of output

      I had wondered how long it would take for this to be investigated. I thought it would be a while yet, as there's still a lot of dark fiber out there. Thanks for the update. As we move back and forth between analog and binary quanta it amazes me how swiftly entire fields switch, seemingly overnight, from the one to the other and back again. I suppose one day we'll be communicating with real quanta and we'll be stuck in binary for a while with single photon emitter/detector pairs. And then someone will figure out that time is analog and we'll multiply the bandwidth again with time domain encoding.

      By comparison, you probably know better than me about how LZW and DEFLATE have been superseded. (Which is also hugely interesting in the delay-state-bandwidth-energy computing relationship).

      Of course, it was a long time ago. I haven't seen an RLL drive for 20 years. LZW has lost the crown on compression size but as a lightweight algorithm it does still have some appeal, and the fact that the patents have expired makes it more interesting. You are quite correct that LZW is an important lesson in both programming and information theory. I feel implementing it from the original ACM papers would be a useful lesson for both math majors and comp-sci majors. Definitely worth a few days of a 300 level class.

      --
      Help stamp out iliturcy.
    104. Re:Oh boy. by brilanon · · Score: 1

      > Didn't Microsoft redefine "works well"

      Yes!

  2. Software rendering by Anonymous Coward · · Score: 5, Funny

    What a revolutionary & useful idea.

    1. Re:Software rendering by Hal_Porter · · Score: 1

      It's not too useful at the moment but I could imagine Intel adding a bunch of instructions to next generation processors, beefing up the embedded graphics, load balancing between the two and end with something 10x faster. And at that point they've got something good enough for all but the hardcore gamers. And they have another good reason for people to buy fast processors.

      As a viable product it's useless. As a starting point for optimisation, given that they have time to rev the processor and graphics chip before Windows 7, it's interesting.

      --
      echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
    2. Re:Software rendering by Anonymous Coward · · Score: 0

      "I could imagine Intel adding a bunch of instructions to next generation processors"

      Intel Larrabee...
      http://en.wikipedia.org/wiki/Larrabee_(GPU)

      Based on all current specs on Larrabee, it won't compete with high end GPUs, but its enough to draw Windows.

    3. Re:Software rendering by Bert64 · · Score: 1

      For all but gamers and people doing heavy graphics work an old 2d-only videocard is more than adequate.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    4. Re:Software rendering by Anonymous Coward · · Score: 0

      Please use proper wording. This is called innovation.

    5. Re:Software rendering by AmberBlackCat · · Score: 1

      Very funny. Now think about how people who want to play the latest DirectX games can just spend their money on a second processor or more cores instead of buying a graphics card and bitching about how the drivers are terrible. Think about how, if OpenGL did this, there would be complete OpenGL support on Linux instead of just support for whatever cards got reverse engineered decently. And finally, think about how much this will push the graphics card manufacturers to actually update their drivers since they are now competing with hardware that has built-in support from the operating system.

    6. Re:Software rendering by hardwarefreak · · Score: 1

      This is a result of the class action lawsuit over "Vista Ready", nothing more. Low end systems are getting newer faster CPUs at a rate greater than the increased performance of low end integrated GPUs.

      This is MS's effort to get a "Windows 7 Ready" logo on just about every PC without fear of more class action lawsuits. Call it hedging their bets. They believe that even low end multi-core CPUs are a better way to get decent Aero performance than cheap integrated GPUs.

  3. Yes. by James_Duncan8181 · · Score: 5, Insightful

    In other news, Intel graphics chips said to be designed for minimal power draw rather than all out performance. This power draw is decidedly not beaten by running a software renderer that will stress the CPU till it sucks power like an electric chair as the CPU is only general hardware, not specific. More at 11.

    --
    "To any truly impartial person, it would be obvious that I am right."
    1. Re:Yes. by GigaplexNZ · · Score: 5, Insightful

      Servers are plugged in at all times, and we still want minimal power draw to save money and heat output (and for people who care, the environment). It isn't just about battery life.

    2. Re:Yes. by afidel · · Score: 5, Insightful

      Yes, but running something like a 9600GSO will require less power than pushing 8 cores on the Core i7! The TDP on the Core i7 is 130W, my 9600GSO has a max power draw of 65W. Not only that but you can get PLAYABLE framerates, like 30fps@1080P.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    3. Re:Yes. by GigaplexNZ · · Score: 1

      That is true, but that has nothing to do with being plugged in to the wall. That is being more efficient.

    4. Re:Yes. by Anonymous Coward · · Score: 4, Funny

      Running DX10 games on servers? Get back to work you lazy servermonkey!

    5. Re:Yes. by RonnyJ · · Score: 1

      Intel graphics chips said to be designed for minimal power draw rather than all out performance

      I somehow doubt Intel's key design goal is minimal power draw - cheap mass production is likely top of the bill.

      As an example, just look at the Intel Atom - the CPU is designed to use a tiny amount of power, but the Intel graphics chipset it's usually integrated with uses a lot more power (relatively speaking) than the CPU and other chipsets out there.

    6. Re:Yes. by MadMidnightBomber · · Score: 5, Insightful

      Aye; "wannabe computer companies worry about clock speed. Real computer companies worry about cooling."

      --
      "It doesn't cost enough, and it makes too much sense."
    7. Re:Yes. by msormune · · Score: 1

      Do you have any idea how little for example newer generation Intel CPUs draw power, when compared to their earlier chips? 33 watts for E8400 on full blast. Yeah, try that for an electric chair. It warms nicely, but does not really fry the bacon.

    8. Re:Yes. by ultranova · · Score: 1

      Well, word to Intel: You can plug laptops into power outlets now.

      So you can put the performance back in under those situations!

      But what about the fuses ? I mean, a single copper wire can only carry so much current, and a cherry red glow is kind of distracting.

      Maybe computers should come with branching power cords, so they could be plugged into multiple electric sockets simultaneously, to balance the draw ?

      --

      Forget magic. Any technology distinguishable from divine power is insufficiently advanced.

    9. Re:Yes. by msormune · · Score: 1

      My mistake, it draw 65 watts... Still that's quite low.

    10. Re:Yes. by frdmfghtr · · Score: 1

      Maybe computers should come with branching power cords, so they could be plugged into multiple electric sockets simultaneously, to balance the draw ?

      Now why would you do that? Too many cords. Just get a two-phase 220 vac outlet in the computer room and you're good to go. The side benefit is that you also have a power socket for the portable welder when installing/repairing the copper piping that supplies cooling water.

      --
      Government's idea of a balanced budget: take money from the right pocket to balance...oh who am I kidding?
    11. Re:Yes. by Targon · · Score: 2, Informative

      If you have people running DX 10 games on a server, then you either have major staffing problems, or the server is nothing more than a glorified workstation. There is ZERO need for DX 10 graphics on a true server, and really, the need for a GUI should be near zero(unless the server software vendor can't figure out how to code for a true server).

      Yes, there are good uses for having a GUI available on a server, but for normal maintenance, a command line SHOULD be all that is needed to reduce the overhead. the GUI places on the system.

    12. Re:Yes. by Bert64 · · Score: 1

      More importantly, you might actually want to use your CPU for other things....

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    13. Re:Yes. by Bert64 · · Score: 3, Interesting

      A GUI on a server should be entirely optional, and never the default...
      Serial consoles enable me to rebuild my servers without traveling to the location where they are hosted. Even if the OS is screwed to the point i can't login using it's existing remote logon features, i can get on via serial and fix it or do a complete reinstall.
      I've never found a need for a GUI on any of my servers, because everything i've ever needed to do was possible from the CLI. I would try to avoid any server software which required a GUI, as you pointed out poorly coded, and failing that i would install it locally and copy the configuration if possible. Having to install GUI libraries and the like would end up doubling the footprint on most of my machines, and therefore double the patching requirements.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    14. Re:Yes. by Anonymous Coward · · Score: 0

      Not compared to the MIPS that weighs in at about 12W.

    15. Re:Yes. by Spy+der+Mann · · Score: 1

      You make a very interesting point. I never wanted to buy a graphics card - mainly for the price, but also for its vampiric power suction.

      From an electricity bill point of view, it's bad when you buy a graphics card. Yes, I want the graphics to render faster, but couldn't they be only 4 or 5X faster than CPU-only and keep the same energy usage, instead of being 30X faster by sucking 10 times the energy usage of CPU-only?

    16. Re:Yes. by CableModemSniper · · Score: 1

      There is ZERO need for DX 10 graphics on a true server, and really, the need for a GUI should be near zero(unless the server software vendor can't figure out how to code for a true server).

      Render farm taking advantage of the GPU?

      --
      Why not fork?
    17. Re:Yes. by Kjella · · Score: 1

      You mean both wanna-bes that went for the 1GHz mark? Or just that little wanna-be corp that introduced the Pentium IV? And at the moment I'd say they're both very "real" companies. It's a lot less simpler in my opinion, chart the laptop market share vs attention to power consumption. When it shifted from power consumption == utility cost to power consumption == battery life, that's when everyone started paying attention. Because in the big scheme of things, I don't think running my computer is expensive at all. One trip back home with flight + airport transport costs me more than leaving a 200W system running 24/7. And that's without making any effort to save expenses, if I wanted to I could probably run an EEE Box PC @ 20W for 8 hours / day for 1/60th of the cost. In fact, my Internet bill is bigger than my power bill any way I count it so it's not going to be that cheap anyway. That said, I am planning to find myself a nice box to run accelerated video (VDPAU/XvBA) soon that'll probably become my primary computer / HTPC and leave this one for gaming. Noise is important too, and a fast CPU + fast GPU = noise.

      --
      Live today, because you never know what tomorrow brings
    18. Re:Yes. by Anonymous Coward · · Score: 0

      30fps isn't a playable framerate

    19. Re:Yes. by atrus · · Score: 1

      I'd suggest getting your house on a 3 phase feed, and run 208V everywhere. After all, its how datacenters do it :)

    20. Re:Yes. by Macman408 · · Score: 1

      That's only low until you have to design something like that. Or design its power supply, PCB layout, or heatsink. I designed a board once that had a 25W chip on it, and that had plenty of challenges. Of course, we had some extra restrictions that most systems don't have - extraordinarily limited space for a heatsink, another nearby heat source, very large ambient temperature range, and no fan allowed.

      The highest-power Pentium 4 was 115 Watts, which is quite frankly ridiculous. From what I can tell, the Prescott had a 122 mm^2 die; that puts it at a power density of 94 W/cm^2. The E8400's 107 mm^2 die comes in at about 61 W/cm^2. A slide I've seen in many places, originally attributed to Intel, pegs a nuclear reactor at 200 W/cm^2.

      With the P4, Intel realized that you eventually reach a point where it's not realistic to cool the CPU. Since then, basically everybody has started towards finding more ways to save power. After all, about half of those 65 Watts are from transistor leakage - the electricity is just wasted, not actually doing any work. As process technologies shrink, the leakage power increases. So, to be able to continue to cool our CPUs, they'll have to figure out ways to do more work with less energy.

    21. Re:Yes. by Macman408 · · Score: 1

      I think graphics chips are moving in the direction of saving power like CPUs have tended to do. For example, the MacBook Pro, which will run either the discrete GPU or the integrated one, depending on whether you prefer performance or battery life. Laptops will definitely go in this direction, because it's easier to reduce your power consumption than to increase the capacity of the battery without making it larger and heavier.

      Desktops graphics cards will get better at power savings too, but probably not as much, because it's something most people seem to care about - energy is readily available, cooling isn't too difficult, and people buy computers based on initial price and performance, not long-term costs like energy usage.

    22. Re:Yes. by Targon · · Score: 1

      GUI != GPU

      What I was saying is that a GUI adds CPU/GPU overhead to the system, and for a server, the resources would be better spent on what the server SHOULD be doing, being a server for clients.

    23. Re:Yes. by MadMidnightBomber · · Score: 1

      The GP was talking about servers. Talk to a DBA and they'll be demanding I/O throughput, not clock speed. Talk to your mail admin and he'll want the same. There are not very many compute-bound workloads out there, and a lot of the serious ones need clusters anyhow. We used to house one of the top 500 SCs in a machine room so I have a fair idea what I'm talking about. Short of CPU? Buy a rack of dual quad-core DL380s- they're commodity hardware these days. What you will have to worry about is rack space, power and cooling.

      --
      "It doesn't cost enough, and it makes too much sense."
    24. Re:Yes. by CableModemSniper · · Score: 1

      I was specifically addressing the "There is ZERO need for DX 10 graphics on a true server" statement. I should have been more surgical in my quoting.

      --
      Why not fork?
    25. Re:Yes. by badkarmadayaccount · · Score: 1

      This is OT, but did any body else see through intel's naming scheme? i786 Get it?

      --
      I know tobacco is bad for you, so I smoke weed with crack.
  4. Quickly, bash microsoft. by Anonymous Coward · · Score: 1, Insightful

    they've had an idea, those BASTARDS!

    1. Re:Quickly, bash microsoft. by Anonymous Coward · · Score: 2, Insightful

      Well in all fairness it's a pretty dumb idea. An 8 core CPU managed 7fps? Whoooopeeee!

      How about instead of wasting time on this, they work with vendors and get properly working drivers for the stand-alone graphics cards?

    2. Re:Quickly, bash microsoft. by RiotingPacifist · · Score: 5, Insightful

      How about the vendors learn to code and stop writing shitty drivers! I mean they have the full spec on the cards and still cant produce a driver as stable as some guys reverse engineering! Vista had a driver model ready for how long? Its not even like the change was unexpected.

      --
      IranAir Flight 655 never forget!
    3. Re:Quickly, bash microsoft. by wwahammy · · Score: 4, Interesting

      I wish I had mod points to use on the parent. The GPU companies (emphasis on Nvidia though) knew the Vista driver model 18 months prior to its release and they still couldn't come up with decent drivers on time or ever two years later. I finally gave up on Nvidia's shitty drivers when a driver update in June caused all AVI files to skip when emule was open. Combine that with Nvidia refusing to implement DVD anti-aliasing on hardware for Vista (something that they have in the XP drivers) I had had enough being a free beta tester for Nvidia. My new ATI card works just fine and I don't have to install additional crapware for its drivers. I don't plan on ever going back to Nvidia.

    4. Re:Quickly, bash microsoft. by Whiteox · · Score: 2, Insightful

      Nvidia and ATI and every 3D card company had serious problems with Vista - not the graphics, but the HDvideo implementation. You see, a Vista capable graphic card also has to be able to play HD Video.

      This has thoroughly been discussed by Peter Gutmann here

      Briefly, the fact that Vista was designed for 'premium content protection' caused long delays to sort out HD graphics driver issues as HD Video will not play unless the Audio can be unscrambled. It's still not fixed, but many companies have designed a work-around that stalls the content protection system. So much for MS OS design, pressure from the RIAA and MPAA.
      Gutmann is very clear on how Vista's design had stuffed up 3D hardware and driver design. A good read.

      --
      Don't be apathetic. Procrastinate!
    5. Re:Quickly, bash microsoft. by LingNoi · · Score: 4, Interesting

      Isn't this the point of openGL? An API to dedicated graphics hardware with a backup software renderer if the hardware isn't supported?

      Whose idea is this again? It doesn't look like much of an idea, more like a step backwards..

    6. Re:Quickly, bash microsoft. by Elrond,+Duke+of+URL · · Score: 4, Insightful

      And it's not just the GPU companies. Creative took their sweet time releasing Vista drivers for their previous generation of audio cards. I believe they were actually released after Vista was, and they're still just dreadful.

      My Audigy 2 is not that old, but after much fighting I still couldn't get 4.1 sound and EAX to work in any capacity. Part of it was Creative insisting on their own competing implementation of how to configure speakers which does not play nicely with the one included with Vista. Other issues are due to the general crummy nature of the drivers. Still other issues apparently only occur on Vista64 with 4 or more GB of RAM. Just awful. Eventually, I had to stop using the Audigy and use the onboard RealTek branded Intel HDA chip which seems to work fine, though the sound is less clean than what I got with my Audigy.

      Another piece of hardware, a Playstation/Gamecube/Dreamcast to USB controller adapter, from EMS Production (http://www.hkems.com) won't work with Vista64 either. Two years in and the company, still alive, has yet to release any Vista64 drivers and the Vista32 drivers are still listed as "beta".

      The annoying thing here is that the damn thing shouldn't even *need* an adapter. In Linux it is simply recognized as a HID gaming device and works fine. Vista actually recognizes it as such and DirectX controller diagnostic program can properly read values from the controller, but Vista steadfastly refuses to list the device in the "Game Controllers" control panel dialog, making it pretty useless for anything.

      Sigh... at least both these pieces of hardware work perfectly well in Linux...

      --
      Elrond, Duke of URL
      "This is the most fun I've had without being drenched in the blood of my enemies!"-Sam&Max
    7. Re:Quickly, bash microsoft. by Ralish · · Score: 5, Informative

      I'm sorry, but Peter Gutmann is not a reputable source for accurate information on Vista graphics, or anything related to Vista at all. Several of his claims have been widely proven to be exaggerated or downright false, and when asked to provide proof, he has refused. His claims have been picked apart on numerous sites both directly and indirectly through the sourcing of benchmarks.

      I suggest you read these articles for instance, which provide a good overview:
      http://blogs.zdnet.com/Ou/?p=673
      http://blogs.zdnet.com/Ou/?p=718

      Some of his points are admittedly valid, there are genuine flaws in the new graphics driver device spec., but he's clearly most concerned with pushing an anti-Vista agenda, even if that requires resorting to FUD.

      Choose your "experts" carefully.

    8. Re:Quickly, bash microsoft. by cbhacking · · Score: 3, Informative

      I presume you're referring to the article in which he described Vista as something like "quite possibly the longest suicide note in history"? I read it back in 06 shortly after it was initially published, I didn't know Gutmann's work terribly well before reading that, but he came highly recommended.

      However, that article cost him about 98% credibility with me. Some of it - even some of the really bad stuff - might in fact be true. However, there were trivially verfiable claims he made which were blatantly untrue (an example being that ATI, nVidia, and other graphics companies were going to need to switch away from unified drivers, and provide a different driver for each card model - which by the time the article went public was an obvious falsehood since you could download and install the beta Vista drivers for any card in a given family and they would work fine).

      If the man can't be bothered to do even that minimal an amount of research (it also didn't help that he refused to disclose any of his sources) then he has no business publishing in anything but tabloids, nor does he have any place in academic circles. I am a student, not a professor, but if I had written such tripe and submitted it to anybody who knew what I was talking about, I'd have been laughed out of the department.

      Incidentally, the article has been edited at leas three times since its initial publication. While I have no objection to revising, it is usually done prior to publication, not afterwards. Furthermore, while some of the more blatantly false claims are missing from the latest version, Gutmann neither addresses nor explicitly retracts those statements. It is although he wishes to remove the original statements entirely, though nothing controversial on the Internet ever vanishes so thoroughly as that.

      --
      There's no place I could be, since I've found Serenity...
    9. Re:Quickly, bash microsoft. by Anonymous Coward · · Score: 0

      I finally gave up on Nvidia's shitty drivers when a driver update in June caused all AVI files to skip when emule was open.

      More likely it was from the shitload of viruses you downloaded, JACKASS!

    10. Re:Quickly, bash microsoft. by wwahammy · · Score: 1

      I agree with the siblings; find a better source if you're going to claim the DRM/HD video issue caused the graphics companies problems. I'm not saying it has nothing to do with the problems; I just don't have the evidence necessary to make that conclusion.

      I also think the argument is tough to believe. You're claiming Microsoft, in order to help out the RIAA/MPAA, made it nearly impossible to write graphics drivers for their OS? Ya the DRM is fucking stupid but is there any evidence that engineers at Microsoft had any worries about how this would affect the quality of the system as a whole?

      I had a card that came out about 6 months before Vista. I think the most likely reason why Nvidia doesn't care that much about the quality of my experience is that Nvidia would prefer for me to just buy a new video card.

    11. Re:Quickly, bash microsoft. by mike_sucks · · Score: 2, Informative

      Actually, the whole point of OpenGL was to provide software- and hardware- vendor agnostic API for writing applications that perform 3D rendering. You've clearly been living in a monoculture too long if you can't see that.

      Software fallback is nice to have but, it's certainly not the reason OGL exists.

      /Mike

      --
      -- "So, what's the deal with Auntie Gerschwitz et all?"
    12. Re:Quickly, bash microsoft. by xlotlu · · Score: 1

      This has thoroughly been discussed by Peter Gutmann here

      You mean here

    13. Re:Quickly, bash microsoft. by Whiteox · · Score: 0, Redundant
      --
      Don't be apathetic. Procrastinate!
    14. Re:Quickly, bash microsoft. by Whiteox · · Score: 2, Insightful

      Well I re-read Gutmann and the blogs and I feel a bit ignorant. When I read it a few years ago, I was by his thoroughness, especially with the driver issues..
      How things change.

      --
      Don't be apathetic. Procrastinate!
    15. Re:Quickly, bash microsoft. by Anonymous Coward · · Score: 0

      You left out the link to Gutmann's comments, or by "here" did you mean generally in /.? Anyway, I went looking and found this. Indeed an interesting read. I knew Vista's driver model was pathetic in part because of "content protection" issues, but I had no idea how awful it is.

      And can you imagine them trying to provide backwards support for this crap in Windows 7? What a nightmare!

    16. Re:Quickly, bash microsoft. by Bert64 · · Score: 1

      Amusing, so in many cases vista now has inferior driver support to linux...

      I have an older pci soundblaster 128 card that doesn't work in xp64 or vista (xp32 picks it up out of the box), and dec tulip ethernet cards which also wont work on 64bit xp or vista. All of these devices work fine on 64bit linux.

      I don't see why i should stop using them, the soundblaster produces better audio than the onboard chipset and the tulip is a perfectly fine ethernet controller, doesn't support 1gb but neither does my switch.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    17. Re:Quickly, bash microsoft. by Anonymous Coward · · Score: 0

      is there any evidence that engineers at Microsoft had any worries about how this would affect the quality of the system as a whole?

      It depends on what you want your system to do. Probably like most of us pretty much everything at one time or another.
      Gutmann makes the point that MS didn't introduce DRM because of pressure from the RIAA/MPAA, but rather that it would step-lock premium services to their platform.
      Others in this tread have pointed out that Gutmann is not a reliable source, but I think he has a point anyway.
      Any engineering team would find it a challenge to create a sophisticated DRM like the one in Vista - theoretically an absolute secure path from media to screen, and as a technical achievement it was a good try - all bases covered BUT
      I bet it left a bad taste in their mouths knowing that it was at the expense of system resources and inefficient.
      That's just my humble opinion.

    18. Re:Quickly, bash microsoft. by Anonymous Coward · · Score: 2, Interesting

      Gutmann was right about one thing: that content protection mechanisms will require parts of hardware specification to be kept confidential.

      We see this with AMD: they are careful about releasing hardware specs for their hardware to X.org community because they have to omit documenting certain components (e.g. UVD engine) because their hardware is not yet safe against hacking to get out unencrypted content. Rendertest is performed to check if the card is genuine. With hardware spec, someone could easily build a fake card (or hook up some kind of emulation on the PCIe port) and rip out unencrypted video bitstream.

      Fortunately for them, people are currently hacking software players, but in a year or two that might become unavailable, so they will have to turn to hardware.

    19. Re:Quickly, bash microsoft. by mlwmohawk · · Score: 1

      How about the vendors learn to code and stop writing shitty drivers!

      How about creating a platform that is well documented and designed, so that drivers written for it don't change drastically from revision to revision and have brain dead restrictions!

      For all of Microsoft's propaganda about APIs being stable, display and the changing direct access systems WinG, DirectWhatever, the API is a crap shoot.

      I can't blame a single vendor for poor support on Windows because it is a horrible and expensive system to support.

    20. Re:Quickly, bash microsoft. by DannyO152 · · Score: 1

      Yes, we know that the Gutmann paper is flawed. However, up above we see you say "not a reputable source for.. anything related to Vista at all." and "Some of his points are admittedly valid."

      So, regarding this particular topic is Gutmann on point or not?

      My question is, why was everyone jumping up and down about how clever they were in sending graphics rendering out the video card's GPU when it's seemingly real clever when Microsoft puts it back in the CPU? What am I not getting?

    21. Re:Quickly, bash microsoft. by UnknowingFool · · Score: 1

      I wish I had mod points to use on the parent. The GPU companies (emphasis on Nvidia though) knew the Vista driver model 18 months prior to its release and they still couldn't come up with decent drivers on time or ever two years later.

      To be fair, nVidia (and everyone else) was dealing MS. Yes they knew it was coming and the change was major but it wasn't like MS wasn't changing signficant parts of Vista constantly and at the last minute. Before someone goes on about how they didn't change XX feature, MS did change YY feature at the last minute. If you've been following MS for a while, this was the norm not the exception. If your job was to write a driver you'd wait too unless MS makes a significant change that essentially negates any work you've done.

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
    22. Re:Quickly, bash microsoft. by TrekkieGod · · Score: 1

      Creative is still surviving because they were the standard back in the DOS days. People remember that if they didn't have a soundblaster, it just didn't work well with their games.

      After serious issues with my own Audigy 2 I just realized that, honestly, the standard AC'97 on-board audio these days is most likely just as good, definitely more compatible with whatever operating system you use, and as advanced as you need an audio card to be (mine is 7.1 with digital out).

      Once more people realize that, they're finally going to go bankrupt, and nothing of value will be lost.

      --

      Warning: Opinions known to be heavily biased.

    23. Re:Quickly, bash microsoft. by Hann1bal · · Score: 0

      "So Gutmann does not deny making these outlandish unsubstantiated claims and he has so far provided zero data. On the other hand, I have do have plenty of hard scientific data to refute Gutmannâ(TM)s claims. AnandTech ran a series of HD DVD and Blu-ray tests with hard CPU utilization numbers. On a low-end Intel E4300 Core 2 Duo CPU, CPU utilization ran as low as 7% for 1080p VC-1 encoded movie when VC-1 video compression decoding was offloaded to the ~$100 ATI 2600XT graphics adapter." Notice he does not say about cpu usage when video decompression IS NOT offloaded to the gpu?

    24. Re:Quickly, bash microsoft. by RiotingPacifist · · Score: 1

      How about creating a platform that is well documented and designed, so that drivers written for it don't change drastically from revision to revision and have brain dead restrictions!

      Any examples of that working?
      mac os [_]
      linux [_]
      windows [_]

      Granted linux drivers change more gradually because they release a new version every few months but if linux worked on the same timescale as windows/mac the changes would still be pretty drastic. perhaps minix has a stable API?

      --
      IranAir Flight 655 never forget!
    25. Re:Quickly, bash microsoft. by RiotingPacifist · · Score: 1

      Its cool we're introducing pulseaudio to a few major distributions to make any better support purely theoretical.

      --
      IranAir Flight 655 never forget!
    26. Re:Quickly, bash microsoft. by i.of.the.storm · · Score: 1

      Umm... no? DirectX did that too, it's just that I think the software renderer hasn't been updated in forever so the performance is basically worthless. I remember when I did DirectX8 stuff you could choose hardware or software rendering, but the software was basically worthless. The point of a graphics API is to make it so that you don't have to deal with hardware-specific issues.

      --
      All your base are belong to Wii.
    27. Re:Quickly, bash microsoft. by Anonymous Coward · · Score: 1, Interesting

      First, I should note that I seem to be missing how the conversation shifted from hardware vendors not supporting devices properly to Gutmann's paper. That said, let's discuss that.

      If you want argue that "numerous sites" have debunked Gutmann's paper, you should at least cite two different ones. If you're going to cite only one site to debunk Gutmann, I don't recommend citing George Ou.

      In the first link, Ou attacks Gutmann's paper on two fronts. First, Ou argues against Gutmann's concerns about downsampling by pointing out that the studios, as of his 2007 article (I have no idea whether this remains the case) had not turned on the ICT flag yet.

      That may be a fair point for much of the market. It might make sense for someone who needn't be sensitive about the issue to just buy and assume that the media protections won't cause them any major issues.

      However, Gutmann is not presenting arguments for the broad market, he's presenting arguments for an audience whose focus is security. Gutmann is paid not to take unnecessary bets, and he wouldn't be doing his job if he recommended betting that studios wouldn't flip on the ICT flag anytime soon.

      Even worse, Ou expands his attack to note that Gutmann hasn't tested his ideas. Frankly, this attack is hypocritical, since Ou has just pointed out that the studios are not producing media with the ICT flag enabled. As a result, one has to ask how Gutmann could be expected to reasonably test his concerns for lack of media...and how Ou could have done any more testing than Gutmann. (Theoretically, one or both could pay the required license fees and produce their own media, but the cost of such an effort dwarfs that of buying the OS and suitable test hardware.) Furthermore, the scant testing results Ou cites in the article fall far short of that required to rebut Gutmann's concerns about false positives triggering media protection.

      That's the point where Ou performs better. The second point he claims to address (as derived from the quote he gives) was that media protection would require that more components remain active when a PC was in power saving mode. To rebut this argument, Ou notes that the CPU utilization he measures when playing videos when the computer isn't "asleep" is reasonable. Chewbacca defense anyone?

      If one assumes that Ou just wrote poorly and wanted to counter the processing power concerns Gutmann raises, then one just has to look at what Gutmann wrote to see that Ou's counter is tripe.

      Gutmann quotes Microsoft (a hint that Ou needs to think harder as writing):

      In the case of premium content, whether video can play back smoothly when using regular AES with uncompressed video will be a function of the resolution of the uncompressed video and the power of the processor. It is unlikely to work well in 2006 for uncompressed HD premium content

      There is a reason the quote refers to 2006; it was the time when Gutmann wrote the bulk of his paper. (He's made a few updates since, the last of which was apparently in 2007.)

      To debate this point, Ou runs some CPU requirement calculations using 3.5 megabytes per second for a 1080p HD stream. Since a quick calculation (1920 * 1080 * 24) shows that a 1080p HD stream to contain nearly 50M pixels a second, it seems pretty clear that Ou is running computational requirements for an compressed stream, rather than the uncompressed one Gutmann worries about. Then, to top it all off, Ou compares his number to a processor that didn't exist when Gutmann wrote his paper. Apples to oranges, anyone?

      I'll grant some points against Gutmann. His paper is very speculative, and his concerns may not be relevant to a good chunk of the market. (In Gutmann's defense, I'll point out that I vaguely recall reading it before Vista was even publicly released, making a healthy dose of speculation unavoidable. Furthermore, the concerns in Gutmann's paper are valid and worthy of strenuous investigation in the security circles for which it was originally written.)

      How

    28. Re:Quickly, bash microsoft. by Monsuco · · Score: 1

      Sigh... at least both these pieces of hardware work perfectly well in Linux...

      Oh boy, now I can play tux racer with a controller!

    29. Re:Quickly, bash microsoft. by tyrione · · Score: 1

      I have a better idea. How about they work on getting their operating system to judiciously select available Cores to offload more work and thus have a more responsive, robust operating system. Once they've accomplished this they should have APIs for developers to effectively do the same for their applications to spread around portions of their applications workload demands across those cores.

    30. Re:Quickly, bash microsoft. by lusid1 · · Score: 1

      That's a different game of "find a working driver".

      When a user drops a Vista laptop on your desk and begs for an XP install, you'll find that not all laptops shipped with Vista have readily available XP drivers.

    31. Re:Quickly, bash microsoft. by jonadab · · Score: 1

      I lump ATI and Nvidia into the same general category of junk. Their graphics cards are toys. I suppose they're okay if you just want to play games all day...

      Of course, for a server you scarcely even need a graphics card (how often are you even going to plug a monitor into it anyway, once for an hour when you do the initial install?), so any old junk will do, including Intel-chipset cheapies or even onboard video.

      But for a workstation I always go with Matrox.

      --
      Cut that out, or I will ship you to Norilsk in a box.
    32. Re:Quickly, bash microsoft. by wwahammy · · Score: 1

      Different tools for different jobs

  5. it doesn't even need MMX or SSE by Anonymous Coward · · Score: 0

    I fucking well hope that doesnt mean it doesnt make use of them. seems like a waste of time coding it so it can run without (what today is considered) standard fucking features of a cpu.
     
    after vista, the os not capable of running useably on older hardware, it seems a bit dumb that they're not trying to claim they can do dx10 in software on an 800mhz cpu. but i guess thats why MS just keeps slipping, they have no fucking direction. you already forced all your customers to buy new PCs for vista. now is not the time to start worrying about running on non-vista capable hardware. those people are no longer your customers

  6. Unbalanced comparison: cost by Anonymous Coward · · Score: 5, Insightful

    How much is an 8-core system going to cost vs the system with integrated graphics? At that point, it seems wiser to invest more money in a graphics card than in faster CPUs if that's what you're going to be doing.

    By far the more useful thing is that it's probably better for development because the driver developers will have a reference point of how the graphics are supposed to render. Also, larger game companies will be able to point out these differences to get bug fixes out of the graphics card companies. "Your graphics card renders this incorrectly with regards to the reference, fix it" is much more forceful than "your graphics card behaves differently than your competitor".

    1. Re:Unbalanced comparison: cost by Lord+Crc · · Score: 5, Insightful

      Also, larger game companies will be able to point out these differences to get bug fixes out of the graphics card companies. "Your graphics card renders this incorrectly with regards to the reference, fix it" is much more forceful than "your graphics card behaves differently than your competitor".

      DirectX already contains a reference rasterizer, which is better suited for that. This thing seems instead to be meant for applications that doesn't necessarily need more than "interactive" frame rates, but do need to run on a broad class of machines. Or for easing development of applications which could benefit from hardware acceleration when available (image processing f.i.).

      From the MSDN page on WARP:

      We don't see WARP10 as a replacement for graphics hardware, particularly as reasonably performing low end Direct3D 10 discrete hardware is now available for under $25. The goal of WARP10 was to allow applications to target Direct3D 10 level hardware without having significantly different code paths or testing requirements when running on hardware or when running in software.

    2. Re:Unbalanced comparison: cost by Anonymous Coward · · Score: 0

      Why didn't they compared Intel graphics card (BTW: which one?) to 1024-cores supercomputer? That would be even more unfair.

    3. Re:Unbalanced comparison: cost by ceoyoyo · · Score: 2, Insightful

      No no, see, now when Windows 7 requires video cards that nobody has but MS puts Windows 7 Ready stickers on all of the new computers anyway, when people say "my Windows 7 Ready computer won't run Windows 7!" MS can point out that yes, it does. Any version of Windows 7. Sure, it takes ten minutes to draw a menu, but it runs!

    4. Re:Unbalanced comparison: cost by Anonymous Coward · · Score: 0

      So how exactly does that contradict what I said. It seems like WARP10 is just the DX10 version of the reference rasterizer. You even quoted it yourself:

      WARP10 was to allow applications to target Direct3D 10 level hardware without having significantly different code paths or testing requirements

      Sounds like a reference rasterizer to me.

    5. Re:Unbalanced comparison: cost by Lord+Crc · · Score: 1

      So how exactly does that contradict what I said. It seems like WARP10 is just the DX10 version of the reference rasterizer.

      You said that "by far the more useful thing is that it's probably better for development because the driver developers will have a reference point of how the graphics are supposed to render".

      I see WARP as more useful for developing actual applications than for debugging drivers.

    6. Re:Unbalanced comparison: cost by Anonymous Coward · · Score: 0

      How would this help as opposed to running this on actual hardware? If you've got an 8-core machine and you're doing graphics development, chances are you can afford a graphics card that can outperform your CPUs (essentially any graphics card).

      The only thing this can do is help as a reference to detect flaws in the graphics card. Potentially it's also a way to decrease development time spent trying to support cards that don't implement certain features in HW, although DX10 defines a minimum set of capabilities, so this would only be a benefit if MS emulated HW features not present in the DX10 spec, which would be strange.

      To me, this smells more like MS got scared of the lawsuit and now they're trying to get Aero/compositing running on underpowered machines to prevent this situation from coming up in the future. I think this makes sense because CPUs are pretty powerful today and MS can't get sued for saying something is "Windows 7 Capable" if it implements the feature, only slowly. That was the problem with the "Vista Capable" slogan - they tried to make a partner happy and thus mislead their customers that their cheaper computers would support the same features.

  7. From the summary: by ben0207 · · Score: 4, Insightful

    "Running Crysis at 800 x 600 with the lowest quality settings, an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics."

    So the game went from unplayable at the lowest settings possible, to being still unplayable at the lowest settings possible?

    Great move MS, youv'e really solved a problem there.

    --
    cmd-q.co.uk - some sort of stupid fucking internet bullshit
    1. Re:From the summary: by Captain+Splendid · · Score: 2, Funny

      managed an average frame rate of 7.36fps, compared with 5.17fps

      But, but, that's like, a 42% improvement! That's like, massive, man! MS are awesome!

      --
      Linux, you magnificent bastard, I read the fucking manual!
    2. Re:From the summary: by ChunderDownunder · · Score: 0, Redundant

      640-core CPUs should be enough for anyone.

      No, seriously, MS have just pioneered a revolution in gaming. It's called "super super super slow-mo".

    3. Re:From the summary: by cheater512 · · Score: 4, Funny

      No, the game goes full speed, you just die randomly however as someone runs in, head shots you, runs out and your computer is still trying to render the first frame.

    4. Re:From the summary: by Pr0xY · · Score: 4, Informative

      As I said in another post:

      Running Crysis isn't the point of the demo. The point was that it was a DX 10 application running entirely in software. In the end, this means that systems without higher end 3D cards would be able to run Aero. THAT's the point.

      They are trying to address the main complaint of the "Vista Capable" debacle. Running Crysis was just a way of demonstrating the capability.

    5. Re:From the summary: by Revenger75 · · Score: 0

      Maybe so, but maybe M$ figured that instead of benchmarking a game where it might actually make a difference and then having all of /. ask "But can it run Crysis?", they decided to answer that question.

    6. Re:From the summary: by hairyfeet · · Score: 4, Informative

      The direct link to their numbers is here, including number for quad and dual CPUs. And here is the inquirer's take on it, which I tend to agree. This is about making sure that Win7 is put on as many machines as possible and doesn't have a "Vista Capable" debacle out of the gate. With this tech as long as they don't fuck up the CPU specs like they did with Vista(A 1GHz with 512Mb of RAM for Basic and 1Gb for all the others? WTF?) they should be able to give the Aero "experience" no matter how shitty of an Intel integrated GPU comes with your laptop. Of course it'll run so damned slow that the desktop will be pretty much the only thing you CAN run, but there won't be any more lawsuits because the machines can't run features. Anyway that is what I'm betting is going on in the mind of MSFT.

      Personally I'll just be happy if Win7 doesn't run like a damned slug. because I'm really getting tired of playing "find a working driver" for all those damned laptops that keep getting dumped on my desk to be "downgraded" from Vista. I shudder to think how all those Best Buy and Wal Mart sub $600 laptops would have run if Vista would have had this "feature" at launch. How about making a nice lean functional OS instead of trying to out pretty Apple MSFT? Because frankly when you try to do Apple pretty you just end up sucking the big wet titty. Just accept the fact that you suck at pretty and move on. Win2K and WinXP weren't pretty and look at how much cash you made. Those of us that work with Windows will take compatibility and speed over pretty any day of the week. Just beg Allchin to come back and make backwards compatibility job #1 again and you'll find your customers will be happy.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    7. Re:From the summary: by this+great+guy · · Score: 1

      One 10-20W Intel IGP: ~$20.
      Two 130W quad-core Core i7 920 processors: $568
      Using the latter to barely manage to match the performance of the first: priceless.

      Remind me to short MSFT when the market opens tomorrow morning.

    8. Re:From the summary: by SeaFox · · Score: 1

      I just find it hilarious these are the result of using a processor that is just now being released on what is considered too low a resolution by today's standards anyway.

      If you can afford a system with a Core i7 processor, having a decent graphics card will not be a concern to you.

    9. Re:From the summary: by danlor · · Score: 1

      Well... then they better try again. It still sounds like a complete failure to me. Since the integrated graphics is equivalent, there is no advantage, and no resolution to the problem. What exactly are you try to get at?

    10. Re:From the summary: by Anonymous Coward · · Score: 0
    11. Re:From the summary: by Anonymous Coward · · Score: 1, Informative

      If they did Aero in openGL they'd have software rendering from the beginning. Oh right, not invented here.. my mistake..

    12. Re:From the summary: by Rockoon · · Score: 1

      "not invented here" but more importantly "hijacked by a bunch of special interest dirt-bags that don't care how crappy opengl remains as long as it still does what THEY need"

      --
      "His name was James Damore."
    13. Re:From the summary: by vux984 · · Score: 4, Informative

      Well... then they better try again. It still sounds like a complete failure to me. Since the integrated graphics is equivalent, there is no advantage, and no resolution to the problem. What exactly are you try to get at?

      Except the integrated graphics on a bunch of 'Vista Capable' laptops DON'T do DirectX10 or Aero... but if a patch to Vista (or Windows 7) will get Aero working on directX10 on the CPU... a buttload of PCs that CAN'T currently do Aero, now CAN.

    14. Re:From the summary: by Anonymous Coward · · Score: 0

      The problem being there are very few shaders in crysis that are sm4 or require dx10 hardware. Running it at the lowest detail settings actually turns nearly all of the shaders off so this software renderer is in fact doing nothing with dx10 or dx9 for that matter.

    15. Re:From the summary: by Anonymous Coward · · Score: 0

      Correct. They just proved they are better at coding than Intel.

    16. Re:From the summary: by Anonymous Coward · · Score: 0

      Running Crysis isn't the point of the demo. The point was that it was a DX 10 application running entirely in software.

      The same could be done for DX 9, couldn't it? Just initialize a DD3DDEVTYPE_REF or D3DDEVTYPE_SW device type instead of D3DDEVTYPE_HAL. Only difference is, games could tell it's not a HAL type and kill themselves on purpose.

      Besides, there's nothing good about "running in software". The important thing is if it's "running well enough in software", and if the average joe is going to need an 8-core system to edge out crappy Intel hardware, odds are it won't be.

    17. Re:From the summary: by WTF+Chuck · · Score: 4, Insightful

      Except the integrated graphics on a bunch of 'Vista Capable' laptops DON'T do DirectX10 or Aero... but if a patch to Vista (or Windows 7) will get Aero working on directX10 on the CPU... a buttload of PCs that CAN'T currently do Aero, now CAN.

      But at what performance cost. If we are talking about the whole "Vista Capable" debacle, aren't we talking about low spec machines that coughed and wheezed when running the low-end version of the OS. Great, lets add 3D rendering to the processor load on those machines.

      I like the idea of rendering the graphics in the CPU rather than an expensive accelerator card for one-off situations, as long as that feature can be turned off. But then, I'm not a gamer, and I'm not into all the eye-candy. If I were a gamer or into eye candy, there's no way this side of hell that I would want to render the graphics in the CPU. I would get the best video card money could buy.

      --
      Note - Liberal use of <sarcasm> tags may or may not need to be applied.
    18. Re:From the summary: by Anonymous Coward · · Score: 0

      Running Crysis isn't the point of the demo. The point was that it was a DX 10 application running entirely in software. In the end, this means that systems without higher end 3D cards would be able to run Aero. THAT's the point.

      If it can only run Crysis (in 800x600 and with the lowest settings, no less) with 7 fps, then how many fps will you get for Aero?

      (This is a serious question, BTW, just to avoid any possible confusion.)

    19. Re:From the summary: by Anonymous Coward · · Score: 0

      In the end, this means that systems without higher end 3D cards would be able to run Aero. THAT's the point.

      Well, that's the last but one objection to me buying Vista dealt with. Just the product activation to go, and MS will have me as a customer.

    20. Re:From the summary: by vux984 · · Score: 1

      But at what performance cost. If we are talking about the whole "Vista Capable" debacle, aren't we talking about low spec machines that coughed and wheezed when running the low-end version of the OS. Great, lets add 3D rendering to the processor load on those machines.

      Seriously, its not the CPU utilization on these low end units was the bottleneck. Low RAM, crappy video cards yes, but even on an 4+ year old Pentium 4 vista doesn't exactly tax the cpu much.

    21. Re:From the summary: by ultranova · · Score: 1

      But at what performance cost. If we are talking about the whole "Vista Capable" debacle, aren't we talking about low spec machines that coughed and wheezed when running the low-end version of the OS. Great, lets add 3D rendering to the processor load on those machines.

      I suspect that the purpose here is to be able to show before a judge that they weren't technically lying.

      --

      Forget magic. Any technology distinguishable from divine power is insufficiently advanced.

    22. Re:From the summary: by Anonymous Coward · · Score: 0

      perhaps what they're really trying to keep a step ahead of developments such as Adobe Photoshop CS4 (and others). Which was discussed a while back here, with the introduction of video acceleration as part of the package. This might simply be a move by Microsoft to allow this along with other products (including but not limited to the Aero UI) to be used over Virtual Servers and Terminal Server thin client style environment setups.

      They would be shooting themselves in the foot more-so if they continue to pass this over to Mac workstations as they have in the past.

    23. Re:From the summary: by jonaskoelker · · Score: 4, Insightful

      be able to run Aero. Running Crysis was just a way of demonstrating the capability.

      I think running Aero at would be a better way to demonstrate that capability.

    24. Re:From the summary: by blahplusplus · · Score: 1

      "So the game went from unplayable at the lowest settings possible, to being still unplayable at the lowest settings possible?

      Great move MS, youv'e really solved a problem there."

      The software render is for older games, not for the cutting edge. The theory is - as CPU's get faster, games will be able to be playable. It's not a bad idea IMHO because many games 'break' on GPU's if they are not supported in the driver as hardware changes. Software compatability is something that is a bit frustrating with GPU's, to give an example: Final fantasy 7 for the PC will not work properly on modern cards without issues. In these cases the software render is a godsend provided it intelligently maintains the needed functionality for such games and doesn't botch it like many GPU manufacturers.

    25. Re:From the summary: by Bert64 · · Score: 3, Insightful

      And if the CPU is pegged rendering the GUI, what effect is this going to have on whatever the user is actually trying to do?

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    26. Re:From the summary: by Anonymous Coward · · Score: 0

      Yeah I'm sure if a laptop isn't Vista-capable to begin with, it will have enough "spare" CPU cycles to emulate the 3D rendering by the CPU. (sarcasm)

      IMHO, the real reason for this WARP 10 thing is to make more games _theoretically_ able to run on more machines, so they can sell more games. Even though they will be unplayable, user-experience doens't matter much to MS, look at the latest Xbox upgrade. If you look at boxes in the shop, you will see they always require a certain Nvidia or ATI chipset. Now they can say "Look, the specs say it should work on your cheap Vista box" and refuse you a refund.

    27. Re:From the summary: by Bert64 · · Score: 1

      Although, OpenGL/X11 has always supported remote displays, and it will do the rendering on the workstation's GPU instead of the server... Give it a try, it's surprisingly quick.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    28. Re:From the summary: by TheRaven64 · · Score: 1

      I don't really understand this. DirectX had software fall-back from the start. Software could test capability flags to see what was done in hardware and what was done in software, and avoid the software paths (something that OpenGL doesn't provide a standard way of doing, by the way, you can only test whether extensions are supported, not whether they're fast). Wasn't one of the points of DirectX 10 to remove the software paths so that if a machine was rated as running DirectX 10 you knew it could do all of the DX10 things fast? Adding them back in seems a bit like they have just realised they made a mistake with Vista, and most people would prefer slow to non-functional.

      --
      I am TheRaven on Soylent News
    29. Re:From the summary: by westlake · · Score: 1
      Of course it'll run so damned slow that the desktop will be pretty much the only thing you CAN run

      explain to me how much of a load Aero puts on the system when applications are running.

    30. Re:From the summary: by nxtw · · Score: 2, Insightful

      Except the integrated graphics on a bunch of 'Vista Capable' laptops DON'T do DirectX10 or Aero... but if a patch to Vista (or Windows 7) will get Aero working on directX10 on the CPU... a buttload of PCs that CAN'T currently do Aero, now CAN.

      Aero does not require DirectX 10; it only needs DirectX 9 with the right features, enough memory, and a suitable driver.

      How many "Vista Capable" laptops aren't Aero-compatible? The Intel 945GM chipset runs Aero, and it began shipping in January 2006 (a year before Vista came out.) Any Intel Mac meets the hardware requirements to run Aero.

      I suppose there could have been some 915GM laptops sold with Vista, or perhaps laptops with the 945GM that had less than 1024MB of RAM.

    31. Re:From the summary: by Runefox · · Score: 1

      Presumably, slightly more than a GMA 950. :P

      --
      Screw the rules, I have green hair!
    32. Re:From the summary: by somenickname · · Score: 1

      If Aero is anything like compiz, it probably doesn't put any load on the CPU while it's running because it's instead putting the load on the GPU. If you specifically tell it to put the load on the CPU, I suspect Bad Things(tm) will happen to your CPU usage when doing things like, I dunno, scrolling in a web browser.

    33. Re:From the summary: by i.of.the.storm · · Score: 1

      Well, that 945G might run Aero, but probably not well. Also, the RAM thing is a bit sensitive, less than 2GB is really not a good idea with Vista. It works, but when you see it on a system with slightly better GPU and 2GB RAM it's a noticeable improvement.

      --
      All your base are belong to Wii.
    34. Re:From the summary: by i.of.the.storm · · Score: 1

      DirectX has software rendering too...

      --
      All your base are belong to Wii.
    35. Re:From the summary: by nxtw · · Score: 1

      Well, that 945G might run Aero, but probably not well.

      Why are you commenting if you haven't used it? Aero runs well enough for regular use on the 945GM; last time I used it (two years ago) the framerate was lower with single-channel memory, but not so much that it was an annoyance. Plus, those who would notice that kind of thing are likely to have two DIMMs installed.

    36. Re:From the summary: by i.of.the.storm · · Score: 1

      I have used it on other people's laptops with the 945G and 1GB RAM and it was not a pleasant experience. The only alternative was to use XP on their ancient, virus-ridden PIII machine that originally ran 98, so it was still the better choice. Above all else, Vista is RAM hungry, which is actually kind of a good thing since DDR2 is so cheap these days. I guess most people who would buy laptops and care about Aero performance wouldn't get a laptop with the 945G but when you compare performance of the Flip3D switch between a real GPU and the 945/915G class GPUs, it's clearly apparent that the Intel solution is far from adequate. To be fair, the older nVidia and ATi IGPs with similar performance (Radeon 9000, Geforce 6100/6150) also struggle with Aero, but those haven't been sold as recently for the most part.

      --
      All your base are belong to Wii.
    37. Re:From the summary: by 427_ci_505 · · Score: 1

      A newer laptop (vintage Sept 08) with Intel integrated video seems to run Aero just fine, by my experience.

    38. Re:From the summary: by i.of.the.storm · · Score: 1

      That's probably running one of the newer Intel IGPs though, either X3100 or X4500.

      --
      All your base are belong to Wii.
    39. Re:From the summary: by icsx · · Score: 1

      Well how good will Aero run then? Would be (sounds like) pretty shitty user experience for me but hey, it runs! Something to brag about for Microsoft and then users who use it will say hey this windows 7 lags more than my Vista because someone screwed them again upon buying decent pc.

    40. Re:From the summary: by icsx · · Score: 1

      Maybe it runs Crysis, but not Aero. ;)

    41. Re:From the summary: by poot_rootbeer · · Score: 1

      The point was that it was a DX 10 application running entirely in software.

      The point is that the performance demonstrated with this software rasterizer is no better than 25% of what I believe most people would consider the minimum, MINIMUM acceptable level.

      It's the same bullshit as the "Vista Capable" debacle. Microsoft is asking the question "does it run?" when they should be asking "does it run WELL?". Obviously, the answers are different.

    42. Re:From the summary: by MightyYar · · Score: 1

      Who cares? It'll be "ooohh, pretty"!

      For a little while in Win 3.1 I had a driver that used the CPU to drive the internal speaker to emulate a sound card... it locked up the whole PC while it did it's thing, but damnit... I had SOUND! :)

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    43. Re:From the summary: by jonadab · · Score: 1

      > But, but, that's like, a 42% improvement! That's like, massive, man

      I know you're not being entirely serious, but it's worth pointing out that in computing, any improvement of less than an order of magnitude (e.g., going from 5.17 fps to something like 10.3) is usually not user-noticeable. A "significant" improvement would be more like two magnitudes (going from 5.17 to something like 21 fps) or better, and "massive" would be even stronger than that.

      (Yeah, I know, extreme overclockers will torture themselves and their hardware to get an extra 5% performance, but they have to use benchmarks to see the results. A normal person would never notice the difference.)

      --
      Cut that out, or I will ship you to Norilsk in a box.
  8. Grrrreat! by Chordonblue · · Score: 4, Insightful

    Does anyone else remember the 'good old days' when certain 3D graphics cards (the ViRGE comes to mind), were actually SLOWER than software renderers?

    The term used then was 'decelerator' and I think MS's stupid decision to (once again) bow to Intel on this should share the same term.

    How long will it take for true 3D acceleration to become an expected standard feature on PC's?

    --
    "...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
    1. Re:Grrrreat! by WiiVault · · Score: 1

      Sadly, never as long as the GUI works most Joe and Jane sixpacks will be just fine; and yes I do know about the Vista debacle but I think the point is still valid.

    2. Re:Grrrreat! by vux984 · · Score: 1

      Sadly, never as long as the GUI works most Joe and Jane sixpacks will be just fine; and yes I do know about the Vista debacle but I think the point is still valid.

      Then you'd be mistaken. Both OSX and Microsoft effectively require hardware 3D acceleration for their desktop effects. All new Macs and any PC that actually meets Vista's real requirements feature 3D acceleration.

    3. Re:Grrrreat! by A+Life+in+Hell · · Score: 5, Insightful

      Sadly, never as long as the GUI works most Joe and Jane sixpacks will be just fine; and yes I do know about the Vista debacle but I think the point is still valid.

      How is that sad? If people don't need it, it seems like a waste of money to me.

      --
      Commodore 64, Loading up the dance floor!
    4. Re:Grrrreat! by WiiVault · · Score: 1

      I have some cheap as hell NVidia integrated that tells me a different story.

    5. Re:Grrrreat! by hairyfeet · · Score: 4, Interesting

      Actually I think you are wrong, and here is why. I have been having lately a lot of customers, including older folks that wouldn't play a game if their life depended on it buying graphics cards. As a matter of fact on Monday I'll be picking up a passive cooled Geforce 6200 for a guy that I know hasn't played games since the age of DOS.

      So why the sudden interest in graphics cards? One word: Video. Folks are getting these nice cheap LCD monitors that do 1400 or 1600 res and they are quickly finding that while the integrated will render the screen, the second they try watching videos on it full screen it really starts to suck. So they come to someone like me asking for a video card since the want to watch....well, videos. Having a dedicated card for video, even an older one like I am going to get this gentleman(he has PCI only and a limited budget) with dedicated RAM simply stomps any integrated I have yet come across. Integrated may work well for rendering office docs and excel spreadsheets, but full screen video at a decent resolution? Not so much. So I think folks will be getting upgrades for those cheap desktops and wanting more and more to see videos run on the laptops before picking them up. Otherwise it really isn't a good viewing experience IMHO.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    6. Re:Grrrreat! by Chordonblue · · Score: 1

      But this is my point - when Microsoft advertises Vista they show all the fancy-pants graphic effects and video in full force. By continuing to allow manufacturers like Intel to get away with some sort of 'Ready for Vista' sticker, it's deceptive at worst, not entirely accurate at best.

      --
      "...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
    7. Re:Grrrreat! by neumayr · · Score: 1

      Won't apply to someone stuck on a PCI only system, but the general notion that IGPs suck for anything but office stuff is no longer true.
      My IGP (AMD780G/HD3200) is fast enough for 1080p video, and the few games I actually play on my PC (Q3A and HL2).
      I'm sure that's not the only one that can do that..

      --
      Truth arises more readily from error than from confusion. -Francis Bacon
    8. Re:Grrrreat! by smallfries · · Score: 1

      How does that work exactly? Given that your integrated Nvidia part is still a hardware accelerator.

      --
      Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
    9. Re:Grrrreat! by maxume · · Score: 1

      You are giving them bad advice. On this 2 year old laptop with Intel 945GM graphics, outputting a 512 by 224 mpeg4 (divx) video to the second monitor, fullscreen at 1680 by 1050, takes less than 10% of the cpu (a core duo at 1.6 GHz) using directX output. A 720 by 288 video takes less than 15% cpu.

      This is using VLC in hardware overlay mode, perhaps the software they are using is a little less efficient.

      --
      Nerd rage is the funniest rage.
    10. Re:Grrrreat! by Bert64 · · Score: 1

      I think the virge was the first consumer level 3d accelerator...
      And software rendering worked differently in those days, most code was specifically written for software rendering because thats all the majority of users had.

      Nowadays games are written for hardware 3d, and software rendering is just unusable despite massively faster processors. For a comparison, try running original quake in software mode, and compare it to glquake running in software mode... glquake will usually be massively slower.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    11. Re:Grrrreat! by Bert64 · · Score: 1

      Trying to watch video at that resolution will just saturate the PCI bus, it might even suck worse than the integrated card.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    12. Re:Grrrreat! by Ash-Fox · · Score: 1

      This is using VLC in hardware overlay mode, perhaps the software they are using is a little less efficient.

      Doesn't VLC use software rendering for everything?

      --
      Change is certain; progress is not obligatory.
    13. Re:Grrrreat! by TheRaven64 · · Score: 1

      The ViRGE wasn't slower than the software renderer on my machine (Cyrix M2 at 200MHz). With Microsoft's software opengl32.dll I got 0.3fps in GLQuake. With the ViRGE opengl32.dll I got 3fps (with the VooDoo 2 I got 40fps). Some games, like Escape Velocity used the ViRGE effectively - I got the same frame rates with the software and hardware renderer, but the hardware renderer was nicely antialiased and had blended fog.

      --
      I am TheRaven on Soylent News
    14. Re:Grrrreat! by drinkypoo · · Score: 1

      Does anyone else remember the 'good old days' when certain 3D graphics cards (the ViRGE comes to mind), were actually SLOWER than software renderers?

      The S3 Virge was faster than the PC alone providing the same graphics quality - remember, Virge games looked better than their un"accelerated" counterparts. Also, you lost maybe 1 or 2 fps at most on a high-end system; it was only in the low-end (granted, where you need acceleration the most) where performance suffered significantly. Intel is talking about a future with "thousands" of cores, and when we get there, the 3d accelerator will be a thing of the past. I don't believe 8 cores is there, though.

      The term used then was 'decelerator' and I think MS's stupid decision to (once again) bow to Intel on this should share the same term.

      Bow to intel? What the hell are you talking about? They're not removing support for 3d accelerators. Sky falling much?

      How long will it take for true 3D acceleration to become an expected standard feature on PC's?

      It's a stupid idea to have dedicated hardware if you don't need it. Right now, we are in a cycle where we need it. Sooner or later, we will be in the part where we don't.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    15. Re:Grrrreat! by maxume · · Score: 1

      It might default to software, but there is a check box for hardware overlay in the version I am using (8.6 on XP). The feature matrix suggests that there is good output support across all platforms:

      http://www.videolan.org/vlc/features.html

      Perhaps you are thinking about decoding, which is different than output (I don't think VLC can take advantage of hardware decoding, but it is not all that important for viewing movies on computers made in the last 5 years)?

      --
      Nerd rage is the funniest rage.
    16. Re:Grrrreat! by Anonymous Coward · · Score: 0

      Pepperridge farm remembers...

    17. Re:Grrrreat! by UserChrisCanter4 · · Score: 2, Informative

      Just a heads up: the PCI 6200 has some known problems with video playback. They were all driver-related, but as far as I know, NVidia never fixed them because the 6200 was always a fairly low volume unit and has now been dropped altogether.

      Google "GeForce 6200 video won't play" or something similar and you'll see the number of forum threads and posts where people complain about how this version of the drivers works but not this version and so on.

      The solution that's usually thrown about: disable hardware acceleration.

    18. Re:Grrrreat! by EvilIdler · · Score: 1

      A system without PCI-e or AGP can't be from the past five years, so the CPU probably also sucks. Hence the 6200, which might help in both decoding and displaying video.

    19. Re:Grrrreat! by ceoyoyo · · Score: 0, Flamebait

      Never, since MS keeps releasing new versions of DirectX that demand more features. The Intel 915 IS a "true" 3D accelerator. It just doesn't offer any of the DirectX 10 features, a few of the DX9 features or one of the DX8 features.

      WIndows 7, or Windows 8, is probably going to required DX11, so most or all of the current 3D cards aren't going to work. Does that mean they suddenly cease to be "true" 3D accelerators? Nope. It just means MS wants to try to force all the gamers to use their new OS for the "better graphics" and regular desktop users get caught in the squeeze.

    20. Re:Grrrreat! by maxume · · Score: 1

      Yeah, I missed that. That has a lot more to do with bus bandwidth than it does integrated/dedicated though (which is essentially irrelevant for everything but screaming 3d performance).

      --
      Nerd rage is the funniest rage.
    21. Re:Grrrreat! by hairyfeet · · Score: 1

      Well then could someone suggest a card in the $50 range for my customers then? I haven't used ATI since the bad old days of the Rage Pro and I have yet to find an easy "translation chart" that will show me which ATI=which Geforce and the only pci I am seeing in that range from them on Newegg is DX7. I need to order him one on Monday so he'll have a way of running his new $99 LCD special he picked up at the Staples Black Friday. So any suggestions?

      I choose the 6200 because that is what my oldest nephew has in his machine and it seems pretty solid and fanless means no noise. The only other card I could find from Nvidia at that range(his budget is $55 shipping included) is the FX5200 which I believe was nicknamed the "leaf blower" for how noisy they are. I admit I haven't had to mess with PCI in years, as I get a lot of work from SMBs and those machines usually have AGP. But I am finding now that a lot of those Wal Mart and Best Buy specials have PCI only so I need to have a card to recommend with those units. So could anyone with PCI experience tell me where to get a good video card for that price range. And if it is ATI could you please compare it to a Geforce model so I have a point of reference? Thanks in advance.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    22. Re:Grrrreat! by UserChrisCanter4 · · Score: 1

      The Radeon 9250 is frequently available in that price range in passively-cooled versions. The 9250 is, IIRC, a rebrand or slight tweak of the 8500 and is a DX8-series card according to Wiki.

      Tom's Hardware, Anandtech, or similar sites often have benchmark breakdowns for even older/budget models still available in their archives, and would probably be the best place to locate approximate comparisons across the range for 3D purposes. As far as video playback is concerned, your original thread said it all: anything with dedicated VRAM should do the trick.

    23. Re:Grrrreat! by i.of.the.storm · · Score: 1

      Well, to be honest that Geforce 6200 is barely better than the Intel GMA 950. I didn't know they even sold those anymore. I'm certain that it doesn't do any video decoding though. I would think you would want either a Radeon 3450 or Geforce 8400 for some basic video decoding, and actually quite capable gaming for the price (both of those should be around $30-50 if I recall correctly). Barely use any power either. Buying really old tech isn't always the way to go.

      --
      All your base are belong to Wii.
    24. Re:Grrrreat! by i.of.the.storm · · Score: 1

      The latest series of IGPs from AMD and nVidia are actually really good, but they're fairly recent and a lot of slightly old laptops don't have them. My tablet PC that I got in May has the craptastic Geforce 6150, although about two months later they updated their models to have the HD 3200, which really pissed me off.

      --
      All your base are belong to Wii.
    25. Re:Grrrreat! by hairyfeet · · Score: 1

      PCI ONLY remember? No PCIe means no 3450 OR 8400. And as for the earlier poster asking about CPU, it is a budget box with a 2.2GHz Pentium 4. You have to remember that those Compaq and Emachines sold at Best Buy and Wal mart up to the last year or two really didn't give you much. I have a 2.8GHz Compaq sitting in the closet waiting for me to get around to rebuilding it and it has a grand total of 3 PCI slots and that's it.

      ,p>Anyway I figure ANYTHING has to be better than the 915 chip it has in there now. And the 6200 has 256Mb of VRAM which should help with rendering. I know that according to their website the 6200 supports MPEG2 and H.264 and VC-1 decoding. He is an older guy and that means mostly DVD with a few videos from the web thrown in. For the money I'm betting on the 6200 and the fanless design is a plus.

      I'm just glad it isn't a laptop. I had to work on a Best Buy cheapo laptop that was about 2 years old last week and they are truly terrible. The thing has a Celeron DESKTOP chip in it, so the battery life is about 30 minutes and will fry your leg off, it has 256Mb of RAM to run XP Home, and the crappiest old Radeon card, which was 2 generations behind when it was new. Just a truly crappy POS laptop. But folks buy for price above all and those things are cheap. Just like this past Black Friday when I couldn't believe how many of the cheap laptops were running desktop CPUs in them. Just awful piles of junk.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    26. Re:Grrrreat! by i.of.the.storm · · Score: 1

      Here you go: PCI 8400
      Or ATI 2400, not as good as the 3450 on power consumption but still has the UVD chip. ATI's UVD has been generally found to be better than nVidia's but I would probably chose the 8400 over the 2400 since it's a newer generation and I believe a smaller process, so lower power consumption.

      I guess the 6200 will probably be better than the 915 chip (is that GMA 900?) but barely. My parents' laptop has the GMA 900 and it's about the same as the Radeon 9200 I installed in their desktop. What I meant was while the 6200 might accelerate video decoding, it doesn't have the dedicated chip that the newer Geforces and Radeons have that make video decoding use hardly any CPU or GPU at all. From what I can gather the 6200 will accelerate the video decoding, but not very much. Wikipedia says that it only works on SD content, so I guess it might work for his purposes but he'd really be better served with the PCI 8400 or HD 2400. He might not want HD video now, but if you're still upgrading that computer now the chances are that before he's done with that PC he'll begin to wonder why his HD videos play crappily on his new monitor. http://en.wikipedia.org/wiki/PureVideo#Generation_1_PureVideo

      --
      All your base are belong to Wii.
    27. Re:Grrrreat! by AmberBlackCat · · Score: 1

      How long will it take for true 3D acceleration to become an expected standard feature on PC's?

      When it no longer adds over $100 to the cost of the computer, and most users actually care.

    28. Re:Grrrreat! by hairyfeet · · Score: 1

      No IIRC the 915 is before the GMA900 and is really shitty. And the problem is that he is retired, so like I said he wants cheap above all, preferably $55 or less including shipping. Thanks for the info and I will pass it on. I just have to really watch the power requirements with this thing. Because this box is a SFF and has a really funky proprietary PSU and I shudder to think the amount of work I'd need to get done to fit a normal ATX in this thing. I'm thinking it is around 300w. I know my nephew has a 300w in his budget gamer box(don't look at me, his momma bought it) and it plays nice with the 6200 AGP I had left when I upgraded my GPU to a 7600GT. But I thank you again for the info and I will surely pass it on.

      Since you seem to know a little about GPUs and this is Slashdot, one question: My nephew's PC has an 8x AGP so my question is whether he would get a better experience with the 8400 PCI or with a 7600 AGP like mine? I'm afraid I haven't messed with graphics all the much, since I only play very lightly and my customers are mostly SMBs. So would the difference between the 8xxx and the 7xxx GPU be negated by the PCI bus, or will it still be worth the upgrade? I know the 6200 is getting long in the tooth and haven't had any trouble playing demos of the newer games at 1078x768(which is what he and I both run our displays at) but I haven't a clue as to whether the bus would kill the architecture gains from going 7xxx AGP to 8xxx PCI. And as far as I can tell there simply aren't any AGP card for the 8xxx series.

      Finally with the economy in the crapper I can't afford to completely shitcan his box and start over right now. So I have to stick with the DDR RAM and IDE drives which means AGP. I am planning to pick up this board to get him through until I can get enough ahead to build him a new machine from scratch, since his board is getting flaky and this supports his AGP, DDR, and CPU. But his PC has an old Celeron 3.06GHz in it and none of the newer boards supports it and with Xmas around the corner I simply can't buy a new CPU and RAM right now. So if anyone knows whether the PCI 8xxx or AGP 7xxx is better please let me know. Thanks in advance. Sorry if I come off as a noob but I've spent all my time working on office PCs and really haven't had time to get into gaming. I think the newest game I play is from 2004(Riddick) so on GPUs I'm afraid I'm out of my league.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    29. Re:Grrrreat! by i.of.the.storm · · Score: 1

      Oh sure, no problem. I'm fairly certain that the 8400 should work in a 300W PSU, since a variant of it powers the 9300 IGP version. SFF case kind of makes me shudder but I think those cards should be fine.

      As for your nephew, I would suggest getting anything AGP over anything PCI simply because it seems like PCI cards carry a definite price premium because they're not that common anymore. However, as for the 8400 vs 7600, from what I can see the 8400 is actually slower than the 7600. These charts are a good reference for comparing different GPUs since naming schemes don't make any sense these days. I just glanced at the 3DMark06 scores and the 7600 was nearly twice as fast as the 8400. I think the 8600 is basically a die shrunk version of the 7600, and the 8400 is a more cut down version of that, or something.

      For gaming graphics on AGP, these cards would be your best bet since as you mentioned nVidia hasn't made any of its 8 series boards on AGP. The 3450 will edge out the 8400 by a little bit so if you're on a budget that's probably your best bet. The 3650 however has three times as many stream processors so it's a bit over three times faster (higher clockspeeds too) so if you're willing to spend the $80 + shipping/tax go for that one. I'm not sure whether the Celeron would be a bottleneck for the 3650 but probably not since at 3.06ghz it sounds like it's a high end Celeron. I wouldn't recommend anything above the 3650 though. You could also look at the 7600GT AGP but the 3650 compared to the 7600 is significantly faster, and it seems like on newegg the 3650 is actually $5 cheaper. Here's a handy comparison between some of your options, although I don't know whether they are PCIe or AGP. I don't think there will be much of a difference at all between AGP 8x and PCIe x16 for those cards in any case.

      Good luck with your shopping, and I'm happy to help if you have any more questions. I follow graphics cards a lot more than I should and it's fun for me to do comparisons and stuff.

      --
      All your base are belong to Wii.
    30. Re:Grrrreat! by i.of.the.storm · · Score: 1

      Oh and the 915G is the chipset which includes the GMA 900, and I think 945G includes GMA 950. Both are not very good. http://www.intel.com/products/chipsets/915g/index.htm But it's a confusing distinction to make. I don't know which chipsets include the X3100 or X4500 but those at least provide decent DX9 performance.

      --
      All your base are belong to Wii.
    31. Re:Grrrreat! by hairyfeet · · Score: 1

      Thanks for the information, I always had trouble figuring ATI VS Nvidia, and after getting burnt by the Rage I haven't really messed with them. But I think the nephew will be getting one of those HD3s for Xmas, which one depends on how many customers I have between now and then. Like I said I need to slap a new mobo in while I am at it so I figure might as well get a card and kill 2 birds.

      Well the SFF has a first gen 915 and the performance in Windows is truly horrible. At the correct resolution for his new monitor anything more than 1 or 2 windows and it starts to have redraw issues bad. And videos just suck the big wet titty. So pretty much anything would help him a great deal. But seeing those prices I will try to talk him into the 8400. It says here that they max at 71 watts so he should be good. Like I said, thanks again. Anything that will get my nephew another year or 2 out of his PC is a good thing. Later on I can pick him up a 3.6GHz Cedar Mill like I am running and it should get him another couple of years with the HD3 card. Thanks again.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    32. Re:Grrrreat! by i.of.the.storm · · Score: 1

      No problem, like I said I spend too much time looking at graphics cards and I don't buy a whole lot of them (well, I've bought a grand total of one) so glad that my time spent fiddling around on sites like Tom's Hardware hasn't completely gone to waste.

      --
      All your base are belong to Wii.
    33. Re:Grrrreat! by Anonymous Coward · · Score: 0

      You're not doing it right!

      No nV 6200 PCI is going to help his video enough to be worth it...and there are plenty of onboard chipsets now that trounce that particular card in 3d and 2d video acceleration. Most are affordable, and work well with low power cpus.(I'm sure his dinosaur is eating enough power to make up for the cost of upgrading)

      You really need to do some more research if you think onboard video is still the picayune performance point Intel has stuffed down our throats for years.

      I realize his budget is limited...but all you're doing is setting him up for disappointment.

    34. Re:Grrrreat! by toddestan · · Score: 1

      I play videos full screen on a 1600x1200 monitor just fine using an ATI 9250 with 256MB on the PCI bus. It even works fine off of an external 1394 drive (using a PCI 1394 adaptor card) with the sound being piped out through a SB Live (also on the PCI bus).

    35. Re:Grrrreat! by toddestan · · Score: 1

      I would second the ATI 9250 recommendation. I have a few in service and they work fine. The GPU is fairly old and mature, it seems the bugs have pretty much all been worked out at this point. They also have pretty good support in Linux using the open drivers. I would tend to stay away from the never GPU's on the PCI (and AGP) buses, as they tend to be low volume products and don't tend to be tested well in older hardware and the vendors don't seem to care about bugs with them either.

    36. Re:Grrrreat! by toddestan · · Score: 1

      I have an ATI 3650 in AGP here, and I would advise caution when buying them. First problem is, that ATI doesn't officially support this GPU on the AGP bus, so you can't use the official drivers with the card. You can either try ATI's unsupported "AGP Hotfix" drivers which may or may not work depending on how they hacked the GPU onto the AGP bus, or you have to use the manufacturer's (such as Sapphire or whoever actually built the thing) drivers which can be horribly out of date.

      Anyway, the problem I had with the ATI 3650 AGP is that playing video, I got horrible tearing artifacts. It was usually between 1/2 to 2/3 the way down the bottom of the screen, and was really noticable. Interestingly though, it seemed to be a driver issue as I got the same artifacts using my old video card (ATI 9600Pro) with the AGP Hotfix drivers. I tried several versions of the drivers, both the latest, older versions, and the manufacturer's drivers and they all had the same problem. I eventually had to use a driver cleaner program to wipe out the drivers and install a regular version that supported my old 9600Pro to get the tearing to go away. I also couldn't get the hardware acceleration (the Unified Video Decoder) to work either using the AGP hotfix drivers, which was the reason I bought the thing. So I just went back to the ATI 9600Pro for now and the fancypants ATI 3650 is sitting on my desk.

      I did try the card in another old computer of mine (P4 1.5Ghz) for kicks and I did not have the tearing problems. Searching the internet it seemed thet generally those who was having problems like mine were also running a AMD Socket A systems (I have an old Sempron on an nForce2 board) whereas the people who did not have as many problems were running Intel systems of some sort, so it may be some kind of chipset/compatibility issue or maybe something to do with the driver assuming that the CPU has SSE2?

      Anyway, I did notice that in the latest Catalyst release, ATI says they addressed some video tearing issues, but they haven't made an AGP hotfix of that release (yet?) so I don't know if it will resolve my problems.

    37. Re:Grrrreat! by i.of.the.storm · · Score: 1

      Ah, I didn't know about that. Might want to reply to the guy who asked the question since he's the one buying. I just assumed that AGP versions worked the same as PCI-e. I guess he probably wants to do more research on AGP issues then.

      --
      All your base are belong to Wii.
    38. Re:Grrrreat! by Bert64 · · Score: 1

      What resolution is the source video?
      It's probably a low resolution video, being sent over the PCI bus at low resolution and then scaled by the GPU...
      Try playing 1080p video on that.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  9. Great news then... by tftp · · Score: 3, Funny

    Running Crysis at 800 x 600 with the lowest quality settings, an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics."

    So they compared one unusable (and dirt cheap) setup to another, super-expensive and still unusable one, and then they brag about sucking 20% less?

    This is typical for MS. They are mostly a software company, and there are too many people who advocate software-only solutions that make no sense, just because that's the only thing they know how to do (maybe.)

    1. Re:Great news then... by Anonymous Coward · · Score: 0

      you're 100% correct. we should side with software and hardware vendor lock-in. we should all switch to apple.

  10. can i do this? by sirmonkey · · Score: 1

    call "dee dee dee" on microsoft?

    or are they off limits? like real down syndrome people and such?

    --
    bored? try this http://jadmadi.net/blog/2005/01/27/linux-wine-how-to-running-windows-viruses-with-wine/
  11. For server use, I guess? by White+Flame · · Score: 1

    Many server motherboards come with some chintzy onboard video, yet have plenty of CPU and RAM to throw around.

    But who is going to be running D3D10.1 apps on a server? Is MS going to rewrite their GUI layers on top of their 3d API a la Apple?

    1. Re:For server use, I guess? by DigiShaman · · Score: 5, Insightful

      My guess is that Microsoft wanted their next OS to be virtualized on a server and yet still be able to run applications written for Direct-X.

      --
      Life is not for the lazy.
    2. Re:For server use, I guess? by SanityInAnarchy · · Score: 3, Insightful

      Is MS going to rewrite their GUI layers on top of their 3d API a la Apple?

      They did that in Vista. They did it so poorly that customers sued over being sold "Vista-capable" machines which weren't -- including Intel video cards that weren't enough.

      Meanwhile, Ubuntu runs on Compiz, which does just fine on Intel -- and Apple has been so far ahead that someone took the audio from one of the original Vista presentations, and combined it with video from Tiger, thus showing that really everything "new" about Vista was just playing catch-up with Tiger, while Leopard was just around the corner.

      More to the point: I believe it's now possible to run a Windows Server without a video card -- or, indeed, any GUI at all, depending on what apps you need.

      --
      Don't thank God, thank a doctor!
    3. Re:For server use, I guess? by smash · · Score: 1

      Aero runs just fine on my second machine from 2005. The aero-compatible bullshit is more to do with cheap hardware vendors being a little ambitious, than microsoft's product sucking arse.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    4. Re:For server use, I guess? by Bert64 · · Score: 1

      No, windows server still needs a video card, and it still loads the graphics layer..
      All you can do is have a copy of cmd.exe running in a window instead of explorer.exe.

      It's a far cry from linux/bsd/solaris/etc, which boots in text mode, doesn't load any video drivers and can run over a serial console on a machine which doesn't have any video capability whatsoever.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  12. Heck Yeah! by WiiVault · · Score: 0, Redundant

    This is awesome, I can finally run a game from last year at 7FPS on my brand new eight core machine that just came out a few weeks ago. Thanks MS! But wait, why not try the same thing on my 'ol 800 mhz P3 oh man does that baby have a whole new life!

  13. Well...I think it's kinda cool. by Antlerbot · · Score: 3, Insightful

    Say you get a new computer with a decent CPU, but no graphics card for work. You guys remember that thing, right? Work? Spreadsheets and documents and...yeah. That stuff.

    Anyway, now you can play Tomb Raider on it. The original one. Sweet.

    1. Re:Well...I think it's kinda cool. by Antlerbot · · Score: 1

      Or, you know. Call of Juarez. If you want to get all technical. Flight Simulator X. That sort of thing.

      Whatever.

    2. Re:Well...I think it's kinda cool. by LingNoi · · Score: 1

      Or you could stop wasting time and do some damn work!

      regards,
      your boss

    3. Re:Well...I think it's kinda cool. by The+MAZZTer · · Score: 1

      Actually I can play Tomb Raider on a virtual machine without needing any silly WARP.

      Also it didn't use DirectX, it was DOS based. I understand there are alternate rendering DLLs you can drop in to give it hardware acceleration on modern platforms, though.

    4. Re:Well...I think it's kinda cool. by Antlerbot · · Score: 1

      Yeah, hence my fixed reply. RTFA, yeah, yeah. :)

  14. ...and kills their own argument / lie... by Hymer · · Score: 4, Interesting

    ...about the impossibility of running DirectX 10 on Windows XP.
    If you can run it on software you'll be able to run it on any OS version. Gee... that was another lie from Redmond, why am I not surprised... maybe 'cause I do run he DirectX 10 hack on my XP and no it didn't raise the CPU usage (as claimed be the union of MS Windoze Vista Fanboyz)... it lowered it.

    1. Re:...and kills their own argument / lie... by Anonymous Coward · · Score: 0

      Unfortunately, too late, even if it makes it technically possible.

    2. Re:...and kills their own argument / lie... by Hymer · · Score: 1

      Unfortunately, too late...
      That is correct, at least from Microsofts point of view, but Win XP is still the most used OS in the world.
      It wouldn't have mattered earlier either, MS wouldn't have kept XP alive.

    3. Re:...and kills their own argument / lie... by Suddenly_Dead · · Score: 2, Insightful

      How does that even make sense? Not to defend Microsoft's bullshit, but how does coding a software renderer on one OS suddenly mean it should work with every OS? There's no possible logical leap there. Hell, why not DOS?

      maybe 'cause I do run he DirectX 10 hack on my XP and no it didn't raise the CPU usage (as claimed be the union of MS Windoze Vista Fanboyz)... it lowered it.

      What? There is no way to use DX10 on XP at this time; the only "hacks" are game-specific, allowing you to use DX10 games on DX9, or bump up the graphics detail on games when in DX9 mode to something closer to what they do in DX10 mode. All that proves is that these particular games don't actually need DirectX 10 to run, or that their DirectX 9 modes are being intentionally crippled.

    4. Re:...and kills their own argument / lie... by jimicus · · Score: 1

      ...about the impossibility of running DirectX 10 on Windows XP.
      If you can run it on software you'll be able to run it on any OS version.

      Gee... that was another lie from Redmond, why am I not surprised... maybe 'cause I do run he DirectX 10 hack on my XP and no it didn't raise the CPU usage (as claimed be the union of MS Windoze Vista Fanboyz)... it lowered it.

      It was always a lie. Basic computer science theory shows that.

      Now, if they'd said "We can't port DirectX 10 to XP because it requires a whole bunch of changes to the underlying graphics system which simply aren't practical to make", or "We can't port DirectX 10 to XP because the sales and marketing teams won't let us", either of these may have been closer to the truth. But would have been significantly less snappy, and when was the last time Microsoft explained the reasoning behind their decisions in any great detail?

    5. Re:...and kills their own argument / lie... by Hymer · · Score: 1

      so... I even don't know that hack, do you ?

    6. Re:...and kills their own argument / lie... by ozphx · · Score: 2, Insightful

      I know, they could have backported all of Vista to work on Windows 98! Some people said it wasn't technically possible! All they had to do was put the 98->XP upgrade and then the XP->Vista upgrade on one DVD. But they didn't....

      Oh wait... thats because its a _completely retarded idea_. Adding DX10 to XP would mean backporting a bunch of kernel mods, the new driver model, etc - which while "possible" would certainly be a hell of a lot of work.

      So shut the fuck up freetard and just buy the new OS... you had the old one for six fucking years, do you expect it to be supported forever?

      And no, those DX10 hacks don't "support DX10 on XP" - they emulate a bunch of crap, and they emulated it badly.

      --
      3laws: No freebies, no backsies, GTFO.
    7. Re:...and kills their own argument / lie... by Suddenly_Dead · · Score: 1

      so... I even don't know that hack, do you ?

      What?

    8. Re:...and kills their own argument / lie... by LingNoi · · Score: 1

      I think the GP was trying to focus on the fact that Microsoft lied from the beginning rather then suggest someone should get to work on implementing it in XP.

    9. Re:...and kills their own argument / lie... by abigsmurf · · Score: 1
      How the hell did you come to that conclusion? It's a software driver. To the OS it's probably little difference than a hardware driver.

      DX10 cannot run properly on XP, for one thing it can't do the the sound (Vista has an extra software layer designed to prevent crappy sound drivers crashing the system). What you're doing it Forcing DX10 and It looks at your system, see loads of features it can't do and puts it in a compatibility mode with lots of DX9 filling all the gaps.

      You can hack these things into any OS you wish but unless an OS is designed around these features, you have to put in extra software layers and code yourself in knots. The end result will be a slow, buggy implementation.

      You can fool yourself by pretending that because you can hack a game famed for it's inefficient coding and poor implementation of DX10 to give you a game that's dx9.5 and tell you it's dx10, that XP can do dx10 without issue but it won't be correct.

    10. Re:...and kills their own argument / lie... by raistlinwolf · · Score: 1

      What 'directx 10 on xp' hack, I'm imagine. Where's the link?

    11. Re:...and kills their own argument / lie... by Bert64 · · Score: 2, Insightful

      DirectX is middleware between the hardware and software, there's no reason you couldn't implement the frontend side of things, regardless of how it's actually handled on the back end... Just look at wine.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    12. Re:...and kills their own argument / lie... by Ash-Fox · · Score: 1

      Oh wait... thats because its a _completely retarded idea_. Adding DX10 to XP would mean backporting a bunch of kernel mods, the new driver model, etc - which while "possible" would certainly be a hell of a lot of work.

      It is entirely possible to get the DX10 API working on top of DX9 or OpenGL.

      I'm not talking about getting the hardware features working of DX10 cards, but the API it self, it was entirely possible for Microsoft to do and some people have even gone that far to try to implement it for Windows XP.

      --
      Change is certain; progress is not obligatory.
    13. Re:...and kills their own argument / lie... by Anonymous Coward · · Score: 0

      What directx 10 on vista? The only project I know of, died long ago. Quit lying moron.

    14. Re:...and kills their own argument / lie... by Pyroja · · Score: 1

      If you can run it on software you'll be able to run it on any OS version.

      QFT.

      Excuse me, I have to return to running PhotoShop CS4 in DOS.

      --
      [Trojan.]
    15. Re:...and kills their own argument / lie... by Anonymous Coward · · Score: 0

      Were you born an asshole ? Or did it take many years to prefect being one ?

    16. Re:...and kills their own argument / lie... by Suddenly_Dead · · Score: 1

      What 'directx 10 on xp' hack, I'm imagine. Where's the link?

      Like I said, the only "hacks" I've seen are for specific games, and don't actually allow you to run DX10 on XP. EX: Shadowrun / Halo 2, Crysis.

    17. Re:...and kills their own argument / lie... by Anonymous Coward · · Score: 0

      So, WARPed Windows is what you need to not see the Vista?

    18. Re:...and kills their own argument / lie... by csartanis · · Score: 1

      Link to the DX10 on xp hack? Google suggests most of the attempts never got off the ground.

  15. The software is STILL IN BETA.... by BulletMagnet · · Score: 1

    Or Alpha even ... when it comes out, they'll most likely crank up the FPS to something blistering in the neighborhood of 11 or 12 maybe....or if they're REALLY good, double it to 14 and some change!

  16. lol by DigitalisAkujin · · Score: 5, Insightful

    /. is silly

    they made this to run the desktop effects

    not crysis xD

    1. Re:lol by WiiVault · · Score: 4, Funny

      If so then why would they demo Crysis?

    2. Re:lol by RiotingPacifist · · Score: 1

      Arg the first insightful post but in such a stupid form!

      On the one hand I want you to get modded up, as on a modern system being able to use software to render desktop effects will be useful (I wonder what the FPS on aero is though). However I also wish you to die a horrible and painful death for using the subject "lol" and ending an unpuctuated post with an emoticon.

      --
      IranAir Flight 655 never forget!
    3. Re:lol by SanityInAnarchy · · Score: 1

      To prove that their implementation is complete, and doesn't completely suck, even if it mostly sucks.

      And, if you think about it, this could be good for Larabee, which is supposed to be just a bunch of x86 CPUs on a card.

      --
      Don't thank God, thank a doctor!
    4. Re:lol by Cyberax · · Score: 1

      Why not?

      It certainly shows that their software renderer is fast and feature-complete enough to run fairly recent games.

    5. Re:lol by RiotingPacifist · · Score: 1

      Do you run benchmarks copying one file from home to tmp and back again?
      Crysis is a good place to test dx10 performance in a way people will understand, would you rather they said they could do x thousand polygon operations per second vs n thousand?

      --
      IranAir Flight 655 never forget!
    6. Re:lol by WiiVault · · Score: 1

      I certainly see the point you are making, its valid, however I would never, as a company trot around how the fastest Intel CPU's available (just recently) with the most cores can produce a slideshow of Crysis at crap resolution and detail.

    7. Re:lol by WiiVault · · Score: 1

      Crysis came out in Nov 2007. The i7 is brand spanking new and hardly anybody buys (can afford) the 8-core. Please tell me this is a +1 Funny post. Nobody in their right mind would buy an 8-core CPU to play 3-4 year old games at 30+ FPS, or a year+ old game a 7.

    8. Re:lol by Cyberax · · Score: 3, Insightful

      DirectX 10 on CPU is _NOT_ intended for games.

      It'll be used for rendering the Aero interface. And it requires several orders of magnitude less computing power. Hell, even my old 4-year old ATI Radeon 9600 can render Aero just fine.

      Games make a useful test-case, though.

    9. Re:lol by Barny · · Score: 2, Interesting

      But then, isn't the whole point of aero (excusing the prettiness) to get load OFF the CPU and onto something else?

      --
      ...
      /me sighs
    10. Re:lol by SanityInAnarchy · · Score: 1

      Then what would you use for such a demonstration of the completeness of your software engine?

      --
      Don't thank God, thank a doctor!
    11. Re:lol by Cyberax · · Score: 5, Insightful

      Sure. But you also need good-quality 3D drivers. This way Microsoft will be able to run Aero even on plain VESA framebuffer.

      Also, consider this: the upcoming Intel Larrabee graphics card will consist of 64 independent programmable x86-compatible cores. NVIDIA CUDA also allows direct GPU programming.

      I bet this renderer will be adapted to run directly on such GPUs bypassing their 'native' rendering pipelines. That'll give Microsoft freedom to experiment with new feature such as ray tracing without any help from hardware vendors.

    12. Re:lol by Anonymous Coward · · Score: 0

      Vista Ultimate, of course.

    13. Re:lol by WiiVault · · Score: 1

      Geez perhaps a DX10 demo, floating cube or some such, certainly MS could conjure up something that would be less embarrassing.

    14. Re:lol by Anonymous Coward · · Score: 0

      Sure. But you also need good-quality 3D drivers.

      Yes and we used to have them before Microsoft fucked up the driver model!

      I bet this renderer will be adapted to run directly on such GPUs bypassing their 'native' rendering pipelines.

      WE ALREADY HAVE THAT!!!?! *facepalm*

    15. Re:lol by Anonymous Coward · · Score: 1, Informative

      amusingly, the larrabee drivers, ie the software renderer which runs on the larrabee cores, work in exactly the same way.
      also, the main guy behind those, is an ex-ms guy (abrash) who also previously has worked on a software DX renderer for RAD...

    16. Re:lol by vux984 · · Score: 1

      Geez perhaps a DX10 demo, floating cube or some such, certainly MS could conjure up something that would be less embarrassing.

      Right, because the world would be impressed with a floating cube... in 1990.

    17. Re:lol by JohnBailey · · Score: 1

      It certainly shows that their software renderer is fast and feature-complete enough to run fairly recent games.

      For a very impractical definition of "run". I get a sneaking suspicion that marketing has been telling the techs what to do again.

      --
      It is difficult to get a man to understand something when his job depends on not understanding it.
    18. Re:lol by Anonymous Coward · · Score: 0

      This way Microsoft will be able to run Aero even on plain VESA framebuffer.

      What's the point of that? Aero uses the GPU because it's unnecessary work for the CPU. If you don't have a fancy GPU from the last 5 to 10 years, you likely don't care about graphics that much in the first place. Last I knew, Windows had a software desktop renderer, and it won't take up anywhere near the CPU cost of what Aero using software DX10 would.

    19. Re:lol by cerberusss · · Score: 1

      they made this to run the desktop effects

      not crysis

      If so then why would they demo Crysis?

      Because Windows 7 will use Crysis as the default startup theme. Duh.

      --
      8 of 13 people found this answer helpful. Did you?
    20. Re:lol by drinkypoo · · Score: 1

      But then, isn't the whole point of aero (excusing the prettiness) to get load OFF the CPU and onto something else?

      NO. The goal is to increase prettiness. The fact that it requires 3d acceleration (when it clearly could have been implemented in software, albeit at a high cost in CPU time - as we have seen here) was an attempt to provide the prettiness at an acceptable level of performance.

      Aero is about eye candy, nothing more, which is why it is not as usable/featureful as what OSX has today, to say nothing of Compiz with Xgl (when it works) or whatever is replacing it (when it gets here.)

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    21. Re:lol by Bert64 · · Score: 1

      No, it's to provide eyecandy so that windows doesnt look ugly and dated when compared to OSX and linux based machines.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    22. Re:lol by SanityInAnarchy · · Score: 1

      Potential investor/customer/partner/whatever: "But can it run any real applications? How about actual games?"

      Never mind how visually unimpressive a floating cube would be.

      I suppose displaying Aero might be better, as that's what this is targeted at. However, I think what they've accomplished here is damned impressive -- and I think it's better to impress the intelligent people and look stupid to the stupid people, than the other way around.

      --
      Don't thank God, thank a doctor!
    23. Re:lol by euri.ca · · Score: 1

      I imagine a lot of people will be running exactly that on Windows 7, after Vista kind of failed the whole copying-a-file test.

  17. Wish they had done a press conference... by WiiVault · · Score: 5, Funny

    Just imagine the demo. "Here is the slooooow intel extreme, geez what a dog, they should be ashamed! Now check out the BRAND NEW straight out of the labs tech, this will blow your mind (cues 7fps slideshow). I know, I know, we do seriously kick butt.

    1. Re:Wish they had done a press conference... by garphik · · Score: 1

      ... And now that you have 7fps, you will just need a tiny bit of extra which you can get through a "Time Compressor" (easily available at your local science labs (joke)) Now the computation happens in a different dimension where time is 3 times slower that the dimension you are breathing in, so when the transformation takes place you don't even notice the difference. Thats easy.

  18. Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by WiiVault · · Score: 4, Insightful

    To think that anybody would want to run a DX10 game on an 800mhz no SSE CPU is insane, even considering the company involved. Perhaps for DX 7,8 and perhaps 9 games this might be reasonable (though not likely) but jesus, no thanks!

    1. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by aurasdoom · · Score: 1

      You seem to have missed the point.
      It's not meant to play games, but to run Direct X 10 based apps without a direct x 10 card.

    2. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by twomi · · Score: 1

      To think that anybody would want to run a DX10 game on an 800mhz no SSE CPU is insane, even considering the company involved.

      I want to see someone run DX10 capable Windows on a 800Mhz CPU first, then we can talk about running games on top of that!

    3. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by smash · · Score: 1
      this is notihng to do with running games on low end hardware. its more to do with enabling DirectX 10.x to be a standard that Microsoft can use for various other bits of software for the UI, office apps, etc.

      If they code to DX10, and it can be run in software, then any vista platform will be able to run that app.

      If you buy a system to run software directX to play games on, you're a fucking retard - that's NOT what this is about.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    4. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by WiiVault · · Score: 1

      Ok guys, would you want to run ANYTHING DX10 on a 800mhz GPU. Jesus.

    5. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by WiiVault · · Score: 1

      Correction----CPU that is.

    6. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by Joce640k · · Score: 2, Funny

      So you're saying the next Office will require eight cores to run? (and only be as fast as on an Intel IGP...)

      --
      No sig today...
    7. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by Anonymous Coward · · Score: 0

      Could be useful for a do-it-yourself render farm. A room full of shuttles or no-name mini-atx machines rendering out a DX scene would be useful. Not so much for real time gaming, I know.

      If you can render out a complicated scene this way and output to uncompressed video, you can compare the output of the reference implementation against that of the GPUs, which in turn means that you have an effective way to hunt down bugs (or graphics cheats). Or just show off a game at 'maximum quality' when you know no GPU can run it at those settings without going into meltdown.

    8. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by Spit · · Score: 1

      By the time windows 7 is out in 2011 the low end will be eight core with heavy system integration, like Sparc is now.

      --
      POKE 36879,8
    9. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by smash · · Score: 1
      You can read whatever you like into my comments, it doesn't mean that the inference was intended, or logical.

      In other words, no. But it saves microsoft doing 2 versions of the UI - simply write for DX10.x and it will run on whatever hardware is available.

      By the time the next version of office comes out, 8 core machines will be commonplace anyhow.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    10. Re:Imagine a DX 10 game on an 800mhz CPU -SSE/MMX by AmberBlackCat · · Score: 1

      He got the point. He's just hoping enough other people miss the point to get him modded insightful.

  19. It's truer than ever by FranTaylor · · Score: 4, Funny

    "Every time Andy gives us more power, Bill takes it away".

  20. impressive innvotation by Anonymous Coward · · Score: 0, Interesting

    Did we not do this already back in 1993? From the MESA project an excerpt: "August, 1993: I begin working on Mesa in my spare time. The ..."
    So what is the news.

    1. Re:impressive innvotation by Nagrom · · Score: 1

      I'd say this link would be a little more relevant: http://www.transgaming.com/products/swiftshader/

  21. Re:WOW! Someone buy microsoft a clue. by Pr0xY · · Score: 4, Informative

    Running Crysis isn't the point of the demo. The point was that it was a DX 10 application running entirely in software. In the end, this means that systems without higher end 3D cards would be able to run Aero. THAT's the point.

    They are trying to address the main complaint of the "Vista Capable" debacle. Running Crysis was just a way of demonstrating the capability.

  22. So does MS hate get an automatic upmod? by Sycraft-fu · · Score: 5, Interesting

    Seriously, this is a good thing. One could compare it to Mesa 3D. You have the option of running graphics in software, if you lack the hardware to accelerate it. This is highly useful in two situations:

    1) You have something intensive and need to see it on a computer that lacks the requisite accelerator. Though it won't be fast, at least you can see the output rather than just being SOL.

    2) You have a non-intensive task and don't wish to purchase dedicated hardware. While Crysis crawls, I'm going to guess something like, say, Thief wouldn't.

    This is just a software lawyer to allow the OS to do 3D rendering even if there's not an accelerator present. I'm sure that 99.99% of people who do 3D in any capacity will use an accelerator as they are extremely cheap and extremely high performance. However it isn't a bad thing to have a software implementation. MS has actually had one for a long time, however it only comes with the development version of DirectX. It allows you to check the expected output for a program against the reference renderer as compared to an actual card.

    Sounds like this is the same thing, just sped up and packed for end user use, rather than just developers.

    Could have applications in the future too. For example what will computer hardware be capable of in 15 years? Processors are likely to be much faster as compared to today. Well, this might allow for 3D to be useful when emulating Windows for old programs. People remember people emulate DOS today (see DOSBox) for various purposes. I don't think it is out of the question that a decade or two later people will emulate Windows 7. Ok however part of that will be dealing with the 3D layer. A large number of apps today make use of Direct3D. Well, if Windows 7 has a software 3D layer, and processors are blazing fast you are good. Just use that. If it doesn't you then have to make your emulator emulate the 3D hardware, since I'm guessing a decade from now the 3D subsystem will be vastly different than it is now.

    This is not intended to be a "Oh you don't need a graphics card ever," thing. It is intended to give people the option to get 3D without having to have a graphics card. It won't be as good, but at least it'll work.

    1. Re:So does MS hate get an automatic upmod? by Anonymous Coward · · Score: 1, Interesting

      Even with 3D hardware, this can be useful. If you have the previous generation 3D adaptor, you will be able to run 80-90% of the code accelerated, but you will still be able to see the effects only otherwise available on newer adaptors. This encourages developers to just go for the newest version of DirectX instead of having a chick and egg-problem, where there needs to be enough adaptors developed and sold before game developers start using that standard and vice versa.

      Apple is working on technology making this even better. They basically do runtime optimization of the code, so all the conditionals (use software or use hardware) are optimized away and the correct path just used, essentially allowing a single software implementation to automatically adapt itself to all adaptors.

      This, of course, is also beneficial for hardware developers, which can release feature-complete drivers for new adaptors much earlier, by simply not implementing difficult things and later releasing updated drivers.

    2. Re:So does MS hate get an automatic upmod? by UberMongoose · · Score: 1

      This seems more forward looking than currently practical. With intel talking about doing graphics completely on the CPU again, and Nvidia talking about more general purpose GPU's, the line between the two will be blurring quite a great deal. Allowing the directx stack to take care of where the code is actually executed is a nice way of allowing developers to concentrate on the actual work, without worrying about which hardware they want it to run on in x years time.

    3. Re:So does MS hate get an automatic upmod? by Anonymous Coward · · Score: 0

      wow, now if only Mesa3d weren't irrelevant buggy crap that will never see wide adoption...i'd be impressed.

  23. VNC by ChunderDownunder · · Score: 1

    Have a 'server' with a decent GPU and use the magic of TCP/IP.

    Performance surely couldn't suck any worse than running it locally?

    1. Re:VNC by Nagrom · · Score: 1

      In the sense that 0 == 0, no.

    2. Re:VNC by 4D6963 · · Score: 1
      --
      You just got troll'd!
  24. is this a "feature"? by the1337g33k · · Score: 1

    If only Microsoft would open source these things... Heck if DirectX was open sourced imagine how much farther Linux would have come in the gaming arena. Yet, there is wine and that is pretty damn close to an open source version of DirectX and windows altogether now. I mean its just as buggy as native windows, sometimes even less for me. Counter-Strike: Source doesn't crash wine like it crashes XP and Vista.

    Microsoft needs to fix bugs before putting in new "features"

    P.S: Where can I score one of them 8-core processors, that was the most interesting part of the announcement for me.

    1. Re:is this a "feature"? by bersl2 · · Score: 1

      The DirectX code would likely be a pile of dung. Most of the interesting things happen at the driver level and at the hardware level. After all, isn't Direct3D supposed to require hardware acceleration?

    2. Re:is this a "feature"? by pandrijeczko · · Score: 1

      DirectX doesn't need to be Open Source, there's already an open graphics programming standard called OpenGL that runs on just about any platform.

      DirectX is "better" for games companies because Microsoft are very good at documenting APIs that lock people into their products - so for games programmers, it's probably quicker to use DirectX than OpenGL.

      --
      Gentoo Linux - another day, another USE flag.
    3. Re:is this a "feature"? by LingNoi · · Score: 1

      Yes, problem is OpenGL just sucks and needs to be replaced by something better. I was hoping the new release of OpenGL would do it but here we are..

    4. Re:is this a "feature"? by S3D · · Score: 1

      The OpenGL is harder mostly because it was developed for client-server architecture, with all the functions merely wrapping for the commands to server. That caused OpenGL states model to be quite complicated. As OpenGL evolved it got more and more states and extensions, some of which work little differently for different platforms/manufacturers. The plan was for OpenGL 3.0 to break legacy support, clean mess, get rid of states and introduce nice objects instead. However the task proved to be too difficult, and after three (or was it two?) years Khronos Group produced specification with very little improvements on the classic OpenGL model. Though the rumors saying that the reason of this was that the manufacturers (NVIDIA nad ATI) could not agree on standards.

    5. Re:is this a "feature"? by Ash-Fox · · Score: 1

      Direct X being opensource wouldn't of done anything. It would be very difficult to adapt DX windows code to x11, Linux stuff.

      The DX APIs calls are fully documented for everyone though, and Wine has made significant progress in reimplementing the majority of DX.

      --
      Change is certain; progress is not obligatory.
    6. Re:is this a "feature"? by TheRaven64 · · Score: 2, Informative

      The client-server model of OpenGL works well because GPU programming is a client-server model - the application is the client, running on the CPU, and the server is on the GPU. You need to transfer data to a coprocessor and process it remotely. Direct3D does exactly the same thing.

      If you'd looked at the OpenGL 3 spec, instead of reading tabloid reports, then you'd see that it has some pretty major changes. The entire fixed-function pipeline is basically gone (although it can be emulated easily in shaders) and a load of stuff is marked as deprecated, and will be completely removed in the new release. There is a clean and simple subset of OpenGL 3 that is forwards-compatible, and another subset that is backwards compatible. Bringing out OpenGL 3 which was completely different to OpenGL 2 would have been pointless - why switch to it rather than another API?

      --
      I am TheRaven on Soylent News
    7. Re:is this a "feature"? by HunterKing · · Score: 1

      hey bersl2 it's Hunter King and I don't know how else to contact you other than hoping you'll check on old replies I'm semi in-charge of this http://neworleans.craigslist.org/grp/941917409.html It would be cool to have you there. The craigslist e-mail leads to mine

  25. Ummmm by Sycraft-fu · · Score: 4, Interesting

    3D accelerators are an expected feature on standard PCs. I can't think of one you can get these days without one. All the current integrated Intel and ATi and nVidia chips are 3D accelerators. Not powerful ones, but they do the trick. Any ad in card is, of course, an accelerator.

    However here's a better question: How long until we don't need that anymore? Personally, I'm not thrilled with the idea of having to have lots of dedicated hardware. The whole point of a PC is a general purpose machine that can do pretty much anything because it is all programmed in software. You replace dedicated units that did only one thing with a general purpose computer that does everything. Ok well that is somewhat undermined by the requirement of specialized hardware.

    Now, I understand the need for it. Graphics are intense and there is just no way, at this time, for a CPU to handle it. A dedicated processor optimized for the kind of math graphics need is the way to go. However wouldn't it be nice if that weren't the case? Wouldn't it be nice if the CPU again did everything?

    We won't see that day tomorrow, but perhaps we'll see it in a decade or two.

    I look back to the changes in audio production and hope to see it come to graphics as well:

    Originally, PCs used in audio production were little more than interfaces for complex dedicated hardware. A normal PC simply couldn't handle it. You had a PC that was loaded full of Pro Tool cards, which were massive bunches of specialized hardware, to do anything. Well as CPUs got better, you started to be able to do more on a regular PC. At first it was still nothing really useful in the pro market. You had to do everything non-realtime, spend lots of time rendering a change then listening to it and so on. But at least you could actually do it on normal computers. Yet more time passed and now non-destructive realtime software was available on normal systems. You could overload it pretty easy, you still had to bounce tracks and such, it wasn't the unrestricted power of an accelerated solution, but it worked pretty well and in fact lots of project studios did just that.

    Then we come to now. Now, the hardware accelerated audio production system is a relic. They are still made, but they are unpopular. Most professional studios don't bother, they just get a nice powerful PC (by PC I mean personal computer, Macs are included in this) with a couple of multi core processors and go to town. The CPUs easily handle large number of tracks with multiple effects and so on all in realtime. There is simply no need for dedicated hardware, and not using it means much greater flexibility. Everything is just changed in software.

    So, I'd love to see that same sort of thing come to graphics. At this point, CPUs have a long way to go. But then, technology moves fast. Everything I'm talking about in the audio world has happened in about 2 decades. In just 20 years or so it went from something you could only do with amazingly expensive special hardware to something that is easy for a $1000 computer to handle.

    20 years from now, may be the same deal with graphics.

    1. Re:Ummmm by NormalVisual · · Score: 2, Insightful

      Now, the hardware accelerated audio production system is a relic. They are still made, but they are unpopular.

      This isn't quite true. Certainly the mixing, EQ, effects processing and a lot of signal generation (softsynths, etc.) is done on board the host PC nowadays, but where the rubber meets the road and there's a need have to have really good sample-accurate synchronized input/output in real time without the possibility of clicks and pops, people are still relying on outboard hardware, usually in the form of a pricey rack-mount FireWire interface that's offloading a *lot* of effort from the host computer in addition to performing A/D/A conversion. It's not usually signal processing per se, but still necessary since none of the OS's used in audio production are hard realtime, and consequently can't maintain accurate timing without the help of that extra hardware.

      --
      Please stand clear of the doors, por favor mantenganse alejado de las puertas
    2. Re:Ummmm by Sycraft-fu · · Score: 3, Informative

      No, those soundcards aren't anything more than that. They just get the signal and convert it for the computer. I've owned a few, and worked with many more. Thus far I haven't seen any that do anything past conversion, routing, and perhaps basic mixing (basically those that have more advanced routing). Their function is to convert the sound to a format the PC can use and hand it off, nothing more. That they are external has nothing to do with it. That is done for convenience (hard to pack a lot of inputs on a PCI card) and noise (don't need to worry about dealing with all the RF from the computer). Firewire is often used since it has DMA and thus works well for low latency sound, but there's others that use their own PCI card and interface (MOTU does both, for example).

      Now I leave open the possibility there are ones I haven't encountered that do something more, but those I've seen are just soundcards.

      You forget that timing isn't an issue on the computer. Everything on there is asynchronous, clockless. The audio is just a stream of bits on disk. The computer never processes it at the sample rate, that is just a number stored in the file. So soundcards don't do anything special in this regard other than have a good clock to which everything is slaved (or perhaps a word clock input for external clocking). Once the audio has been converted and handed off to the system, timing isn't an issue anymore. The only difference between a cheap consumer card and an expensive pro card in this regard is the quality of timing source, and perhaps if everything is locked to a single source.

      In fact, you'll find that there is often more processing done on consumer cards, than on pro cards. Pro cards just convert the signal from analogue or S/PDIF or whatever and feed it in to the computer. Consumer cards often do sample rate conversion, and sometimes various other kinds of processing. In fact the card with the most muscle I'm aware of (leaving out dedicated hardware like the HDAccel) is the SoundBlaster X-Fi. That can handle 128 different sound sources in hardware, do SRC on all of them, place them in space, and perform effects on them. Compare that to a MOTU HD192 which does little more than deal with audio input and output, and mix/route it as you specify.

      The money/hardware in pro cards is in high quality circuitry, mostly in the analogue phase, not any special processing.

    3. Re:Ummmm by Targon · · Score: 1

      AMD has been working on their Fusion project for a while now, where there will be a GPU core on the CPU. I don't know what the current status of that project is at this point, but the AMD Fusion brand has been expanding, so I suspect it may be here sooner rather than later.

      When it comes down to it though, Microsoft has already had this in mind when they released Vista. The whole "performance rating" benchmark in Vista is based on a certain "minimum" and "recommended" rating. Having DX 10 in software means that the overall rating of a system can be used to counteract the graphics score being sub-standard. To be honest, I wish Microsoft would implement one of these for all versions of DirectX, and the compatibility tab would allow older games to run properly since changes to DX can cause undesired side effects.

    4. Re:Ummmm by Marwood · · Score: 0

      I think you've just described the decades old concept of The Wheel Of Reincarnation

    5. Re:Ummmm by klapaucjusz · · Score: 1

      In just 20 years or so it went from something you could only do with amazingly expensive special hardware to something that is easy for a $1000 computer to handle.

      Audio is far from being the only areay where hardware acceleration became unnecessary and disappeared. Intel processors still have instructions for accelerating Cobol decimal number processing (AAA etc.).

      20 years from now, may be the same deal with graphics.

      No, it will take four years at most. I bet for 2012.

    6. Re:Ummmm by Anonymous Coward · · Score: 0

      > where there will be a GPU core on the CPU

      The problem is, unless they extend the memory bus to be 512 bits (which will lead to extremely expensive mainboards, I would guess the cheapest would then be > 100 $/EUR compared to 25 today, probably have bad performance with normal CPU load and more expensive, incompatible memory) performance will still be useless.
      A measly 6-12 GB/s just can not compete with 120 GB/s, and you can't really get the 120 GB/s with a standard CPU memory interface.

    7. Re:Ummmm by NinthAgendaDotCom · · Score: 1

      Actually, timing *is* an issue on computers. If you go to any music production / home recording type forums, you'll see that trying to get input and output latency as low as possible is a big topic.

      --
      -- http://ninthagenda.com/
    8. Re:Ummmm by Chandon+Seldon · · Score: 1

      No, it will take four years at most. I bet for 2012.

      In 2012, AMD is talking about shipping 8 and 12 core processors with DDR3 RAM. I haven't looked at the Intel roadmaps, but let's give them the benifit of the doubt and say they're shipping shipping 16 core processors with each core twice as fast as a Core i7 core today.

      Assuming everything scales linearly, that's about 30 fps at minimum graphics settings in Crysis - and that's for a top of the line single socket workstation. My bet is that in 2012, gamers will still buy $200 video cards and mainstream machines will have quad core processors with reasonably powerful dedicated graphics hardware on-die.

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    9. Re:Ummmm by Targon · · Score: 1

      You seem to have missed that current systems tend to use at LEAST two channels when connecting to memory. Fusion will in no way be competition for systems with a dedicated video card, but if the graphics portion of Fusion is good enough, you will end up without needing a graphics controller on the motherboard. So, you end up with a new tier in the overall levels of systems.

      Fusion will basically allow for low-end chipsets without graphics to hit the market that will still provide better graphics performance than what you see from a system with Intel integrated graphics. These systems may very well be cheaper and due to the graphics issues faster under DirectX 10 than you may expect. The CPU may be a bit more expensive for the CPU power, but with the decreased chipset/motherboard cost, the overall system cost will probably be lower.

      Current integrated graphics will never compete with a dedicated video card, in much the same way that a GPU core added to the CPU will never compete with a dedicated video card. A GPU core added to the CPU MAY compete well with integrated graphics though.

    10. Re:Ummmm by Anonymous Coward · · Score: 0

      The Future is Fusion...

    11. Re:Ummmm by Sycraft-fu · · Score: 1

      That's totally different from timing sync issues. In fact, you'll find that many integrated audio cards are capable of extremely low latencies when used in WDM/KS mode. A lot of pro software doesn't support WDM/KS so it wasn't something widely known but ASIO4ALL, which is basically an ASIO-WDM bridge revealed that many integrated cards are extremely low latency when asked to be.

    12. Re:Ummmm by Anonymous Coward · · Score: 0

      Before you go off on how great software audio is, consider this: Good soft-audio requires around 10-15% (or more) of your CPU for 3D and environmental effects. Even a decent integrated hardware audio solution only generally cost $15USD. A 10-15% faster CPU costs $750+. Of course what really happened was Intel integrated another CPU core to handle it when the megahertz race ended due to heat problems. Yes your 2nd core can be used for non-audio purposes, but if you are using audio you basically got suckered into buying the most expensive audio solution of all time.

    13. Re:Ummmm by LandDolphin · · Score: 1

      So in 4 years a new computer will be able to play a 5 year old game?

      --
      Spelling and Grammar errors have been added to this post for your enjoyment
  26. manycore apps by Anonymous Coward · · Score: 0

    Glad to see we have found a way to consume those last cycles. Bravo!

  27. Seriously... by deanston · · Score: 1

    ... the marketing department came up with the name - WARP - before they started creating the software, right? Now Ballmer or Ozzie can stand up on stage some point in the future and say, "... and for Windows 7 (SP2) we introduce WARP 11! Even the USS Enterprise couldn't go that fast! (..hey I bet the geeks will love us now, yeah!)

  28. 42??? by ryzvonusef · · Score: 0, Offtopic

    Dear God, its, like, the answer to life, the universe and everything, man!!! Way to go MS, you are the second coming of Deep thought! (sorry, couldn't resist...)

    --
    I am an ACCA student. Got a query on Accountancy/Finance? Maybe I can help!
  29. It got 7fps on Crysis... by mseidl · · Score: 1

    but, couldn't run Aero. Lets see... Where did I put those "Vista Capable" stickers?

    1. Re:It got 7fps on Crysis... by LingNoi · · Score: 1

      There is of course a difference between "capable" and "practical".

  30. Re:WOW! Someone buy microsoft a clue. by pandrijeczko · · Score: 1

    The point was that it was a DX 10 application running entirely in software. In the end, this means that systems without higher end 3D cards would be able to run Aero.

    I don't get this statement at all.

    Why would you need a high end graphics card to run a desktop GUI? If you're a gamer with a good graphics card, then if you're into eye candy there's no reason not to turn up the settings on Aero to "full" to get a cool looking desktop while that graphics card isn't dealing with shifting game frames.

    But if you're someone who writes letters, surfs the Internet and uses whatever Windows is on the new PC you buy, why would you buy a high-end graphics card just to be able to do that?

    And what about server systems? What about if, as a server admin, you just want a slim GUI to run some admin tools on but nothing more graphics intensive than that.

    Personally, I slim down any GUI I use to be as fast and light as possible (whether Windows or Linux) so eye candy does nothing for me - but I get that other people like it and if they've already got the machine to cope with it then why not.

    But what you're saying really doesn't make any sense (unless I'm missing something), especially when we're also being told that Windows 7 is going to be a lot more modular.

    --
    Gentoo Linux - another day, another USE flag.
  31. Virtualization by Anonymous Coward · · Score: 0

    So MS is doing this to avoid a "Vista ready" disaster with Windows 7?? That makes almost NO SENSE.

    Windows 7 is going to be what, 3, 4, 5 year after Vista?? Even the shittiest integrated chipsets fully support Vista now, and by proxy, Windows 7.

    However, this will be GREAT for future virtualization. 10 years from now, Linux with virtual Windows 7 on a thousand core computer will be able to run Aero, whoopie!

  32. Why wait? by Krakadoom · · Score: 1

    Hm two things. If this is so great, why are they waiting to incorporate it in Windows 7 - why not add it to Vista to try and regain some momentum for that pile of fail?

    The other thing to note is probably that it's a decent use of all the extra idle cores on newer systems - as long as programs aren't taking advantage of multicores more than they are, why not use half the cores to do graphics processing?

    Will it somehow 'stack' with dedicated GPUs?

  33. Seriously? by Verble · · Score: 1

    No one else has commented that WARP 10 is impossible? Even if MS pulls off WARP 10, it'll just turn them into some weird lizard thing. Though for Ballmer that could be a good thing.

    1. Re:Seriously? by cnettel · · Score: 1

      Well, we're before the 24th century, so it's the n^3 scale, i.e. warp 10 being 1000c.

    2. Re:Seriously? by jmobley · · Score: 1

      Thank you for reminding me of that terrible Voyager episode. Do you know how much Vodka it took to forget I ever watched that train wreck? :(

  34. Oww it hurts! by Whiteox · · Score: 3, Funny

    Running Crysis at 800 x 600 with the lowest quality settings, an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics.

    and this is ball-slapping good news?

    --
    Don't be apathetic. Procrastinate!
    1. Re:Oww it hurts! by Legion303 · · Score: 2, Insightful

      Man, that's like 2 whole fps more. With further optimization they might even crank it up to 15fps, which would get it close to the framerate I got from Crysis on medium settings with my laptop. And the best part is, you can run it on your enterprise-class server when you aren't busy serving up hundreds of thousands of SQL searches! Why pay $400 for a lousy video card when you can buy a $20K server instead?

  35. This is a swap file on a ram disk. by Saysys · · Score: 1

    The point of the DX10 graphics in windows is to look cool and offload some of the processing onto the 3D card. It reminds me of putting the swap file on a ram-disk, wasting system resources to do something meant to get around having limited system resources.

    That said, at least we won't see nearly as many lawsuits because of varying levels of "windows 7" compatibility.

  36. Re:WOW! Someone buy microsoft a clue. by pandrijeczko · · Score: 1

    The core problem with Microsoft is that it has absolutely no concept of competition or how to work with other companies in partnership.

    I work for a US-owned telecoms company that is already well established in business PBX, call-centres, Voice-over-IP and voice messaging.

    Last week, we had a team meeting and one of my fellow consultants gave us a quick overview of Microsoft's Office Communication Server. To cut a long story short, rather than embracing the strengths of their own corporate desktop presence and working *with* telecoms vendors like us to integrate with our systems, they're actually going all out to lock us out with proprietary codecs (for VoIP stream encoding & decoding) rather than using open codecs that already do the job perfectly well.

    The problem is that Microsoft is trying to control far too many new markets rather than only focusing on what it does well - this means that it's spreading itself far too thin when it's trying to take on the already big players in those markets; Google is a classic example of this.

    I don't see the logic behind it at all - look at the Internet in 2008 and compare it to, say, 1998, and the most popular and liked applications and protocols are those where the creators have embraced and used open standards (okay, with the exception of iTunes and AAC formats maybe).

    I don't get it with Microsoft at all...

    --
    Gentoo Linux - another day, another USE flag.
  37. "WARP"ing back in time? by Anonymous Coward · · Score: 1

    When I think of WARP I still think of OS/2 Warp. Seems fitting considering the reaction I'm seeing to this news.

  38. Re:WOW! Someone buy microsoft a clue. by Sparks23 · · Score: 3, Informative

    Because, quite frankly, people were upset that their 'Vista Capable' computers couldn't run Vista with Aero enabled. The integrated cards don't have the 'oomph' for Aero's glassy transparency effects, but Microsoft had tooted the horn of 'Look! Shiny!' loud and long, so people expected that functionality. In addition, there are other places extended graphics capabilities are used (the Vista DVD maker program, for instance), where if your card isn't up to snuff, you can't use those programs.

    By showing 'we can make this work in software, slowly, but work,' they're trying to address that. This isn't for gaming, despite the demo. This is an attempt to solve the problem out of the gate in Windows 7 so that they don't have another Vista Capable type class action suit.

    --
    --Rachel
  39. But that's the GOOD news! by Anonymous Coward · · Score: 0

    we can play things at 7fps with ultra low settings

    The bad news is that they were rendering a black cube on a black background. Color and lighting are expected for the new version, at lower fps.

  40. Bad Summary! by AlgorithMan · · Score: 1

    I already read about this in a different news mag - and the main purpose of this thing is FOR DEVELOPERS to test the rendering of their stuff on cheap, low-end PCs (which are often used for development)

    --
    The MAFIAA is a bunch of mindless jerks who will be the first up against the wall when the revolution comes
    1. Re:Bad Summary! by Joce640k · · Score: 2, Insightful

      Which part of "eight core machine" is cheap and low end?

      I'm sure a $50 graphics card is cheaper (and would whip this things ass).

      --
      No sig today...
    2. Re:Bad Summary! by Spatial · · Score: 1

      That's pretty poor reporting. There are no eight core i7 models. They're all hyperthreaded quad cores.

    3. Re:Bad Summary! by petermgreen · · Score: 1

      I find it unlikely that a company doing any serious 3D development work wouldn't buy thier devs at least half-decent graphics cards.

      It seems to me that the main point of this driver is to allow non-gaming DX10 stuff (e.g. aero and whatever it's windows 7 successor is called) to work on machines that don't have hardware support (think virtual machines, remote desktop and so on).

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  41. Yay! by zmollusc · · Score: 2, Insightful

    Hurrah! In the future, when i switch off pointless Aero crap, it will free up lots more cpu cycles for the annoying microsoft apps i need to run to see simple 2d spreadsheet data sent to me by retards who use proprietary microsoft file formats. Microsoft FTW!!

    --
    They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
  42. oh! I see.... by Anonymous Coward · · Score: 0

    so after removing the software renderer from DirectX around ver7/8, they now plan to put it back?

    woooo...way to go guys..your back to where you where last century...ROFLMAO

  43. what it really says by Bizzeh · · Score: 1

    this is yet another miss quoted summery from someone who is just trying to slam microsoft... what it actually says in TFA is that DX10 in windows 7 CAN run on the gpu OR the cpu alone, but, in most instances, will use both together to get the maximum effect required.

  44. solution in search of a problem by Tom · · Score: 1

    So there's still a geek or two left at M$, as it seems.

    Aside from the cool factor, what exactly is the purpose of this hack? Nobody wants to play Crysis at 7fps in the real world, and dedicated 3D hardware is moving into even the smallest devices. 10 years ago, when notebooks didn't have 3D cards, this would've been commercially interesting. In 2008, it's a simple "because we can" hack.

    --
    Assorted stuff I do sometimes: Lemuria.org
    1. Re:solution in search of a problem by Anonymous Coward · · Score: 0

      3D acceleration and the Aero interface will now work within VMWare clients.

      Microsoft perhaps see some value in this. Branding is important to them and being able to have a common look&feel for folks working with both virtualised and native machines has perhaps been worth the effort of creating WARP.

    2. Re:solution in search of a problem by SpinyNorman · · Score: 1

      Right now it may only make the less complex games playable (maybe also enabling the full Vista aero UI without a graphics card), but given the pace of processor speed improvement that will change. Double the CPU clock and the FPS doubles, double or quadruple the number of cores and the FPS again goes up by the same factor...

      I don't think it's be too long before most games have playable frame rates using this software renderer, even if not at the highest resolution, or with the highest quality settings enabled, but it'll be a process of continual improvement. In the meantime even if the most demanding games arn't playable under this, there's much other less demanding software for which it'll be immediately useful, and given that even the most-stripped down budget machines still have todays insanely fast processors (even if not the bleeding edge), it brings an ever increasing level of capability to the lowest level.

    3. Re:solution in search of a problem by Tom · · Score: 1

      Double the CPU clock and the FPS doubles, double or quadruple the number of cores and the FPS again goes up by the same factor...

      Meanwhile, GPU speeds have been growing faster than CPU speeds for years, and the requirements of games will grow with them, i.e. faster than CPU speeds grow.

      --
      Assorted stuff I do sometimes: Lemuria.org
    4. Re:solution in search of a problem by Tom · · Score: 1

      3D acceleration and the Aero interface will now work within VMWare clients.

      With adoption of Vista being in the single-digit percentages, I can only imagine where it is among VMWare users, where resources are even more limited. I do run XP in VMWare Fusion every once in a while. The thought of running Vista in there (and DX10 is Vista-only, remember?) did never so far cross my mind.

      (yes, /., it's been less than a minute since I posted my other comment. Some of us don't do peek-and-type anymore! Grow up already!)

      --
      Assorted stuff I do sometimes: Lemuria.org
  45. Larrabee? by WittyName · · Score: 1

    Seems mighty useful on an 80 core X86 coprocessor..

    Maybe 560 fps would be overkill, but:
    140 fps at 1600x1200
    35 FPS at 3200*2400

    And Win 7 will supposedly ship next year, but with schedule slip, etc.
    Lets figure 2010. About the time for larrabee..

    --
    The law is a weapon of the government, not a protection for the likes of you. Surely you understand that.
    1. Re:Larrabee? by poetmatt · · Score: 2, Funny

      Still won't be vista capable.

    2. Re:Larrabee? by badkarmadayaccount · · Score: 1

      Ah, yes, but x86 is not a Scalable Processor ARChiteture with descent SIMD instructions and monster RAM bandwidth, as any true GPGPU would require. BTW, wouldn't it be better if they all had some sort of generic interface, possibly an Instruction Set based Architecture?

      --
      I know tobacco is bad for you, so I smoke weed with crack.
    3. Re:Larrabee? by Anonymous Coward · · Score: 0

      Descent SIMD? Bandwidth?!? Have you seen the Larrabee slides floating around?!?
      Try googling a bit..

      X86 is a generic interface. Several companies make chips that use it. It is also
      emulated widely. It can be implemented in RISC, and CISC styles. How much more generic do you want?

      The main improvements I see compared to current GPGPU's would be a common instruction set, and proper 64/80 bit floats! And x86 has these..

  46. A good feature and still the endless bashing by abigsmurf · · Score: 2, Interesting
    How dare MS maximise compatibility for Windows 7 and implement what will be a handy feature for low end systems, particularly netbooks (it's the chipset that draws all the power in atom based systems, not the CPU).

    Improving performance over a dedicated graphics chip (albeit a weak one) is still a respectable achievement, especially when you consider games typically use ~100% cpu anyway. Whilst it may be unplayable for crysis, I can see it giving a solid frame rate on things like WoW.

    1. Re:A good feature and still the endless bashing by miknix · · Score: 1

      How dare MS maximise compatibility for Windows 7 and implement what will be a handy feature for low end systems, particularly netbooks (it's the chipset that draws all the power in atom based systems, not the CPU).

      Yes, but CPUs also drain less power when running at lower frequencies (speedstep, cpuscaling whatever) and low cpu load.

      Transferring the graphical bloat to the cpu will make the CPU power saving management less efficient.

      For example, word processors don't require much from a CPU. Although, with software rendering you will get CPU usage spikes every time you open a menu (think on the shadows, anti-aliasing, ..), scroll the document, etc..

      So IMHO that's no good for laptops, specially when integrated graphics solutions are available and low-end graphic cards are cheap.

    2. Re:A good feature and still the endless bashing by basicio · · Score: 1

      Dedicated graphics cards drain (lots) of power, too. The advantage this has is that it doesn't have an additional hardware requirement--fitting a dedicated graphics card in a netbook isn't generally something that works very well.

      As the parent up from you said, chips like the Intel Atom draw *very* little power themselves, and it's Intel's bad chipset design that is accounting for most of the platform's power draw. CPU power saving doesn't come into it all that much.

    3. Re:A good feature and still the endless bashing by ceoyoyo · · Score: 1

      Did you not even read the summary?

      Their software renderer performs barely better than the worst integrated graphics they could find, when running on an 8 core i7 system. That's not a respectable achievement. They just picked a computer that was big enough to give them the result they wanted.

      The point of this thing is to fill in missing features (poorly) so that aero or aero2 or whatever will technically run with any video card, even if it's so slow you've got plenty of time to get up and pop some blood pressure pills while waiting for the start menu to open.

      The availability of Windows XP will be extended yet again to take care of (well, maintain a presence in) the low end, netbook market. As you might have noticed, 8 core i7 >> any-netbook processor. Therefore, crappy-integrated-graphics >> any-netbook-running-software-DX-renderer.

  47. Heres the big picture: by moniker127 · · Score: 1

    This is a huge step in the right direction. Keep in mind that win7 supports up to 256 cores. By the time we have that many, we wont need video cards any more.


    In 10 years, gamers will tout their specs by saying "Oh yeah, well I get 200 fps on wow4 with my Intel c256 8g. Suck on that!"<br>

    If I held stock in Nvidia id sell it now. :D

    1. Re:Heres the big picture: by afidel · · Score: 1

      Perhaps, but the G92 in my graphics card already has 96 cores and they are significantly more powerful for graphics than a general purpose CPU core. Do I need to mention that the G92 launched 2 years ago? The GTX280 has 240 stream processors so they are roughly doubling ever 18 months as expected.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    2. Re:Heres the big picture: by Casandro · · Score: 1

      Well I guess by then, the main problem will be memory bandwidth. So it doesn't matter how fast the single core is, it's the memory bandwidth which is the bottleneck.

      General purpose CPUs just are cheaper as you can also use them for general purpose computing.

  48. Re:WOW! Someone buy microsoft a clue. by LingNoi · · Score: 1

    You do realise that Microsoft has been about control from the beginning right? The Microsoft master plan is total domination over everything computer related.

  49. In other news.... by Joce640k · · Score: 1

    Intel does real time ray tracing on an eight core machine.

    http://blogs.intel.com/research/2007/10/real_time_raytracing_the_end_o.php

    --
    No sig today...
  50. Lie? by Anonymous Coward · · Score: 0

    They're not going to cater the very small group of high-end PC gamers. Modify the kernel to satisfy the wishes of 1% of all Windows users? Not worth it.

    1. Re:Lie? by Bert64 · · Score: 1

      Which is why a single general purpose OS isn't good from a user perspective...
      Complain all you like about the multitude of linux distributions, but there can easily be one that caters to the 1% of users who need a particular configuration.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  51. Colour me unimpressed by Anonymous Coward · · Score: 2, Insightful

    "an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics"

    So a $5 single cheap chip runs crysis slower than an 8-core $1000 200W behemoth CPU?

    Wow. Cheap affordable 3D is here!!!

    FFS.

  52. BUT... by Anonymous Coward · · Score: 0

    but will it run Crysis?

  53. tim sweeney predicted this by Anonymous Coward · · Score: 0

    "If this is a case, then the 3D hardware revolution sparked by 3dfx in
    1997 will prove to only be a 10-year hiatus from the natural evolution
    of CPU-driven rendering."

    http://arstechnica.com/articles/paedia/gpu-sweeney-interview.ars

  54. Re:WOW! Someone buy microsoft a clue. by drsmithy · · Score: 1

    In the end, this means that systems without higher end 3D cards would be able to run Aero. THAT's the point.

    Then it's pointless. Bottom-of-the-line 3D cards have been capable of running Aero for years. Any system with enough CPU grunt to emulate DX10, will have a video card in it that can run Aero.

  55. So How Can MS Use This? by Anonymous Coward · · Score: 0

    To fsck over Linux?

  56. I'm just waiting for the the patent to be awarded by mike_sucks · · Score: 1

    "Novel method for rasterizing three dimensional scenes in absence of specific hardware support."

    /Mike

    --
    -- "So, what's the deal with Auntie Gerschwitz et all?"
  57. Meh by Vexorian · · Score: 1

    You know, it would have been possible to make aero not use 3d in case there is no acceleration available, now you have UI that requires 4 cores and will still be slow, great compatibility!

    --

    Copyright infringement is "piracy" in the same way DRM is "consumer rape"
    1. Re:Meh by ceoyoyo · · Score: 1

      It would have been possible to make Aero not use DirectX 10, in which case pretty much every computer actually capable of running the rest of Vista would run Aero just fine.

  58. Well, now if someone other coded that.... by Casandro · · Score: 1

    Seriously, if the coding quality is the same as with Microsoft's MPEG4 decoder, we can expect alternative software 3D engines to get _way_ faster.

  59. Don't fracture a dilythium crystal by MrKaos · · Score: 1
    Warp 8.2 Warp 9.0 Warp 9.0a Warp 9.0b Warp 9.0c Kirk: Scotty, Weee neeeed Mooorreee Poooowwwweeeerrr Scotty: I've given yeah all she's got Cap'n , Ye canny Break the laws of physics

    Which is why you will never see warp 10. Obviously no-one at Microsoft has seen Star Trek otherwise they would already know this is impossible. Or perhaps they are trying to tell us something

    --
    My ism, it's full of beliefs.
  60. GPU+CPU working together on graphics? by Anonymous Coward · · Score: 0

    so, would I be able to play DX10 game on my DX9 card? It should do things GPU is not able to do on CPU, right?

  61. Please explain then by Anonymous Coward · · Score: 0

    How some people have managed, without access to the build source and tools, to get DX 10 working on XP?

    If they have to backport kernel mods and all that sort of crap to the extent of being unable to do it with free and complete access to the build process of XP and access to the people who wrote it, how did a bunch of people without these resources manage???

    Possibly because that argument is a load of bullshit

    PS Why is MS so cock-a-whoop about this? Why, even more strangely, are people on /. who should know better so impressed. Ever heard of Mesa? Software rendered. DX has a software render mode. Hell, DX5 (IIRC) is now ONLY handled in software because the API calls for DX5 are available in the DX8+ driver system.

    It was ALWAYS possible to run DX is software. So why the big song and dance here and now? That they've finally decided to put a software render call for ALL API calls for DX10 in the driver?

    PS Why is the label on how long you've waited gone? This post has waited 20 minutes about and still demands that I wait because /. doesn't like quick posting.

    1. Re:Please explain then by abigsmurf · · Score: 2, Informative
      Because Microsoft, amazingly enough, realised there were going to be plenty of cheapo DX10 cards which aren't fully featured, it would run the unsupported elements using DX9 functions or emulation taking a visual or performance hit. These hacks almost certainly cause DirectX to look at your DX10 card in XP and all it will be able to see is a cheapo card which can't handle a lot of functions.

      With XP you're never going to have full DX10 support. The kernel can't physically do a lot of the functions itself. With DX10, Microsoft wanted to give developers a whole new framework without having to worry about legacy DX code.

      Get someone to code a tech demo comprised of nothing but DX10 specific functions (the large texture sizes etc.) and you'll be able see the difference. It's hard to tell the difference at the moment because stuff like Crysis implements DX10 poorly.

    2. Re:Please explain then by Ash-Fox · · Score: 1

      Get someone to code a tech demo comprised of nothing but DX10 specific functions (the large texture sizes etc.) and you'll be able see the difference.

      I have, I didn't see a FPS increase at all doing things from spinning cubes to cloud formations.

      Infact, running a clean windows xp, windows vista install on the same hardware with my various tests I found:

      1) Better FPS under Windows XP with DX9.
      2) FPS not showing any difference with Vista's dx9->dx10 compared to the demo that was written using dx10 functions only.
      3) DX10 demo was slower than DX9 demo running under Windows XP.

      Same logic used, same features used from both.

      I also did not perceive any better quality from DX10.

      If it can't do simple things better, faster. How is it going to handle the more complex stuff better?

      --
      Change is certain; progress is not obligatory.
    3. Re:Please explain then by ozphx · · Score: 2, Informative

      If it can't do simple things better, faster. How is it going to handle the more complex stuff better?

      I guess you failed CS 101. A more sophisticated API is always going to be slower than "poke xxx".

      Graphics hardware is moving from being a specialised device which can handle basic primitive drawing to a full fledged massive vector processor. Doing this makes it less efficient at the original task.

      Your time might be worthless, but in the real world developer time costs money - and at the rate of hardware improvement its pretty clear that MS and its customers are happy to take a few % performance hit to have a more featured/safe/simpler/etc API.

      v10 brought much better memory management and reworked the shader model to add geometry shaders. The smallish hit in state management / data stream overhead is made up by the fact that half the procedural geometry can be done in the GPU.

      v11 will introduce compute shaders. Presumably the older API functions will be slower on equivalent hardware again. But I'll have compute shaders to play with (and it won't be by using some dodgy ass API which is the equivalent of poke).

      Only counterstrike tards care about getting 100fps vs 110fps. Hint: That shiny new intel quad core would get smoked at performing FFTs by an ASIC. Does that make the Q6600 the lesser CPU because its traded raw performance on certain tasks for generalisation?

      TBH I don't get where all this whinging is coming from. I have a 9600GT, its one of the cheapest cards you can get, it runs everything fine under vista? Why the butthurt?

      --
      3laws: No freebies, no backsies, GTFO.
    4. Re:Please explain then by Ash-Fox · · Score: 1

      I guess you failed CS 101. A more sophisticated API is always going to be slower than "poke xxx".

      A API could also be implemented more efficiently than it's predecessor, granting better performance.

      Graphics hardware is moving from being a specialised device which can handle basic primitive drawing to a full fledged massive vector processor. Doing this makes it less efficient at the original task.

      I'm really not interested in the excuses. As a developer, I find it slower. I don't see these great improvements at all that you initially stated.

      Your time might be worthless, but in the real world developer time costs money - and at the rate of hardware improvement its pretty clear that MS and its customers are happy to take a few % performance hit to have a more featured/safe/simpler/etc API.

      Really? Because Vista certainly is not very popular with average Joe. Many people really do not like Vista's new interface at all and have been extremely vocal about it. This from average Joe users to techies. So, I'm not willing to accept this unanimous perception you have that everyone thinks this way.

      Additionally, people have made a large stink about the performance hit experienced with Vista on top of it.

      Only counterstrike tards care about getting 100fps vs 110fps.

      Well, what I saw was more than a mere 10fps difference. You're making assumptions.

      TBH I don't get where all this whinging is coming from. I have a 9600GT, its one of the cheapest cards you can get, it runs everything fine under vista? Why the butthurt?

      Where did I say I couldn't run the stuff under Vista? I said it was slower and I literally, do not see the improvements for any of my uses.

      You implied people like me would see the differences.

      --
      Change is certain; progress is not obligatory.
    5. Re:Please explain then by ozphx · · Score: 1

      A API could also be implemented more efficiently than it's predecessor, granting better performance.

      Unlikely. You are putting a nice object model on top of some rather horrible data streaming and parellel processing tasks.

      The nicer you make the object model, the worse it will match up with the hardware model, and the slower it will perform. This is called a mapping impedance - and is a common bitch with ORM tools. So sorry, no, forcing pixel/vertex/geometry shaders into a nice HLSL interface is never going to make it faster, and people with _commercial_ development experience are going to understand the reasons for making a more rational API with a performance hit. If you want your complex game then you are either going to have it fast and buggy with a shit API, or a little slower and less buggy with a decent one. Same reason people go for managed platforms like Java or .Net.

      As a developer working in industrial visualisation (SCADA HMI tools - flat shaded polygons and not many of them), this directly impacts me. I'm even working in managed code, which is slower again. Nobody gives two tosses, because frankly a month of hardware progress is enough to make up any difference in API performance (and I'm hitting hundreds of FPS anyway).

      I also really cannot be fucked responding to the usual loud butthurt whining about Vistas performance / UI / whatever the fuck. It works fine, can look like XP, on non-crap hardware. Cry some more.

      People don't see the difference. The benchmarks I've seen using real games show a negligable difference. 10% is an exaggeration. Your personal experiences don't mean shit, frankly, and neither do the whinings of people online who are having to measure this crap with the FPSometer in Fraps.

      --
      3laws: No freebies, no backsies, GTFO.
    6. Re:Please explain then by Ash-Fox · · Score: 1

      The nicer you make the object model, the worse it will match up with the hardware model, and the slower it will perform.

      I didn't say nicer, I said more efficient. I actually spent quite few years of my life working on a AmigaOS clone, messing with some very low level and high level API abstraction.

      I also really cannot be fucked responding to the usual loud butthurt whining about Vistas performance / UI / whatever the fuck. It works fine, can look like XP, on non-crap hardware. Cry some more.

      I've not cried. I've merely disagreed with you. I don't see the benefits, period. You can go on and on spewing all your non-sense, but the point is still: I don't see the benefits. My personal experience shows no amazing advantages.

      I am not impressed with what Microsoft have been promoting at all.

      People don't see the difference.

      There are plenty of people "crying" about it - as you put it. Which is it? Are they crying (or is it now 'whining' as you put it, later on in the post) or aren't they?

      Your personal experiences don't mean shit, frankly, and neither do the whinings of people online

      I'm not the one whining.

      Back to your initial post: I still disagree and I am still someone who went out of his way to "code a tech demo comprised of nothing but DX10 specific functions" and reimplemented the code in DX9 too. Guess what? I was "able see the difference" and it wasn't the rubbish you came up with. I had lower performance from DX10 and Vista, period.

      --
      Change is certain; progress is not obligatory.
    7. Re:Please explain then by Anonymous Coward · · Score: 0

      Because Microsoft, amazingly enough, realised there were going to be plenty of cheapo DX10 cards which aren't fully featured, it would run the unsupported elements using DX9 functions or emulation taking a visual or performance hit.

      Ummm ... MS repeatedly said that as opposed to previous DX specs, the only way to get your card certified DX10 was if it supported EVERY feature.

      It's hard to tell the difference at the moment because stuff like Crysis implements DX10 poorly.

      The word you are looking for is "utilizes" not "implements".

  62. What's new? by Anonymous Coward · · Score: 0

    about two years ago i wrote a simple DirectX MMORPG. One of the main problems was that some computers where I had to test it didn't have compatible acceleration cars.

    For those computers I used software rendering vs. hardware rendering. That's alreay present on DirectX, has it been sped up?

  63. GPU Cores on Die by transami · · Score: 1

    In the end I suspect this means they'll be putting GPU support cores directly on the die with CPU cores.

    --
    :T:R:A:N:S:
  64. The minimum CPU spec needed is just 800MHz by miknix · · Score: 2, Interesting

    Sure, just imagine Vista running on a 800MHz computer with software render.

    1. Re:The minimum CPU spec needed is just 800MHz by Anonymous Coward · · Score: 0

      I think you mean "walking"

    2. Re:The minimum CPU spec needed is just 800MHz by miknix · · Score: 1

      Sure, just imagine Vista running on a 800MHz computer with software render.

      I think you mean "walking"

  65. SciFI? by aLEczapKA · · Score: 0

    As stupid as the idea might sound at first, what if in say.. ten or twenty years people come up with new kind of CPUs.. so powerful that specific one-task-optimized hardware won't be necessary anymore ?

    I know that tendency is to move to hardware to have it faster, but what if, quantum computers become real (or any new type of technology), and assuming they gonna be hell of a fast machines... wouldn't be it faster, easier and cheaper to just write software then to do the research, development, design, production etc... all that involves creating hardware....?

    --
    -- All Gods were immortal.
    -- S. Lem
  66. In other words: OS7Warp finally announced by cavehobbit · · Score: 1

    IBM vindicated!

  67. So what effect will this have on Linux? by DJRumpy · · Score: 1

    If DX10 will be handled in the CPU, what effect will this have on Linux in general?

  68. Cycle by DarkOx · · Score: 1

    Ah, the cycle of PC hardware. Starts out with nothing but the bear minimum generalized equipment where the cpu itself is timing the scan rate for the display. Then slowly explodes to the Amigia point where there is a separate chip for everything under the sun. After that everything gets pulled back in again. PCs have been on their balloon out for a while, it started with sound cards, then fancy video cards, then all kinds of expansion buses. Now we are headed back for just one big honkin CPU in charge of everything again.

    --
    Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
  69. How does the performance compare to Mesa? by Ed+Avis · · Score: 3, Interesting

    The news here is not the existence of a software renderer, but one with good performance (such that a high-end CPU is competitive with a low-end GPU for some tasks). I wonder how the trusty Mesa GL renderer compares to Microsoft's latest offering? (They implement different APIs, but Wine provides DirectX 10 on top of OpenGL, so you can get an apples to apples comparison.)

    --
    -- Ed Avis ed@membled.com
  70. OS/2 by bigngamer92 · · Score: 1

    Yeah Microsoft's going to start working with IBM again! Oh wait... This should refresh some memories Seriously Engadget got the Irony before /. whats wrong here?

  71. Re:WOW! Someone buy microsoft a clue. by Ash-Fox · · Score: 0, Redundant

    You do realize that Apple has been about control from the beginning right? The Microsoft master plan is total domination over everything computer related.

    There, fixed it for you.

    --
    Change is certain; progress is not obligatory.
  72. Fallback by OverflowingBitBucket · · Score: 1

    Microsoft promise a lot of things. Let's see if this feature is still there once Windows 7 is out and about.

    Anyway, if you're serious about 3D, you're going to get a dedicated 3D card, or at least a decent integrated one. That doesn't mean what they are doing is a bad idea. It is actually a *splendid* idea. Good on them. The reason I say this is that if a particular feature is lacking on the card you have, you can always fall back to software to use it anyway. Sure, performance won't be great, but you can still use it. This is particularly important if using such a feature is the *only* option- it lets you use software you otherwise could not. It also makes development much easier- you can target a particular feature, knowing that everyone will have it, via the software renderer in the worst case. I certainly hope it is available for hardware renderers to selectively fall back on if they lack a certain feature, but I suspect it'll just be a full software solution.

    (it's a bit late here, please forgive any mistakes in the above)

    1. Re:Fallback by argent · · Score: 1

      On the Mac you just use OpenGL, and it gets handled in software or hardware. I suspect that Microsoft will do the same thing.

      I just wonder why they took so long to get around to it.

  73. warp 10 gives infinity fps according to star trek by Fedarkyn · · Score: 1

    sny star trek fan knows that at warp 10 you can do time travels, enabling the CPU to process all the frames at the same time. any doubts, see star trek 4

  74. Shouldn't it be the other way around? by jlebrech · · Score: 1

    With technologies such as CUDA shouldn't they be putting DirectX fully onto the graphics card?
    Even better would be to have a dedicated "API" card that is loaded with an API and that could then do the processing.

  75. OpenGL please by Anonymous Coward · · Score: 0

    OpenGL please

  76. I hope IBM sues by Anonymous Coward · · Score: 0

    IBM owns OS2 WARP, so I hope they sue M$

  77. Xbox 4 ~ Computer in every house! by soupforare · · Score: 2, Interesting

    I doubt big multicore chips will be cheap enough for the Xbox 3, but win7 probably won't be released in time for it anyway.

    This isn't for Aero on GMAs. This is so you can target both Xbox4 and Win7/GFW without even bothering to think.
    One set-top box, one platform. They've had a hard-on for it for a decade+, it's coming.

    --
    --- Do you believe in the day?
  78. You mean they don't already? by argent · · Score: 1

    I mean, really, both Mac and Linux have software OpenGL so the video card acts as an accelerator... your software is faster with a good video card but you could get all the Quartz effects even on a Powermac 7500, and you can get T&L even on the horrible GMA 950. Sure, it takes up a whole core and makes the system slower than a G4 with a Radeon 9200, but it works.

    I had no idea that Microsoft had completely passed the whole job of 3d graphics off on the video card makers.

    1. Re:You mean they don't already? by Shados · · Score: 1

      Because when they designed it, they (understandably so) figured it was pointless. Who in this day and age wouldn't have a videocard capable of running Aero? Then of course, Intel had to prove them wrong by making the world's crappiest integrated solutions.

      But seriously, we need to push more onto the GPU, since its doing virtually nothing most of the time, not less.

    2. Re:You mean they don't already? by argent · · Score: 1

      Because when they designed it, they (understandably so) figured it was pointless.

      When they designed DirectX? Back in the '90s, when most video cards didn't even do 3d?

      I don't mean "you mean Vista didn't do this already" or "didn't DirectX 10 do this already", I mean "why didn't they do this back when they came up with Direct3d".

      My first Mac running OS X was a Powermac 7600 running OS X 10.1, and it was already doing OpenGL in software back then. On a 132 MHz PowerPC 604e (not even a G3). In OS 9 it was Quickdraw rather than OpenGL, but again it was all the same whether it was software Quickdraw or hardware accelerated Quickdraw.

      This has nothing to do with "doing less in the GPU". This is all about "reducing the distinction between GPU and the CPU ".

    3. Re:You mean they don't already? by Shados · · Score: 1

      Oh! Ok, I misunderstood you. I thought you meant "Why didn't they make it so Vista's GUI could run in software more". Sorry I misunderstood you.

      There is actually a software direct x renderer, and was since pretty much the beggining... it was only ever shipped as part of the development toolkit though, don't know why they never put it as a standard part of Windows. Software OpenGL renderers have been around on Windows since forever too, not just on Mac (I was playing Baldurs Gate 1 with that to get the cool "3d" effects, if you can call them that, at 3 frames per second...).

      So not sure why MS never shipped a consumer version.

    4. Re:You mean they don't already? by argent · · Score: 1

      *boggle*

      So not sure why MS never shipped a consumer version.

      My guess is they wanted to charge extra for it and never figured out a way to do it.

  79. Re:WOW! Someone buy microsoft a clue. by TheRaven64 · · Score: 1

    Why would you need a high end graphics card to run a desktop GUI?

    I don't know if it's part of Aero, but there's a good paper by some MS Research guys about rendering text with pixel shaders (you may be surprised to learn that rendering text is one of the most CPU-intensive things in a modern GUI). In a truetype font, every glyph is stored as a sequence of bezier curves. You can approximate these by drawing two triangles each with the endoints and one of the control points as its vertexes. The pixel shader then removes the errors in this approximation, giving you antialiased text, at any size, with just a small triangle list needing to be stored in RAM for each glyph (rather than a bitmap at each resolution, which is how Quartz and, I think X, does it). This frees up a lot of CPU time and uses less power (since the GPU is optimised for this kind of algorithm).

    There are lots of other things better handled by a GPU than a CPU in a modern desktop, such as compositing the windows together and compositing bitmaps and video within the windows.

    --
    I am TheRaven on Soylent News
  80. Halfway to Parallel Processing by Doc+Ruby · · Score: 1

    The hard part of using parallel hardware to parallel process is that the parallel jobs rarely line up in simple patterns like the simple configuration of the hardware, which makes it hard to specify in software how to parallelize the jobs on the hardware.

    But graphics processing is an exception. Graphics is an obviously parallel job to the rest of the application, with occasional messaging between the two overall jobs between epic bouts of specialized processing within each job. That's why HW graphics acceleration is so successful in exploiting the extra HW, without slowing the main processor to support that basic parallel model.

    I worked with a company almost 20 years ago that built parallel (DSP/FPGA) graphics hardware for a hires (16Mpxl/40bit) digital camera, that grew from the team's previous company that made a dual-processor graphics workstation. One 68040 ran the OS and apps, another 68040 ran the graphics, tied together on a proprietary bus (with another bus behind the graphics CPU to the analog display HW). Splitting the graphics at an API across multiple processors is a very good upfront guess at all the different configurations of demand on the HW for different parallelizable tasks.

    Processing graphics on a separate core is tempting, because the multiple cores have such high bandwidth / low latency on the same chip. But the other cores are not tailored explicitly for graphics, so aren't as internally performant as are GPUs. Putting a GPU core on an x86 chip would be a win. But another win would be moving the higher level graphics software to the separate x86 core, but leaving the product of that processing mid-level instructions and data for the separate GPU. That 3 part structure would be fairly easy and generic to program, since the processing falls into those neat categories. And because the computing load of each stage is roughly proportional to the power of each core.

    What would probably work best would be Windows running DirectX on a separate x86 core from applications and OS, with a bus directly from that core to a separate GPU (PCI-e is probably fine, given enough lanes). Architecting the OS to run on its own core, and applications on their own core(s), probably maps the software to the hardware well generically, then let the mapping overlap when there are fewer cores.

    If Windows doesn't do that, Linux still can (you can do it yourself - it's Open Source :). If Windows does split DirectX to a separate core across an API, then perhaps Linux could still get that package running itself. Maybe Microsoft could make some money licensing DirectX binaries with an open API to game (and pro visualization) companies which run Linux, if Microsoft will accept yet another crack in its Windows monopoly. If not, then Linux will split graphics to a separate core, and run circles (triangles, really :) around Windows. One way or another this architecture is a successful way to compete in the increasingly common (even default) environment of multiple cores.

    --

    --
    make install -not war

  81. OS Who? by Anonymous Coward · · Score: 0

    Hahaha, good ole days...

  82. WARP 10 by quacking+duck · · Score: 1

    Microsoft has introduced a 'fully conformant software rasterizer' called WARP (Windows Advanced Rasterization Platform) 10

    Sheesh, trust Microsoft to forget that Warp 10 is impossible to achieve.

    Or, like their headquarters being located at "One Microsoft Way," is this a subtle message about trying to oversee and control everything, since at warp 10 it's "technically occupying all places in the universe at the same time?"

  83. The argument is dead by Anonymous Coward · · Score: 1, Informative

    Exactly how delusional are you to believe that a piece of software can't be ran on another OS just because some company says that it can't? Have you never heard of Wine? Emulators? Virtual machines?

    The fact that the Wine crew have already reverse engineered part of DX10 and have it running over OpenGL should make it exceptionally obvious that there is no concern that DX10 has that cannot be performed by some other abstraction beneath it. DX10 is not a kernel, it's not it's own operating system. And even if it were, we could sneak another piece of software under it *still* and have it running on other hardware.

    Microsoft's chief argument for DX10 was some bullshit argument about missing features in hardware. Guess what? They just wrote a damned software rasterizer, which does the exact same damned thing that they said wasn't possible for XP. The argument is dead.

    When you wake up from your delusional coma, maybe you'll understand that.

  84. Finally by Anonymous Coward · · Score: 0

    I will be able to play Halo 2!

  85. Stop the bashing already by basicio · · Score: 1

    This is a useful feature, especially for people with laptops. Dedicated graphics cards are large, run very hot, and draw a ton of power.

    Today's integrated chipsets are not nearly where they need to be to handle even simple 3d stuff.

    I have a friend who's considering buying a new laptop because he bought one with integrated graphics a year ago. More recently he started to do some 3d work and has found his integrated graphics card completely unable to cope with even simple stuff. He would love to have something like this right now.

  86. Well... there are some dots to connect... by XDirtypunkX · · Score: 1

    MS has said they're aiming to get Windows 7 down to running on more embedded/smaller platforms. That's what this is really about.

    Intel is going to release even more efficient atom processors and they'll probably also get some of the design benefits from Larrabee. Stick some wider SIMD units on Atom, hey presto, embedded system on a chip and MS is already most of the way there to supporting it.

    Combine it with dense flash storage and the new battery technologies on the way... Suddenly have some very compelling compact devices running Windows 7. And here I was thinking MS was hypersensitive to picking up MS's plotting.
         

  87. Microsoft looking to the future is a good thing by Peter+Cooper · · Score: 1

    Running Crysis at 800 x 600 with the lowest quality settings, an eight-core Core i7 system managed an average frame rate of 7.36fps, compared with 5.17fps from Intel's DirectX 10 integrated graphics.

    This is easy to laugh at (and many Slashdotters have done so above) but this move shows Microsoft is seriously looking at the future, and not in a "we need a new Windows that'll actually make some money" kinda way.

    Running an 8 core machine just to do CPU based graphics rendering is currently impractical and stupid. But in 5 or 10 years? 8->64 core machines will probably the norm and we'll have more CPU power than we know what to do with (we already do for most users in terms of general processing), so there'll be room for this concept.

    The day of many separate pieces of hardware is going to be up sometime in the future, and this is a great step at planning for it. Once you have, say, 100 cores, why not devote a handful to graphics, one to networking, one to IO control.. and effectively have "software as hardware"? Why have a dedicated graphics card *if* the CPU power, technology and bandwidth is there to deal with it in future? At least MS are thinking about something that could be useful 5 years down the line..

    1. Re:Microsoft looking to the future is a good thing by ceoyoyo · · Score: 1

      You do know that pretty much everyone else, including both Linux and OS X already do this right? And have for a long time?

      Microsoft may be "looking toward the future" but it's far more likely they just realized that doing what everyone else did years ago might keep them from getting sued, on a technicality, next time round.

  88. Lower Power Use by Anonymous Coward · · Score: 0

    The biggest benefit would seem to be keeping the GPU turned off during general work with a 3D operating system.

  89. this is for AMD by Anonymous Coward · · Score: 0

    this will lead the way to put better acceleration on the CPU, like the AMD fusion will!!!!

  90. I wonder... by Shaltenn · · Score: 1

    I wonder how long it took them to do this...

    I also wonder how long it would take them to make DX10 work in Windows XP, which I'm sure would have had a much greater impact.

    I dualboot Vista and XP on my desktop and I use Vista solely for gaming with DirectX 10. I would loveeee to use XP for everything and not have both OS's at the same time.

    I guess that's probably too much to ask. Oh wait. Apparently Microsoft has an Alpha for DX10 on XP... ... Maybe I stand corrected?

    --
    If you were offended by anything I said... No, I'm not sorry. Please lighten up.
  91. Microsoft/Vendor separation by minister+of+funk · · Score: 1

    Could Microsoft be trying to extricate itself from the relationships it has with hardware vendors? I read a quote a couple years ago from a Microsoft engineer that said something to the effect of, "Microsoft could make their OS a ton faster on older hardware, but because of relationships with hardware manufacturers like Intel, HP and Dell, they can't."

    (I don't recall the source, so I'm relying on the /. collective to remember or set me straight)

    If Microsoft is free to make the best OS they possibly can, where does that leave the hardware manufacturers? If the demand for new hardware diminishes because the new OS runs as well or better on your existing hardware, what will happen?

    The enthusiast markets will still need to purchase high-end hardware, although we might see an increase in price on the high-end items.

    Now, if Microsoft is left to make the best OS they can, there's no guarantee they will. OEM sales are where they make the most of their OS sales (I'm assuming), but if they can market a bigger, better, faster product, I think they'll see a lot more upgrade sales.

    1. Re:Microsoft/Vendor separation by Shados · · Score: 1

      Could Microsoft be trying to extricate itself from the relationships it has with hardware vendors?

      Thats possible. Remember the beggining of Vista: Videocard and soundcard drivers were by far the biggest cause of headaches (followed closely by OEMs bundle incompatible codecs and drivers even when newer, compatible versions had been released). If you could have simply flicked a switch and have a slower (no big deal for a desktop render...) but well supported implementation, it would have solved the issue and clearly shown that the problem was with the drivers.

      "Oh my god, my computer is crashing constantly, damn Vista!!!" ::knowledgeable person flicks a switch in the control panel:: "Hey...it doesn't crash anymore o.O".

      It would have made for a vastly different scenario, and fingers would have been pointed in a much different direction... Thats probably part of why MS is doing this.

  92. It was a good idea when Apple did it 8 years ago. by argent · · Score: 1

    I have a friend who's considering buying a new laptop because he bought one with integrated graphics a year ago. More recently he started to do some 3d work and has found his integrated graphics card completely unable to cope with even simple stuff. He would love to have something like this right now.

    Nod.

    I've been thinking that I should have gotten a plain Macbook instead of my Macbook Pro, because my Pro runs so hot... and OS X already has this capability: the main difference between running 3d on my G4 Mac Mini with a Radeon 9200 and the Intel Mac mini with an Intel GMA950 is that for some things the Intel is faster and for some things the G4 is faster... they both have pretty much the same OpenGL functionality.

  93. 3D Graphics and actual use of multi-cored CPUs.. by PCMeister · · Score: 1

    With regards to Windows, it's safe to say that many apps out there, with the exception of some higher-end games, don't use modern CPUs to their full potential. Just because Windows acknowledges their presence doesn't mean they're being put to good use. Some will be quick to ask why an office suite (for example) needs multi-core support. My answer is why not!? You forked over the extra cash for the dual-core processor and it would be nice if it was actually put to good use.

    Running what amounts to a virtual 3D-adapter in software is nothing new of course. Whether it will be done properly is another story. Will the performance ever catch up-to the likes of the GeForce 9800 GTX?? Highly unlikely because of it will cause a great deal of strain between the graphics card market and OS vendors like Microsoft. For the VM world, it would be great as Linux users could FINALLY run Windows-based games in a VM without the need to mess with dual-booting. Not to mention Live CDs could finally run Compiz! It would definitely be a paradigm shift as it would help steer more users towards Linux.

    Now THAT would be the year of Linux on the Desktop(tm). Dreaming is still free right?

  94. Then DX10 Should Work on Win95 Too? by Anonymous Coward · · Score: 0

    I find it disturbing that you would so quickly call someone a liar when you are clearly ignorant of the subject you speak of. I find it even more distributing that you were modded up as a result.

  95. Pushing the Envelope by Anonymous Coward · · Score: 0

    Yeah, this is just MS ripping off the cutting edge work done by Lunix.

    Oh, wait...

  96. Facts hurt Microsoft, get over it by SgtChaireBourne · · Score: 2, Insightful

    Choose your "experts" carefully.

    I'll take an expert over a pay-for-say MS "expert" any day. Facts happen to run against MS, get over it. That's why the marketing firms they hire come down so hard on reviewers, evaluators and benchmarkers.

    If you want to get down to the bottom of some of the many, many problems with MS Vista, as well as the OpenGL imitation, then see Peter Gutmann's analysis, A Cost Analysis of Windows Vista Content Protection.

    Running a smear campaign may or may not annoy the author, but it is the facts he is reporting. You can even read Peter's response to the MS attack dogs where he addresses their tactics as well as emphasizes some of the points they chose to skip over.

    MS has a long history of manufacturing abuse of not just critics but also critical data. Money spend on MS products goes into funding unethical, anti-competitive, and, in some cases, illegal activity. Even helping keep the monopoly going, whether intentionally or unintentionally, by not supporting open formats or protocols allows the malfeasance continued funding.

    --
    Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
    1. Re:Facts hurt Microsoft, get over it by Anonymous Coward · · Score: 0

      I'll take an expert over a pay-for-say MS "expert" any day.

      You shouldn't take any obviously biased opinion seriously, and neither biased opinion should trump the other.

      Facts happen to run against MS, get over it.

      That may be, but it doesn't mean that Gutmann wasn't lying or spreading FUD. You're changing the subject. The last paragraph isn't even related to anything.

      So many people on slashdot seem to be victims of black-and-white thinking. Just because you say one thing that doesn't take the worst possible view of Microsoft in one particular instance, doesn't mean that you therefore defend and support everything Microsoft ever did, ever.

  97. We asked for it. by Zaphod-AVA · · Score: 1

    We asked for compatibility to take a back seat to security, and they did it. The end result is Vista. It's best features are either invisible, or downright irritating. It doesn't run a lot of old software. Patches occasionally break things. This is what we said we wanted, and it sucks.

  98. What about older games? by Guspaz · · Score: 1

    If the software rasterizer is faster in newer games, how about older games? I've always been a fan of software rendering engines. Not because they're necessarily better, just because I think they're neat.

    Anyhow, what about Half-Life 1 games (Counter-Strike for example), or Quake 3 engine games, and so on? If they run faster in software mode, then there's some utility here; they'll run fast enough to be playable, and will run better in software than hardware.

  99. the benchmarks i want to see by steak · · Score: 1

    equal computers one with a cuda enable version of this warp 10 and a one of those nvidia tesla c1060 cards versus one with directx 10 and a nvidia gtx 280 (the graphics card which the c1060 is supposedly based on). of course this would effectively be a comparison of a $1200 video card vs. $500-$600 video card, but it would be interesting to see.

  100. DirectX is the least of your worries by Anonymous Coward · · Score: 0

    After the Spore debacle (and the soon GTA debacle) you're going to need more CPU for the DRM than the graphics...

  101. Killing stream processing. by w0mprat · · Score: 1

    I wonder if MS is taking a stab at stream processing, which is a threat to Intel. Since obviously this is step towards enabling rendering on Larrabe, which is essentially a great big multi-core x86 CPU in a GPUs clothing. Little code change would be needed to get a full DX10 render path running on a larrabe chip. Since vendors are not going to stop putting under powered GPUs in laptops and mainstream desktops, it makes sense to meet all needs in the middle. AMD want's to put a GPU and CPU on the same die, fine, but Intel + Microsoft are obviously gearing up for one general purpose many-core CPU to do all things.

    --
    After logging in slashdot still does not take you back to the page you were on. It's been that way for 20 years.
  102. Tired of meaningless comparisons by RazvanHrestic · · Score: 1

    I'm really getting bored of meaningless statistics such as "my software renderer is better then your software renderer because it can run Far Cry at a FPS of 6 rather than 5", which forget the essentials: when it comes to playability 6 IS NO BETTER THAN 5! In fact, 7 is not better than 5, nor is 10, nor is 15. Starting from 20 FPS this might be somewhat meaningful! Please refrain from cluttering teh Interweb with more meaningless statistics Have a nice day, The Spaghetti Monster

  103. Good ol times.... by Anonymous Coward · · Score: 0

    I used to play Weapons Factory for Quake 2 on software rendering and "56K" modem (if you can call it that). I could snipe with 200-300 ping. Now get off my lawn LPBs!!!

  104. There is no '8-core' Core i7 by Anonymous Coward · · Score: 0

    there is only a 4 core i7...with hyperthreading enabled, which should give 8 threads.

  105. Warp 10. We are borg! by Anonymous Coward · · Score: 0

    Am I the only one to see a relation to Star Trek? The picture tag features Bill Gates as a borg. And borg had Transwarp, which was like Warp ten.

    C'mon people this is /. so connect the dots! Its gotta be a joke.

  106. As an engineer... by gillbates · · Score: 2, Interesting

    the brain-dead architecture of the x86 PC strikes me as funny.

    Here, you've got 2, 4, what - now 8 cores which can't compete with a decent FPGA?! The problem isn't the CPU speed. The problem is that CPUs make very poor DSPs. A TI DSP can encode/decode HD video in realtime using only a few percent of the power required by the desktop CPU. A large part of that GPU's performance comes from the fact that it has hardware optimized for video, which, of course, Intel has steadfastly refused to add to their processors. Instead, they push multimedia instructions which, as hard as they try, are still hamstrung by the memory architecture, and hence, non-competitive compared to a GPU.

    What we really need is for PC architecture to include a standard FPGA which can be reprogrammed on the fly by the OS. You need a GPU? Simply program the FPGA for 3D tasks (you need not emulate the entire GPU - just the parts you need at the moment for your application). You want to do audio processing? Filter implementation in the FPGA is as simple as loading the correct software. Instead of writing the algorithm in software, and it being implemented by software, you configure the hardware to do the computations you need directly. That way, you get the flexibility of software with the speed of dedicated hardware.

    But, alas, market forces trump all others. I remember seeing $20 motherboards recently!? When even a Spartan FPGA costs $10 in quantity, I'm not going to hold my breath for a standard FPGA. But it sure would be nice.

    --
    The society for a thought-free internet welcomes you.
  107. I remember.. by skelly33 · · Score: 1

    Back in the olden days applications used to perform some sort of benchmark on the hardware they were running on and simply advise you, self-degrade, or just plain refuse to run if your machine can't handle it. I find the software-only rendering option to be an interesting one, but as someone else pointed out, am unclear on how this is a major advancement over OpenGL. Then again, I won't touch Vista with a 10' pole, so I may never know.

  108. Low compared to what? by Molochi · · Score: 1

    The quad core2 cpus are about twice that. The new i7 also peaks at ~130W. Intel is back down to where they were with pentium3; around 35W per core.

    The reality is that power usage/processing performance (for every manufacturer of processors) generally improves every year and it is software reliance on cpu performance that increases your electricity bill.

    --
    "The Adobe Updater must update itself before it can check for updates. Would you like to update the Adobe Updater now?"
    1. Re:Low compared to what? by msormune · · Score: 1

      Yeah, that's true but i7 is considered pretty much high end now.

      In my perspective it seems Microsoft wants Windows 7 to run on a simple motherboard without dedicated gfx hardware of any kind. Maybe just a mapped frame buffer and video out interfaces. Which makes sense to me, because everyone knows integrated gfx chips aren't actually very good for other than running Aero... So why include them in a strictly office PC in the first place, if the CPU can cope without it.

  109. MSFT stole Stacker to do it... by mccabem · · Score: 1

    They ended up paying them after going to court and such.

    It's almost hard to believe MSFT is still in the OS business given the history of their OS products and how bad they are at it.

    Now they're running 3D graphics on the CPU. How 1987 of them.

    -Matt

  110. Anonymous by Anonymous Coward · · Score: 0

    I've tried to install Apple's Mac OS X on a PC. And even drivers for my GPU was not found, I can see many of Mac OS desktop effects. Minimizing windows, dialog box effects, expo - all runs smoothly without GPU acceleration. May be some thing like this is Microsoft's point.

  111. MESA GL by Rysc · · Score: 1

    Headline: Microsoft invents what the open source world has had for years.

    Great innovation, Microsoft!

    --
    I want my Cowboyneal
  112. i7 has four cores and eight threads by Anonymous Coward · · Score: 0

    The CPU in question is a four core CPU that can do eight threads in hardware. To have eight cores you'd need two CPUs.

  113. Not about the high end by ravyne · · Score: 1

    This thing isn't meant to run the latest whiz-bang games like Crysis -- They only demoed it with Crysis because its a high-profile benchmark that probably exercises the capabilities of Direct3D 10 pretty fully.

    This is aimed at accelerating Aero and casual games like your girlfriend or mom plays. They want to push Direct3D 10 in this casual games space, which currently tends to target older versions of Direct3D, which Intel's integrated graphics has an easier time coping with. Its basically a free Direct3D 10 graphics card for all PCs, making the Direct3D 10 feature set the new baseline, even if performance is a fraction of dedicated cards, because it will still perform well enough for many casual games -- remember that casual and web games are the fastest growing sector of the games industry, and that women over 35 is already a signifigant and fast-growing demographic.

  114. That's why we're patenting it by EmbeddedJanitor · · Score: 1
    Yours,

    MS Intellectual Property

    --
    Engineering is the art of compromise.
  115. How can I be nice when addressing a bunch of cocks by shadders · · Score: 1

    First, look at the MSDN article: http://msdn.microsoft.com/en-us/library/dd285359.aspx unless it's against your religion to do so. Microsoft is touting Warp as a software fallback where the PC hardware is not up to the job. Well done to the ~5 posters who have seen this, your score a +10 on the scale of comprehending the MSDN article. -10 to Linix/Mac fanboys for seeing the word Microsoft, frothing at the mouth. A software fallback makes perfect sense on several fronts: 1. Older software that uses GDI/GDI+ gets a performance increase on Windows 7. GDI/GDI+ performance got walloped with Vista. Try running some VBA in Excel that builds graphs for example. Stupid new diver model and crap implementation from device manufactures. 2. Direct 3D games. The article mentions device CAPS, that's capabilities if I recall. Game coders have to examine what the device (graphics card) is capable of doing before trying to do it. EG: If its ATI and it's model X do routine Y, otherwise do Z. The code should just run in software emulation. It will be easier for developers to at least run the code without it crashing and identify where the fallback to software is. They are then free to develop a better workaround without having crash dumps to understand. EG quicker game development. 3. And in my opinion, most important. The GUI, Windows Presentation Foundation (WPF) was introduced with Vista and on XP. While I think it's next to impossible to truly describe the potential impact of this, I'll have a go. a. It's vector based. Resolution and device independent graphics for example. Coordinates are double precision, if someone makes a massive monitor or a monitor with 600 pixels per inch capability, WPF will still cope without the display looking crap. b. Floating point precision colour. Put this in perspective, most colour values are byte based. One byte for red, another for blue, another for green and one more for alpha or opacity (transparency). A byte = 256 values. Ignoring the alpha channel, that's 256x256x256 colours, or 16,777,216 colours. A single (32bit) can be positive or negative in value, so lets be conservative and only use positive values; that's 3.402823e38 or 340,282,300,000,000,000,000,000,000,000,000,000,000. Now multiply that by 3 for red, green and blue and you get: 120,846,900,000,000,000,000,000,000,000,000,000,000,000,000. That should be enough colours to be getting on with. I imagine that display technology will take a while to catch up with a figure like that. c. 3D. Developers can use 3D in their applications without an intimate knowledge of how the Direct 3D API works. d. Recently, .net 3.5 SP1 introduced the ability for programmers to use shaders as well. It's early days yet; so useful implementations are bit thin on the ground. But you can introduce motion blur when a user is scrolling through a list of photos for example. All of this stuff does not come cheep, so MS are offloading as much as they can onto the GPU. But if the GPU fails for some reason, they don't want GUI developers trying to understand why and code round the problem. Warp enables them to at least have the app run, regardless of hardware. They are then free to either fix the problem or up the hardware limits. This technology has bugger all to do with playing Crysis, it's used as an example for gods sake.

  116. Re:WOW! Someone buy microsoft a clue. by Anonymous Coward · · Score: 0

    What? That you can get better performance in software on an 8-core system, than with integrated graphics? So how many real systems will this help run Aero? I suspect most 8-core systems being sold will have sufficient graphics capabilities, and the ones with the shitty integrated graphics will probably still be better off with integrated graphics than software rendering.

    I don't think this is for running Aero on machines with under-performing integrated graphics, and that isn't what Microsoft say it is for. I don't quite understand what the purpose of this is, but according to the MSDN page on it

    We don't see WARP10 as a replacement for graphics hardware, particularly as reasonably performing low end Direct3D 10 discrete hardware is now available for under $25. The goal of WARP10 was to allow applications to target Direct3D 10 level hardware without having significantly different code paths or testing requirements when running on hardware or when running in software.