Slashdot Mirror


AMD Releases 900+ Pages Of GPU Specs

An anonymous reader writes "Ending off the X Developer Summit this year, Matthew Tippett handed off ATI's GPU specifications to David Airlie on a CD. However, the specifications are also now available on the X.org site. Right now there is the RV630 Register Reference Guide and M56 Register Reference Guide. Expect more documentation (and 3D specifications) to arrive shortly. The new open-source R500/600 driver will be released early next week."

325 comments

  1. Its by Anonymous Coward · · Score: 5, Funny

    over NINE HUNDRED!!

    1. Re:Its by Anonymous Coward · · Score: 0, Redundant

      I've seen real register reference PDFs for ATI chipsets and what was released was generic 2D support register references. Don't expect OSS to build *real* 3D OpenGL implementations based on what was provided. I don't want to sound too negative though, because this is a good start for AMD. But for slashdotter's praising this action as the end all be all to disclosed technical details, be aware that there are many details missing (like anything useful for fixed or programmable pipeline 3D setup and usage)

    2. Re:Its by ToasterofDOOM · · Score: 0, Redundant

      meh. It needs moar mudkipz

      --
      I am Spartacus
    3. Re:Its by Knuckles · · Score: 2, Informative

      I've seen real register reference PDFs for ATI chipsets and what was released was generic 2D support register references. Don't expect OSS to build *real* 3D OpenGL implementations based on what was provided.

      Um, the summary says "expect more documentation (and 3D specifications) to arrive shortly".

      --
      "When I first heard Daydream Nation it quite frankly scared the living shit out of me." -- Matthew Stearns
    4. Re:Its by Anonymous Coward · · Score: 0

      I don't get it... This is Sparta?

    5. Re:Its by Anonymous Coward · · Score: 0

      While I welcome AMD's commitment, why the hell does it need a 900 page spec (minus all the whitespace that is due to the tables) for just 2D? Does that mean they cannot do things the right, the simple, way? Don't worry, NVidia does not seem to do it right either. They have almost one file for each card revision (e.g. nv40.c, nv41.c, etc.) that make up the 5+ MB binary Linux blob, known as nvidia.o in the package, up. Hilarious.

  2. I'll replace my nVidia when I see a good review. by khasim · · Score: 4, Insightful

    The only way to get nVidia to release their specs is to show them that there is a real market.

    I'll do my part and replace my AGP nVidia card with an ATI one as soon as there is a good review of an available card with this driver on Ubuntu.

  3. Sound interesting... by Anonymous Coward · · Score: 5, Funny

    no, wait, the other thing - tedious.

  4. Great by qbwiz · · Score: 4, Insightful

    They've actually done it. It's time to buy an ATI card.

    --
    Ewige Blumenkraft.
    1. Re:Great by StormReaver · · Score: 5, Insightful

      "They've actually done it. It's time to buy an ATI card."

      I've been recommending nVidia cards to everyone who asks, simply because their Linux support has been leagues ahead of ATI (now AMD, for those who haven't been paying attention). If the specs are credible enough to create a quality Free driver, then I'll switch to AMD in a heart beat.

    2. Re:Great by Workaphobia · · Score: 3, Interesting

      Is it really though? That's not rhetorical. Without RTFAing, I want the slashdot opinion - is this or is this not the proof that ATI is the solution for linux graphics? I was almost certain that my next card would be an nvidia, but this may change that.

      --
      Evidently, the key to understanding recursion is to begin by understanding recursion. The rest is easy.
    3. Re:Great by icepick72 · · Score: 1

      You're not jumping on a bandwagon, you're creating one. Kneejerk reaction! Care to expand on anything like "it", what "it" means to them and us, etc?

    4. Re:Great by amiga3D · · Score: 2, Interesting

      I've got a new in the box nVidia card I was going to put in my new tower I'm building. It's off to e-bay for it and back to the store for an Ati card. amiga3D

    5. Re:Great by 777a · · Score: 1, Interesting
      Quoting myself from a few weeks ago:

      For the last several years I've alternated between nVidia and ATI (nVidia 4200Ti, ATI 9800XT, nVidia 6500, ATI X1900XTX).

      My last card (the ATI X1900XTX) cost ~$600, since then I've started paying a lot more attention to Linux, and unless ATI do some decent drivers it'll be the last ATI card I buy.

      I retract my previous statement, it looks incredibly likely my next card will be an ATI.

      Looking like I'm going to be becoming an ATI fanboi.

    6. Re:Great by Xtravar · · Score: 5, Insightful

      I've been recommending nVidia cards to everyone who asks, simply because their Linux support has been leagues ahead of ATI (now AMD, for those who haven't been paying attention). True, true. My girlfriend got a laptop a couple weeks ago and one of the deciding factors was nVidia vs ATI. I haven't even considered an ATI card for myself and I recommend that others get nVidia.

      It's about fucking time that companies realize the trickle-down effect of abusing nerds. Who do the ignorant masses go to when they need advice? Their nerdy friend...

      ATI lost market share for almost the exact reason that IE did (albeit to less extent).
      --
      Buckle your ROFL belt, we're in for some LOLs.
    7. Re:Great by EzInKy · · Score: 3, Interesting


      is this or is this not the proof that ATI is the solution for linux graphics?


      For those of us who absolutely refuse to use closed sourced drivers their older cards have already proven to be the best solution for desktops.

      --
      Time is what keeps everything from happening all at once.
    8. Re:Great by poopdeville · · Score: 4, Insightful

      Is it really though? That's not rhetorical. Without RTFAing, I want the slashdot opinion - is this or is this not the proof that ATI is the solution for linux graphics? I was almost certain that my next card would be an nvidia, but this may change that.

      It will be, in a few weeks. Moreso in a few months as the drivers improve. Performance tuning is one of the open source methodology's strengths.

      --
      After all, I am strangely colored.
    9. Re:Great by TheRaven64 · · Score: 2, Informative

      I've had a quick look at the PDFs. It seems like they have a lot of registers for setting up the display modes, including dual display. They don't, however, seem to specify the registers for doing things like loading vertex arrays. The specifications to date don't seem to include enough information to produce a good 3D driver.

      --
      I am TheRaven on Soylent News
    10. Re:Great by antdude · · Score: 1

      Agreed. Let's wait for the actual results. This could be months to years. Let's see what the outcome is. I'd like to switch to ATI/AMD too because NVIDIA drivers for Windows suck. The only reason why I still use them because their drivers rock in Linux.

      --
      Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
    11. Re:Great by Anonymous Coward · · Score: 0

      The documents do not cover 3D. Those documents will come later.

    12. Re:Great by Anonymous Coward · · Score: 0

      Yeah like many I'm sure, I've stuck with nvidia because I know their linux drivers will work. ATI has never been a real contender. This changes it. Now ATI is available for comparison. I don't expect with the release of some specs/docs today that awesome drivers will show up tomorrow, but then again I'm not shopping for a new card yet. If this situation develops like many of us hope it will, then I'll even buy the ATI if it is slightly inferior or slightly more expensive on the weight of this action. So I'm still in a wait and see mode, but this is probably gonna score some big points.

    13. Re:Great by Anonymous Coward · · Score: 0

      lucky me, i happen to have one of both(ati x1550, geForce 6200 OC)

    14. Re:Great by putzin · · Score: 5, Informative

      Yeah, but the article clearly states 3D specs sometime soon (next week was it?). So why complain?

      --
      Bah
    15. Re:Great by RAID10 · · Score: 1

      these specs are only to produce the 2d driver for starters. 3d specs will follow.

    16. Re:Great by pxc · · Score: 2, Interesting

      The specifications to date don't seem to include enough information to produce a good 3D driver. Dead on! Those are only the 2D specs. The 3D specs will allegedly be released "soon" (and judging by AMD's expedience in releasing the 2D specs, I think I actually believe them).
    17. Re:Great by Dolda2000 · · Score: 3, Insightful

      If the specs are credible enough to create a quality Free driver, then I'll switch to AMD in a heart beat.
      Yes, me too. I was actually planning to do an upgrade of my box in a little while, for the first time in many years now, and I was sure that I was going to get an nVidia card, but this might turn that decision around completely.

      However, I have to wonder -- I really have no idea about ATI GPU parts, but the impression I got is that they are releasing the specs for the new top-of-the-line units, and since I don't even play games, I'm not interested in such things. What I'm interested in is having dual-display, TV output, 2D acceleration and XV working on the budget cards (and without making VGA BIOS calls, thank you very much), but I have yet to hear whether these released specifications will cover enough to create a truly free, fully featured driver for the budget model GPUs.

      Also, apart from budget models, how will these specifications apply to older cards? I still have a Radeon 7500 lying in a drawer doing nothing just because I never got the TV output working on it in Linux. As a side story on that one, I even engaged in communications with ATI to try and get some specifications on that card in order to enhance the X driver with TV output support, but even when I managed to get my hands on documentation, it conspicuously excluded any information on the registers controlling the TV output encoder (even though I had explicitly requested that information...). That's when I resigned myself and bought a GeForce 5200 instead.

    18. Re:Great by xSauronx · · Score: 1

      i dont know about tv outs on the 7500, but the T40 thinkpad has a mobile 7500 in it and the performance is quite good enough to happily run Beryl with its fancy cube and whatnot. i was surprised. its not useless, by any stretch, even if the tv out doesnt work.

      --
      By and large, language is a tool for concealing the truth. -- George Carlin
    19. Re:Great by Hooya · · Score: 1

      In the same vein, DON'T ever buy a Matrox card. From what I remember of the common wisdom from back in the day (before nVidia), matrox was all that and more because they were best supported in linux and were high quality cards.

      Recently I figured I'd use 'em for high quality 2D (I don't play games so don't really need 3D/OpenGL). Went with a high end Matrox since I was planning on 4+ heads. Big Big mistake. They seem to have the opinion that they have enough paying clients within the Oil/Military industry and that Linux users, in their opinion being synonymous with Freeloaders, don't generate any revenue and therefore can go to hell. Never mind that Linux user or otherwise, I still paid for their card. Fidgeted with messy non-supported drivers for a while and switched back to nVidia with 2 heads.

      Now that ATI/AMD looks like they'll come through big for my platform of choice, I'm already a fan. Now if I could drive 4+ 1600x1400 DVI displays, I'd be in heaven.

      Going forward, ATI/AMD it is.

    20. Re:Great by HeavyD8086 · · Score: 1

      http://www.phoronix.com/scan.php?page=news_item&px=NjA1Mw We're supposed to expect more 3D specs shortly.

    21. Re:Great by innocent_white_lamb · · Score: 1

      I believe the words that you're searching for there are "Intel video chipset".

      --
      If you're a zombie and you know it, bite your friend!
    22. Re:Great by caluml · · Score: 1

      Very true.
      My current GFX card is on its way out - it's scrambling up characters in text mode, and doesn't provide enough output to power one of my monitors. I was certain I was going to get another NVidia - because, aside from binary drivers, it's worked the best. I can't even get Beryl running on my ATI-something at work.
      However, now I'm hoping that it'll last long enough for me to see what comes of this. I don't want super top end performance - I'll take average performance with truly open drivers to super performance with closed drivers anytime.

    23. Re:Great by Kjella · · Score: 1

      Something tells me that by the time you're done digesting the first 900 pages (well, 450 if you're only doing one chip) the 3D specs will be released. When it comes to specs, release early, release often is a good policy. I easily expect those to be 1000+ page mammoths since 3D is a lot more complex than 2D these days. It's not like there's much to gain by waiting until you cna dump it all like a ton of bricks.

      --
      Live today, because you never know what tomorrow brings
    24. Re:Great by rtyall · · Score: 2, Funny

      Yes, thank the heavens that you have two low-end/cheap video cards from different manufacturers.

    25. Re:Great by RobertLTux · · Score: 1

      yes are they going to release the specs for the All-In-Wonder chips (i have a Radeon 7500 AIW and it works more or less but

      1 3-D sucks on it
      2 i can't get close captioning to work on it
      3 No capture
      4 i don't even bother trying to use the tv out-put

      --
      Any person using FTFY or editing my postings agrees to a US$50.00 charge
    26. Re:Great by 3.14159265 · · Score: 1

      It does indeed look promising.
      The specs mention audio, and I can only dream that it will be possible to output video+audio over HDMI: finally, a solution for 6.1/7.1 digital!

      Very promising.

    27. Re:Great by Anonymous Coward · · Score: 0

      Yeah, yeah, nice story. You have a person who listens to a nerd, and that person is a female and is your girlfriend?

    28. Re:Great by Anonymous Coward · · Score: 0

      Good news indeed, but I'll wait until the drivers based on those specs are both released and tested, first, before buying a new video card.

    29. Re:Great by petermgreen · · Score: 1

      BTW at least one firm make a nvidia based graphics card in pci express X1, http://www.tweaktown.com/reviews/1045/index.html . Maybe you could use a couple of those to drive extra displays.

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

      It will be, in a few weeks. Moreso in a few months as the drivers improve. Performance tuning is one of the open source methodology's strengths.

      Weeks? Months? I think you are overly optimistic here. Even if AMD releases the full spec with 3d support and everything, it will likely take at least a year before we've got a fully working, production quality Open Source driver for these cards.

      Just look at how long it's taking the nouveau project to come up with something working. Granted, they're in a much worse position, having to figure out the cards by themselves with no official documentation whatsoever, but still, it's a lot of work to go from 900 pages worth specs to a working, tested and optimized driver that implements all that.

      --
      If it's so secret, then how come I've never heard of it?
    31. Re:Great by Anonymous Coward · · Score: 0

      The older Radeons already have well-working Free drivers! They're up to r4xx series now. r5xx and up (X1300 and up) don't work yet, that's why specs are very much neded.

    32. Re:Great by Anonymous Coward · · Score: 0

      ATI granted NDAs to several Linux developers to work on the drivers before the specifications became available to the general public. Now that the specifications will soon be available, the drivers will be released.

  5. Sweet! by timeOday · · Score: 5, Informative

    Closed-source drivers can be OK, except they tend to discontinue support after a while. Eventually the binary driver won't load into a current kernel and you are high and dry. With open-source drivers, the prospects for long-term support are better.

    1. Re:Sweet! by walt-sjc · · Score: 2, Insightful

      Considering the issues I've had with closed source drivers over the years, I just can't EVER agree with them being OK. Closed source drivers have had all sorts of issues with not only kernel changes, but distro versions, architecture (64bit), xfree86 vs x.org, have issues with redistribution rights, etc.

      Furthermore, BSD / vs linux vs Solaris. No, IMHO closed source drivers just suck in all cases. We need the specs. Specs for all hardware would allow us to have working scanners, webcams, wifi adapters, etc. on Linux / BSD no matter which brand you buy, or which architecture you use.

      Let's let Nvidia know that we will all be moving to AMD/ATI for all future purchases over this issue, unless they too release specs.

    2. Re:Sweet! by walt-sjc · · Score: 1

      Forgot to mention that open source drivers are almost always easier to install due to the fact that they are normally included / maintained by the distro.

    3. Re:Sweet! by tiocsti · · Score: 1

      Slavery can be a good thing too, cept they tend to beat you.

    4. Re:Sweet! by Just+Some+Guy · · Score: 1

      Closed-source drivers can be OK, except they tend to discontinue support after a while.

      Exactly. I have an aging GeForce 4 that's slow by current standards but still quite enough to mess around with Beryl and play Unreal Tournament. And yet, Nvidia has deprecated its drivers. When new kernels can no longer load the old module, I'll have to decide between resigning myself to that kernel version forever on this system or forking out a comparatively huge amount of cash. (Since this is an AGP system, I'll also have to replace the motherboard. And CPU. And memory. And probably throw in an IDE card so I can use the 320GB drive I bought last year.)

      Again, this is a working system. I know it's not new and shiny, but it runs everything I want to run and there's no other reason I'd need to replace it right now. Suddenly that ATI card is looking far cheaper if it means that I won't have to literally upgrade my entire system as soon as they decide to stop providing closed binary drivers for it.

      Oh, to head off the inevitable criticism from k1dd1es: this is what happens when you get old. Sure, I can afford a nice new system and I'm not a cheapskate about it. Still, I'm stubborn enough to not want to if I don't have to and it's the principle of the thing, and get off my lawn.

      --
      Dewey, what part of this looks like authorities should be involved?
    5. Re:Sweet! by Just+Some+Guy · · Score: 4, Insightful

      OMG, you mean you can't run the same computer equipment for 20 years and expect it to before useful work in a modern fashion?

      Your reading comprehension is worse than your grammar. By 20, you mean 5. It still "[something] useful work in a modern fashion", except that it's been artificially crippled by recent driver updates.

      Probably just because they want money. Let's burn them.

      Child, meet Market. Customers don't want to spend more money than they have to. Paradoxically, this often means they'll spend even more money with the companies that don't put the squeeze on them.

      --
      Dewey, what part of this looks like authorities should be involved?
    6. Re:Sweet! by Sark666 · · Score: 2, Interesting

      Well, one time where people were left high and dry, was when voodoo went under. Lots of people had voodoo3s and its brethren. In this case, 3dfx had already released specs for linux (and I believe provided an open source driver at one time? can't recall). Yet with the specs released, the open source driver never achieved the same performance of what the card was capable of (comparing it to windows drivers' performance) and problems with some functionality.

      When investigating for info on updated drivers the general feeling was no one is working on them because none of the devs bother with using 3dfx anymore and most users have moved on so there is lack of interest in further support. And this was only shortly after 3dfx folded. I thought this was a perfect situation for the strength of open sourcee. I actually still have that same v3 and when putting together a myth box last year, I thought it might be good enough for basic video. But in xorg, hardware overlay has been broken for quite sometime, so I couldn't control the brightness/contrast etc of videos. And opengl was pretty shaky.

      I don't think the same thing would happen here, as ati is still doing fine and lots of people have their cards, now. But when a card becomes old and outdated, it will interesting to so how long it's really supported.

    7. Re:Sweet! by Ash+Vince · · Score: 1

      I also will be moving to ATI cards just as soon as I can type "emerge ati-driver" and get a decent open source driver to compile with same ease that all other software does on my Gentoo box (Ok, I might have to change VIDEO_CARDS="ATI" in make.conf as well)

      But I have never had any problems with NVidia closed source binary driver. The ATI one was trash which was I started buying NVidia cards in the first place but if myself as gentoo tweaker has never had any issues they cant be that bad under Linux. Obvoiusly they are trash for everyone on non x86 hardware or running BSD or whatever but not all of them are that bad.

      Maybe the reason that ATI have taken this step is because of the problems they had with their drivers in the past.

      --
      I dont read /. to RTFA, I read /. to offend people in ignorance.
    8. Re:Sweet! by veganboyjosh · · Score: 2

      ...unless you own some slaves, that is.

    9. Re:Sweet! by neomunk · · Score: 0, Troll

      Really? Because Vista works magically with any old hardware you throw at it, right? Or maybe because linux is known for using closed source stuff as a crutch?

      Put more effort into your trolling, I demand at least a modicum of entertainment from my Anon Coward trolls.

    10. Re:Sweet! by timeOday · · Score: 1
      Oddly enough, I also have a no-longer-supported Geforce 4 4200 card.

      My real complaint with upgrading isn't so much the money, but the hassle of spinning the wheel of fortune on a new piece of hardware, a new driver, hardware and software incompatibilies... I just don't need the hassle. Anybody with significant computer experience knows you can never assume stuff will work like it's supposed to.

    11. Re:Sweet! by Maxo-Texas · · Score: 1

      When win98se came out my "hot" sound card stopped working(crashed the system). After a lot of work (by microsoft technical support), it was determined that the driver was the problem. And .. the company was not going to update it (my $145 sound card was now junk). How much of a change would it have to be to go from Win98 to Win98SE. I imagine it was something trivial. But since it was closed source, that was it.

      --
      She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
    12. Re:Sweet! by nunyabid · · Score: 0

      Why would you need to keep the kernel current if you're using old hardware? Isn't it normal to run old OSes on old machines?

    13. Re:Sweet! by Curtman · · Score: 1

      I also will be moving to ATI cards just as soon as I can type "emerge ati-driver"

      More likely you'll want to "emerge xf86-video-ati", that's the free one, and the one we need to stop sucking.
    14. Re:Sweet! by Just+Some+Guy · · Score: 1

      Why would you need to keep the kernel current if you're using old hardware?

      If nothing else, for security fixes. That alone is a huge motivation.

      Isn't it normal to run old OSes on old machines?

      Only on Windows. Other OSes tend to become more efficient with time. OS X has a particularly strong reputation for this.

      --
      Dewey, what part of this looks like authorities should be involved?
    15. Re:Sweet! by vranash · · Score: 1

      I can't talk about 20 year old hardware, since my 386 isn't currently running (And I need to swap in a 16 bit ISA EIDE controller card to replace the MFM card and hard disk I lost the bad sector remapping for.) But I've got 2 10 plus year old servers chugging away right now with 128 megs (4 cpu server) and 256 megs (2 cpu workstation/server), and both of them are still running serious applications under linux as we speak. I have another that if I had the spare power cable to attach it, would be chugging away with FreeBSD within 5 minutes of powering it on. Nevermind that my parents are currently using a Single Proc Celeron that I went through 2 mobos with that is currently running Fedora Core 7, and while it's a little slow starting up, it handles their web browsing and email needs just peachy, and can play Q3 at between 20 and 40 fps (depending on if I bump it from 10x7 to 16x12, and if I swapped in a newer video card it could probably peg at 90 in most places, and in fact HAS when I've had a 9800 in there, so imagine a more modern card, like an X1650, or GF7600, or HD2600Pro AGP.

      Disrespecting old hardware is foolish. A lot of older hardware, esp server grade stuff is fully capable of running until many of us are old or dead, depending on working enviroment (keeping it below 85F if possible, maintenance (Not letting the whole fsckin case fill with dust/hair/crud), and new hard disks (the singular largest point of failure in systems, esp old ones. You'll have some disks last for 20 years, and some barely last one. I've had both, I've got a half dozen 9 gig scsi hard drives that have been chugging for at least the last 2 years straight (and were purchased a good 4 years ago) I've had a few 9/18 gigs fail suddenly. I've had a few smaller and larger IDE drives fail as well. But I've also had more than a few last. Personally I think if you really want to keep your older hardware running for a long time, transitioning to current gen Compact Flash, and IDE to CF adapters is the way to go, since unlike hard drives they're unlikely to fail COMPLETELY at once, and hopefully usually during write, rather than read.) The only other major failure point is the potential for caps to go bad, and as long as you can catch it before it damages something else, you can always take the time and track them down and swap 'em out. Nothing like a twenty or thirty year old system you can say 'I not only have had it running all these years, but I've made sure it'll stay running for my kids to use.

      Just my view on it, YMMV, but I'm sure there are people here with their C64's and VAX's and other oldschool hardware that still have it running, and maybe even have repaired it a few times to keep it doing so. Newer hardware I've had a lot more issues with dying, be it heat, or cold solder joints (no lead, woohoo! ;.;), or just cheap components.

    16. Re:Sweet! by cortana · · Score: 1

      And if you buy an all-Intel system then it will.

      Hopefully these specs will mean the same thing for AMD/ATI systems some day.

    17. Re:Sweet! by cortana · · Score: 1

      Why would you need to keep the kernel current if you're using old hardware? If nothing else, for security fixes. That alone is a huge motivation. Not if you use a distribution that backports security fixes to the kernel present in their stable releases?
    18. Re:Sweet! by dvice_null · · Score: 1

      > But when a card becomes old and outdated, it will interesting to so how long it's really supported.

      That is not really a problem. Once the drivers are implemented and major bugs are fixed, the maintenance work is very small task. So maintainer might not be needed at all.

    19. Re:Sweet! by walt-sjc · · Score: 1

      Oh, you mean like RedHat Enterprise? So yeah, you go down that road, and happily install your binary kernel modules for EMC PowerPath (multi-pathing on a SAN.) Redhat comes out with a new kernel with the same version but with the backported fixes. Oh damn, that binary module no longer works, so I have to revert to the insecure version of the kernel if I want EMC branded and "psedo-supported" multi-pathing. EMC? What do you mean that you have no ETA for a new rev of powerpath that supports the new kernel rev? Note that this is a recurring problem. EMC NEVER supports a current point release of the OS, although sometimes you get lucky and the old module works with the new rev, although that seems to be rare.

      Binary modules are a real problem - even when you backport security fixes.

    20. Re:Sweet! by cortana · · Score: 1

      Oh, you mean like RedHat Enterprise? So yeah, you go down that road, and happily install your binary kernel modules for EMC PowerPath (multi-pathing on a SAN.) Redhat comes out with a new kernel with the same version but with the backported fixes. I actually meant like Debian, where backported security fixes usually do not change te kernel's ABI. If changing the ABI is unavoidable, then the package containing the kernel is renamed (e.g., linux-image-2.6.18-4-k7 -> linux-image-2.6.18-5-k7).

      Oh damn, that binary module no longer works, so I have to revert to the insecure version of the kernel if I want EMC branded and "psedo-supported" multi-pathing. EMC? What do you mean that you have no ETA for a new rev of powerpath that supports the new kernel rev? Note that this is a recurring problem. EMC NEVER supports a current point release of the OS, although sometimes you get lucky and the old module works with the new rev, although that seems to be rare. This sounds like an excellent reason to avoid shitty vendors that release shitty products that rely on proprietary, GPL-violating kernel modules.
    21. Re:Sweet! by walt-sjc · · Score: 1

      I actually meant like Debian, where backported security fixes usually do not change te kernel's ABI. If changing the ABI is unavoidable, then the package containing the kernel is renamed (e.g., linux-image-2.6.18-4-k7 -> linux-image-2.6.18-5-k7).

      Um, no. That's not how it works. Debian revs the package version number (the "4" to "5" in your example) for ALL changes, not just ABI changes. This is how the packing system knows that there is an update.

      This sounds like an excellent reason to avoid shitty vendors that release shitty products that rely on proprietary, GPL-violating kernel modules.

      And therefore, avoiding pretty much all enterprise software. Sorry, not an option in many cases. Oh, I agree the situation sucks, which is why I would like to see ALL binary modules to go away.

    22. Re:Sweet! by Jesus_666 · · Score: 1

      xf86-video-ati would be what Debian uses. Gentoo would rather use xorg-video-ati, as it has already adopted the cutting-edge XFree86 fork called X.Org. ;)

      --
      USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
    23. Re:Sweet! by cortana · · Score: 1

      I actually meant like Debian, where backported security fixes usually do not change te kernel's ABI. If changing the ABI is unavoidable, then the package containing the kernel is renamed (e.g., linux-image-2.6.18-4-k7 -> linux-image-2.6.18-5-k7). Um, no. That's not how it works. Debian revs the package version number (the "4" to "5" in your example) for ALL changes, not just ABI changes. This is how the packing system knows that there is an update. Note that I said they change the package name, not the version.
    24. Re:Sweet! by cortana · · Score: 1
      Oops, hit reply too soon...

      This sounds like an excellent reason to avoid shitty vendors that release shitty products that rely on proprietary, GPL-violating kernel modules. And therefore, avoiding pretty much all enterprise software. Sorry, not an option in many cases. Oh, I agree the situation sucks, which is why I would like to see ALL binary modules to go away. I don't see how "pretty much all enterprise software" requires the use of proprietary binary kernel modules?
    25. Re:Sweet! by makomk · · Score: 1

      Nope; all the X.org driver modules are still prefixed with xf86 for some reason. For example, I might use xf86-video-nouveau (which is now reasonably usable for 2D on my card, if noticably slower for some things than the binary driver).

    26. Re:Sweet! by metalzelot · · Score: 1

      I also have an Geforce 4. Its a Geforce 4 MX 420, a very old model. But its still working and I'm happy with it. There's only one bug working with Beryl. Nvidia told me that they are on the way for a solution. Why I'm responsing is to inform you that it is possible, even with an old Geforce 4 to get the new kernel. All you have to do is to install the Nvidia Legacy Driver. They are still supported and updated, just not as frequently as the normal driver. You can get the legacy driver here: http://www.nvidia.com/object/linux_display_ia32_1.0-9639.html Good luck with them!

    27. Re:Sweet! by Anonymous Coward · · Score: 0

      Since this is an AGP system, I'll also have to replace the motherboard.


      Really? Because I put a fairly recent (6-series) GeForce into an Intel 440BX (AGP 2X) motherboard not all that long ago. And last time I talked to my mom, it was still running strong.

      Just because PCIe is the new hotness doesn't mean that you have to get a completely new PCIe system. There are still plenty of AGP cards being made, including ones made by NVidia.

      Unless you really want to spend $500+ on a video card? While it's true that you can only waste ridiculous sums on PCIe cards, the AGP cards are still available and more sensibly priced. What a tragedy. :)

      (Not to say there aren't sensibly priced NVidia PCIe cards, just pointing out the most expensive NVidia AGP card is - at least - half the price of the most expensive NVidia PCIe card)
  6. It seems to me... by MrNaz · · Score: 4, Interesting

    AMD ie recently making more moves toward the open source community than either it or ATI did prior to the merger.It seems to me that AMD has realized that there is value in not only having the right products rolling off the lines, but also having a greater mindshare.

    Google realized this early, and bought off a great amount of geek awe by using Linux as the basis for its computing grid. This popularity among geeks turned into word of mouth advertising which turned into huge market share (having a great product didn't hurt either). Google still tries to maintain the "we're just a benign bunch of geeks" image (an image which is eroding, as it becomes more apparent that they are more akin to a lovechild of M$ and the NSA than a giant sushi eating LAN party). This appeal to mindshare by making steps toward the community, genuine or not, may be part of what AMD is trying to do, at least to an extent.

    There are other genuine benefits to being more open about its specs, most clearly highlighted by the use of ATI GPUs to process Folding@Home. Therefore it is conceivable that AMD GPUs and GPU/CPU combo chips in the future may, if more openly specced, be used in a wider variety of HPC applications.

    Disclaimer: I am an AMD fanboi.

    --
    I hate printers.
    1. Re:It seems to me... by NerveGas · · Score: 5, Insightful

      Linux kind of carried the Opteron for the first year or so, since it had 64-bit and NUMA support, while M$ obligingly waited to release any such thing until Intel had an offering as well. Maybe AMD learned something from that.

      --
      Oh, you're not stuck, you're just unable to let go of the onion rings.
    2. Re:It seems to me... by RichMan · · Score: 1

      >> Google as it becomes more apparent that they are more akin to a lovechild of M$ and the NSA than a giant sushi eating LAN party)

      Google does search on massive amounts of data and returns relevant results.

      Do you think the NSA did not drive up with a dump truck full of cash and say "give us one" as soon as Google proved their technology worked?

      How many Google stand alone search appliances do you think the NSA has?

    3. Re:It seems to me... by gmack · · Score: 4, Interesting

      AMD ie recently making more moves toward the open source community than either it or ATI did prior to the merger.It seems to me that AMD has realized that there is value in not only having the right products rolling off the lines, but also having a greater mindshare.

      AMD has actually been making moves for a long time so this isn't as large a step for them. Even before the amd64 cpus came out AMD had specs available and a machine simulator as well as several kernel developers working on getting Linux to run on their hardware.

      On the other hand this is a huge step for ATI and I may very well find myself reconsidering my ATI boycott.

    4. Re:It seems to me... by Anonymous Coward · · Score: 1, Informative

      Google realized this early, and bought off a great amount of geek awe by using Linux as the basis for its computing grid.
      Umm.... Are you somehow implying that Google based its initial infrastructure choices on what would gain it the most "geek awe"? In which case, the phenomenal success of their infrastructure is a mere accident? That's patently ridiculous. Google used Linux because it was the right tool for the job (reliable, stable, scalable, Free to be modified, and free to use).

      Sure, the geek cred is a nice by-product of the choice they made. But that was inevitable, since they were geeks making geeky choices about technology (rather than marketing choices).
    5. Re:It seems to me... by MrNaz · · Score: 3, Funny

      Given that Google's initial search technolgy seems to have come out of nowhere and that Google had secretive high government clearance contacts from day one, I think the more likely scenario is the NSA rolling up to Larry and Sergi and saying "we need public sector lovable geek mascots to hide behind while we monitor the population's activities. You two seem suitable and have the right profiles. Here's some search tech, and we'll set you up with the right venture capital connections. No go profile everybody."

      --
      I hate printers.
    6. Re:It seems to me... by MrNaz · · Score: 1

      I don't believe I claimed that was the only reason or even the main reason. It was certainly a factor though, and the cred was exploited for it it was worth. Given that my post was about image, I thought this would be obvious, but as it turns out, there's always an idiot willing to miss even the most plainly visible of contextual qualifications. Well done on it being you!

      --
      I hate printers.
    7. Re:It seems to me... by joeflies · · Score: 1
      Google realized this early, and bought off a great amount of geek awe by using Linux as the basis for its computing grid. This popularity among geeks turned into word of mouth advertising which turned into huge market share (having a great product didn't hurt either).

      Although I think you're probably right on the amount of geek cred google gets for the computing grid, I'll bet that you'll probably need to flip the two arguements around - the word of mouth and market share are driven by great product, not that it runs linux and geeks like it.

      People talk positively and negatively based on the question of "does it work", not "does it run linux". Oddly enough, in AMD/ATI's case, they've had a rocky set of answers for both of those questions.

    8. Re:It seems to me... by N7DR · · Score: 1
      AMD has realized that there is value in not only having the right products rolling off the lines, but also having a greater mindshare.

      Or, as you implicitly suggest later but don't actually say: greater mindshare amongst technologically literate and influential people. My opinion (which is worth what you're paying for it) is that such people wield widely-underestimated power in the mid and long term.

    9. Re:It seems to me... by Somegeek · · Score: 1

      I would say that the buzz about Google came %99.98 from the relevance of their search results as compared to what came before; alta-vista, etc., and %.02 from geek awe over Linux infrastructure.

      --
      And as you tread the halls of sanity, You feel so glad to be, Unable to go beyond. I have a message, From another time..
    10. Re:It seems to me... by LWATCDR · · Score: 1

      I doubt that Google used Linux to gain mind share. They used Linux because they could get the source and was free. It was the cheapest and best solution for building a super large COTS cluster.
      It is nice that AMD gave the 2D specs away. I will go to great when they give out the 3-D specs. I will buy and AMDTI video card when we have good working drivers.

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    11. Re:It seems to me... by Ford+Prefect · · Score: 5, Funny

      I think the more likely scenario is the NSA rolling up to Larry and Sergi and saying "we need public sector lovable geek mascots to hide behind while we monitor the population's activities. You two seem suitable and have the right profiles. Here's some search tech, and we'll set you up with the right venture capital connections. No go profile everybody."

      This conspiracy theory seems incomplete. Did Jimmy Hoffa steal the search technology from Area 51's crashed Roswell UFO, and masqueraded as JFK when FSF supporters attempted to assassinate him for creating the possibility of a faked Apollo moon landing, then went into hiding for many years as Lord Lucan, fathered Princess Diana's unborn child, found Elvis and Marilyn Monroe alive and well in Atlantis, flew an Aurora spy-plane powered by water-fuelled engines through the hole in the North Pole into an unknown hollow Earth down to the South Pole, took this fabled Google search technology to the secret Illuminati base in Antarctica before heading north again, annoyed the Pope and Opus Dei and the long-lost descendants of Jesus Christ and finally became integrated into the Project for a New American Century's headquarters, the NSA - which was almost obliterated when the international Zionist conspiracy felled the Twin Towers with explosives and thermite in the fraudulent September the 11th attacks?

      To be honest, you're not trying very hard. Or giving the real-world NSA lots of credit and assuming no end of competence on their behalf. They've cracked every form of encryption as well, right?
      --
      Tedious Bloggy Stuff - hooray?
    12. Re:It seems to me... by MrNaz · · Score: 1

      People talk positively and negatively based on the question of "does it work", not "does it run linux".

      Dude, seriously? "Does it run Linux?" was such an important question in the common mindset here that it has become a meme.

      --
      I hate printers.
    13. Re:It seems to me... by Anonymous Coward · · Score: 0

      I don't believe I claimed that was the only reason or even the main reason. It was certainly a factor though, and the cred was exploited for it it was worth.
      Okay, fair enough. From my reading it sounded like you were claiming that was a primary deciding factor. Thank you for clearing that up.

      Given that my post was about image, I thought this would be obvious, but as it turns out, there's always an idiot willing to miss even the most plainly visible of contextual qualifications. Well done on it being you!
      Well, I think your post could have been more carefully worded to get your point across. In any case, I think you calling me "an idiot" is uncalled for. I apologize if my comment (e.g. usage of "patently ridiculous") bothered you.
    14. Re:It seems to me... by Anonymous Coward · · Score: 0

      "AMD ie recently making more moves toward the open source community than either it or ATI did prior to the merger.It seems to me that AMD has realized that there is value in not only having the right products rolling off the lines, but also having a greater mindshare."

      Also, having a (hopefully) large developer pool optimize your drivers for free doesn't hurt, either. I'd bet anything that stable well performing driver code produced by the community for open source drivers makes it back into their binary distributions for other platforms in short order (source available upon request, of course).

      AMD's commodity market is hardware, and they already give away their drivers for free, so this does nothing but help them in both the short and long term.

    15. Re:It seems to me... by AJWM · · Score: 3, Insightful

      I think AMD said almost as much when they announced that they'd be releasing specs and open drivers for these GPUs. The next step in processor development will be to combine the CPU and GPU on the same chip, and AMD wants to be sure that Linux and other OSS is there to support it.

      --
      -- Alastair
    16. Re:It seems to me... by tpholland · · Score: 1

      Parent is clearly an NSA asset.

      Anyone with true Agartthan wisdom would know that you can cut your water-fuelled engines when travelling between the poles. The telluric currents are quite adequate for the journey.

    17. Re:It seems to me... by darkwhite · · Score: 1

      Google realized this early, and bought off a great amount of geek awe by using Linux as the basis for its computing grid. This popularity among geeks turned into word of mouth advertising which turned into huge market share (having a great product didn't hurt either). Google still tries to maintain the "we're just a benign bunch of geeks" image (an image which is eroding, as it becomes more apparent that they are more akin to a lovechild of M$ and the NSA than a giant sushi eating LAN party). Has it ever occurred to you that your grasp on reality might be just a little bit loose?
      --

      [an error occurred while processing this directive]
    18. Re:It seems to me... by drsmithy · · Score: 2, Interesting

      Linux kind of carried the Opteron for the first year or so, since it had 64-bit and NUMA support, while M$ obligingly waited to release any such thing until Intel had an offering as well.

      Opteron released: April 22, 2003.
      Windows 2003 released: April 24, 2003.

      2 days != a "year or so".

    19. Re:It seems to me... by markov_chain · · Score: 2, Insightful

      reconsidering my ATI boycott.

      See, the boycott worked!

      --
      Tsunami -- You can't bring a good wave down!
    20. Re:It seems to me... by MikeBabcock · · Score: 1

      AMD was also selling 64 bit desktop CPUs and Windows XP 64-bit was not available for some time. Home users were not likely to be running Windows Server 2003, thus the GP's post I'm sure.

      --
      - Michael T. Babcock (Yes, I blog)
    21. Re:It seems to me... by dreddnott · · Score: 3, Informative

      That's great, except x64-native versions of Microsoft Windows Server 2003 and Windows XP were not released until 2005. Until then, Windows Server 2003 was available only in 32-bit x86 and Itanium-compatible flavours. Grandparent's observation is quite astute.

      --
      I may make you feel, but I can't make you think.
    22. Re:It seems to me... by drsmithy · · Score: 1

      That's great, except x64-native versions of Microsoft Windows Server 2003 and Windows XP were not released until 2005. Until then, Windows Server 2003 was available only in 32-bit x86 and Itanium-compatible flavours.

      Good point. I'd forgotten about that.

      Apologies to the GGP, my bad.

    23. Re:It seems to me... by drsmithy · · Score: 1

      AMD was also selling 64 bit desktop CPUs and Windows XP 64-bit was not available for some time. Home users were not likely to be running Windows Server 2003, thus the GP's post I'm sure.

      I doubt it, more likely he was referring to the x64 version of Windows 2003 lagging a couple of years behind (which I'd forgotten).

      For the vast majority of XP users, 64 bit had pretty much zero benefit (and quite a few disadvantages) until maybe a year - 18 months tops - ago.

    24. Re:It seems to me... by mgblst · · Score: 1

      Google realized this early, and bought off a great amount of geek awe by using Linux as the basis for its computing grid.


      You are being ridiculous. Goolge used Linux on their grid not to impress the meagre geek community. Saying so is moronic and insulting to geeks and to Linux. They did it because Linux is the best operating system for them. If you think anything else you are being very stupid.

      Disclaimer: I am an AMD fanboi.


      Ah, I see the problem. Fanboi means that you enjoy something without actually thinking about it, or it actually being any good. I prefer a much more reasoned approach. I support AMD because we have seen what Intel (and all monopolies) produce without it. Competition is necessary.
    25. Re:It seems to me... by Jesus_666 · · Score: 2, Insightful

      For the vast majority of XP users, 64 bit had pretty much zero benefit (and quite a few disadvantages) until maybe a year - 18 months tops - ago.

      In fact, I know nobody who actually uses 64-bit Windows at home. Everyone has his AMD64-compatible setup, but even Vista goes purely in 32-bit mode because the 64-bit version has compatibility problems, stability or performance issues etc.

      The only boxen I've seen actually make use of their 64-bit architecture were Linux boxen and Macs (kind of; pure 64-bit mode coming with Leopard). Windows users usually treat their AMD64 CPUs as really fast i686s.

      --
      USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
    26. Re:It seems to me... by Jesus_666 · · Score: 1

      In fact, you can do entirely without them if your vehicle was properly designed with the cubic nature of time in mind.

      --
      USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
    27. Re:It seems to me... by Anonymous Coward · · Score: 0

      That was amazing. But you left out the Denver airport. And tinfoil. Tinfoil must fit in somewhere.

  7. Hopefully a meaningful contribution by raydobbs · · Score: 3, Interesting

    One can hope that it actually had the specifications for modern GPUs... and not just stuff you might find in scrap piles or in 15+ year old government computers. Otherwise, it will be like when the RIAA gave a crap-ton of Whitney Houston Christmas CDs as a settlement for their price-fixing practices... technically within the letter of the law, but violating the spirit of the law all to hell...

    1. Re:Hopefully a meaningful contribution by CaptainPatent · · Score: 2, Funny

      it will be like when the RIAA gave a crap-ton of Whitney Houston Christmas CDs as a settlement for their price-fixing practices... technically within the letter of the law, but violating the spirit of the law all to hell... Not to mention violating our ears too!
      --
      Well, back to rejecting software patent applications.
    2. Re:Hopefully a meaningful contribution by Anonymous Coward · · Score: 0

      That is the case -- AMD/ATI is putting out specs for everything on their latest cards, and will do so from here on out. Even better, there's no NDA stuff around it, which was the problem with ATI's previous dip in the open source pool: there were only a small handful of X11 devs with access under the NDA, and when they moved on, all the benefits of the spec "release" went with them.
        I've always had a slight preference for ATI hardware, going all the way back to my DOS 6 machine in the early 90's, which had a very nice little ATI card. The image quality of ATI cards has always been excellent, and they've never been much of a slouch at 3D, it's always just been the crappy drivers that were the problem, so I'm really happy about both these specs, and the new 8.41 driver that Phoronix previewed -- my old Radeon 9800 is finally getting good drivers, and my next card (which was going to be an nVidia up until AMD made this announcement earlier this month) will get free drivers.
        The future's looking up!

    3. Re:Hopefully a meaningful contribution by ThisNukes4u · · Score: 5, Informative

      The specs are for the brand-spanking-new RV630 series, the mid to low range chip in the r600 line. And the specs are only for 2d modesetting at the moment, so they likely apply to the whole r600 series. When the 3d specs are released they'll likely be a separate spec sheet for each specific chip. So to answer your question, they are for the newest cards ATi currently makes not their old, outmoded ones.

      --
      thisnukes4u.net
    4. Re:Hopefully a meaningful contribution by VGR · · Score: 1

      I agree with you, except that the older cards should not be written off. I wonder how many of those older All-In-Wonder cards would be sitting in scrap piles if their video ports were usable in Linux. I know I'd use them. I get the impression they're quite a bit better than BT8xx cards. Sadly, they're currently only usable as generic SVGA cards.

      Yes, I know there have been some attempts at reverse-engineering the older cards' video ports, but it's spotty, difficult to use, and if I remember correctly, requires not only an X module but also patching and recompiling X itself. Plus special binaries instead of v4l or v4l2 support.

      It seems ridiculous to me that so many cards are sitting in landfills for no reason other than ATI's unwillingness to release specs for them. What is there to lose by releasing specs for old cards?

      --
      The Internet is full. Go away.
    5. Re:Hopefully a meaningful contribution by Jugalator · · Score: 4, Informative

      The RV630 chip is anything but "old", and what runs the Radeon HD 2600 PRO and Radeon HD 2600 XT. These belong to the very latest Radeon R600 line -- AMD's Direct3D 10 / Shader Model 4.0 supporting GPU's. These are high end chips from 2007, currently only beat by the Radeon HD 2900 XT in performance, if only speaking of AMD/ATI. The M56 chip is the core of the ATI Radeon Mobility X1600 (released December 2005), which is still a very decent mobile chip, roughly corresponding to an NVIDIA Geforce Go 7600 in performance.

      Hmm, I now also see the ATI FireGL V7600 runs the RV630 too. Maybe that could work out for something too...

      --
      Beware: In C++, your friends can see your privates!
    6. Re:Hopefully a meaningful contribution by Rod+Beauvex · · Score: 0

      Possibly, the fact that that and older card with perfectly optinmised drivers might actually perform better than a newer one with badly written drivers. Seems to me that S3 and XGI learned this the hard way.

    7. Re:Hopefully a meaningful contribution by fastest+fascist · · Score: 1

      What they have to lose is the work it takes to do that. They said there's proprietary IP involved in the specs for the newer cards that they'll have to find a way to not release along with the specs, probably there are similar issues with the old cards. And the older cards no longer generate revenue, so from ATI/AMD's point of view it's not really worth the trouble.

    8. Re:Hopefully a meaningful contribution by Ruie · · Score: 2, Informative
      I just want to highlight a couple of points about released docs.

      In the past the hardware did not change much between releases (which is why one can have unified drivers). In particular, having documentation for one card goes a very long way to figuring out how to use a similar, slightly modified card. What happened after R300 is that ATI ripped out their 2d engine which made it impossible to figure out how to set modes on the new cards. Thus, we did not even have a 2d X-server, let alone 3d. With this documentation one can have a driver that allows distributions to boot directly to the largest supported mode - and then download binary driver if they feel like it.

      Secondly, these documents are now public as opposed to limited to ATI-approved select developers. This makes a huge difference as more people would be able to contribute. Before, with experimental drivers, the outside developers could only provide feedback and getting them approved was not a speedy process. With docs available, not only the driver quality will go up but also one can hope for new and creative use of the chips. I really can't stress enough importance of having public documentation.

      - Eagerly waiting for the 3d part - which is also needed for old-fashioned 2d graphics..

    9. Re:Hopefully a meaningful contribution by VanessaE · · Score: 1
      It's been my experience, having owned both an AiW (128 pro) and now a BT848-based card, that the BT848 has a bit better quality, plus of course the ATI was problematic from the get-go (because of the closed-source drivers, thanks goes to the GATOS team back then for figuring most of that out).


      Plus, the BT848 doesn't conflict with my nVidia 6800, and it's drivers have been open-source and in-kernel for ages.

  8. Nice, but... by Briareos · · Score: 4, Funny

    ...will that GPU run Linux?

    Just imagine an SLI'd Beowulf cluster of these!

    np: Masha Qrella - Insecure (Luck)

    --

    "I'm not anti-anything, I'm anti-everything, it fits better." - Sole

    1. Re:Nice, but... by Anonymous Coward · · Score: 0

      ATI uses Crossfire, you insensitive clod!

    2. Re:Nice, but... by the_humeister · · Score: 1

      That's not an invalid question. GPUs nowadays are programmable enough that they're being used for general purpose computing. The question is, are current graphics cards general purpose enough to be able to actually run their own operating system? If so, which ones can currently do that?

    3. Re:Nice, but... by Briareos · · Score: 1

      Well, duh... all the better reason to ask for SLI, then... :P

      --

      "I'm not anti-anything, I'm anti-everything, it fits better." - Sole

    4. Re:Nice, but... by pakar · · Score: 1

      Would be cool to run the window-manager on the gfx-card instead of the cpu.. :)

  9. Wow! by nonsequitor · · Score: 3, Funny

    This is amazing news, not only that the specifications have finally been opened, but that the open source community has immediately utilized them to update the driver with a turn around time of only 2 weeks.

    I guess we can thank Dell for pressuring ATI for better Linux support.

    1. Re:Wow! by jmitchel!jmitchel.co · · Score: 5, Informative

      Actually, (IIRC) the driver has been sitting in somebody's desk drawer for months, waiting for AMDTI to bless it. The developer got the specs under NDA or something quite a while ago.

    2. Re:Wow! by lasse_dk · · Score: 2, Funny

      if it only takes a week to write a driver with 900 pages of spec, how long can it take to write one with no spec?

    3. Re:Wow! by mtippett · · Score: 4, Informative

      Bzzzz - wrong.

      Read David's blog - http://airlied.livejournal.com/ - there are a whole pile of potential problems about that driver. David accepts that it was on questionable ground, and so it will probably never see the light of day.

    4. Re:Wow! by revengebomber · · Score: 1

      DIV0

      great, now you've crashed my brain.

      --
      09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
      45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
    5. Re:Wow! by marcosdumay · · Score: 1

      "I guess we can thank Dell for pressuring ATI for better Linux support."

      Or maybe the reason AMD brought ATI was to make a FOOS friendly system.

      Too bad ATI had to go near banckrupt and become another company before even thinking about releasing specs for the community.

  10. ati on ubuntu dell laptops? by wpegden · · Score: 1

    Maybe now this means we can have some dedicated video cards on the Ubuntu Dell Laptops, instead of just the onboard Intel chip they're using now because of the driver issues?

  11. Hmm... by Rod+Beauvex · · Score: 3, Insightful

    I wonder if this has more to do with trying to get mind and market share over intel than them really beleiving Open Source is the future of the market. maybe it's both.

    Nice bit of good news anyway.

    1. Re:Hmm... by ScrewMaster · · Score: 1

      From their perspective, who cares if Open Source is the future of the market (it's doubtless the future of some portion of the market.) What does matter is that non-Microsoft, non-proprietary systems are a significant and growing aspect to modern computing, and they'd just continue to be fools to ignore it.

      --
      The higher the technology, the sharper that two-edged sword.
  12. Re:I'll replace my nVidia when I see a good review by plague3106 · · Score: 1

    Good luck finding an AGP card; I don't think they're making many these days.

  13. Re:I'll replace my nVidia when I see a good review by DaedalusHKX · · Score: 3, Interesting

    Actually this is the fun part. Governments have been "enforcing" open source as gimmicks. The only way to show there is a REAL market is to have an actual producer get involved and actually PROVIDE the goods and support. Red Hat did its part, various OSS groups did their part, etc.

    They weren't tax supported, but they did a better job than all the tax supported wealth consuming agencies out there :)

    I agree, once the cards hit my neck of the woods, if they're well implemented in hardware, I'll gladly supplant my 7800's in my SLI rig :) (Or build another one.)

    --
    " What luck for rulers that men do not think" - Adolf Hitler
  14. 900 pages? by 26199 · · Score: 5, Funny

    Come off it... that's not even enough for an Office document standard.

    Worthless!

    1. Re:900 pages? by cnettel · · Score: 1
      But this is only the 2D parts, no 3D formulas (no specified shader language, akin to spreadsheet formulas). And I would damn assume that they have left out the VESA BIOS specs, so no backwards compatibility!

      (jk, not troll)

    2. Re:900 pages? by TheRaven64 · · Score: 1

      Considering that this 900 page spec is very tersely written and is only enough to control basic monitor setup really puts ODF into perspective.

      --
      I am TheRaven on Soylent News
    3. Re:900 pages? by moonbender · · Score: 1

      Maybe AMD/ATI should have used XML!

      --
      Switch back to Slashdot's D1 system.
  15. My next box will go ati by FictionPimp · · Score: 1

    I have not bought an ati for a long time after I switched to linux. My last ati card was a 9800. I will next time.

    1. Re:My next box will go ati by Anonymous Coward · · Score: 0

      My currently-in-use card is an 9800, you insensitive clod. SCNR, it really is. ;-)

    2. Re:My next box will go ati by Anonymous Coward · · Score: 0

      My last ATI was a Rage128 :) Has perfectly working FLOSS drivers, even 3D :)

    3. Re:My next box will go ati by pak9rabid · · Score: 1

      My last ATI card was a.....oh wait, i've never bought one, but that could very well change in the near future.

    4. Re:My next box will go ati by marcosdumay · · Score: 1

      I'll wait untill they publish the 3d specification before buying them. But I probably won't buy anything in the meantime anyway.

  16. HOLY FREAKING CRAP by tietokone-olmi · · Score: 1

    I say again. HOLY FREAKING CRAP.

    If this turns out the way it seems from an optimistic reading of the /. post, I think there's absolutely no reason not to recommend an ATI display card anymore. (Well, except for Intel's or VIA's integrated graphics in a non-gaming desk- or laptop.)

    1. Re:HOLY FREAKING CRAP by pppppppman · · Score: 1

      .. apart from the fact they only open specced the 2D display registers. What would be nice is to open spec the 3D parts and... THE DRIVERS THEMSELVES!

      Maybe next month...

  17. Amazing, we're saved! by Duncan3 · · Score: 3, Funny

    Bits 12 and 13 of D2CRTC_TRIGB_CNTL are D2CRTC_TRIGB_RISING_EDGE_DETECT_CNTL !!!

    Hurray, now all Linux graphics problems are solved, it will autodetect all graphics cards like Windows 1.0 did and penguins will dance in the streets.

    --
    - Adam L. Beberg - The Cosm Project - http://www.mithral.com/
    1. Re:Amazing, we're saved! by jamarsa · · Score: 1

      Sorry for ruining your joke, but windows did recognize the cards not by itself, but by installing manufacturer's provided drivers....

    2. Re:Amazing, we're saved! by Workaphobia · · Score: 1

      On the one hand, it sounds like you've never touched hardware with a 10 foot poll.

      On the other hand, you have a 5 digit UID.

      --
      Evidently, the key to understanding recursion is to begin by understanding recursion. The rest is easy.
    3. Re:Amazing, we're saved! by Duncan3 · · Score: 1

      It's funny to us embedded systems types. 900 pages of register definitions alone does nothing.

      --
      - Adam L. Beberg - The Cosm Project - http://www.mithral.com/
    4. Re:Amazing, we're saved! by makomk · · Score: 1

      On the other hand, it does give them a way of making sense of the register accesses the closed-source driver is doing. (I think there are ways of intercepting register writes to graphics cards.)

      Then again, since I'm currently staring at a list of register writes to a chip (a DVB-T demodulator now owned and sold by Intel, no public register documentation) with no way of working out what they mean, I would say that.

  18. Re:I'll replace my nVidia when I see a good review by Anonymous Coward · · Score: 2, Insightful

    Are you on crack? Did anyone say anything about government? Has government ever made a peep about video cards and closed or open drivers? Did you read the summary before you spouted off?

    STFU, FOAD, and take your paranoid attitude with you.

  19. Re:I'll replace my nVidia when I see a good review by ryszards · · Score: 3, Insightful

    Actually there's a good number of modern AMD D3D10 products available on AGP now, and the older R5-series hardware had good AGP presence as well. Not the high-end R600 I should say, but RV630 and RV610 (HD 2600 and HD 2400) are both available. And the Windows Vista driver sucks, somewhat hilariously.

    --
    - 'sup, G?
  20. You may be right ... by khasim · · Score: 1

    So what you're saying is that I should get a completely new system, all AMD-based?

    And that I should send a copy of the receipt to AMD along with the explanation that the only reason I spent that money was so I could run Ubuntu with the new Free video driver? And that I should say that the system I'm replacing was Intel/nVidia?

    Well, if you say I should, that's good enough for me.

    And you go on to say that everyone who buys an ATI card because of this should also send a letter (not email) to ATI saying the same thing?

    1. Re:You may be right ... by Rolgar · · Score: 2, Insightful

      I would say the message would come across better if you send it to nVidia.

      The big effect will be if every Linux OEM started shipping Radeon in every box, that could be a pretty big number of lost potential sales that they weren't considered for solely based on software.

      This could really be huge in the progress towards making Linux mainstream. The last few times I've installed Linux, installing my 3D drive for nVidia has required a few steps most users wouldn't or couldn't do. Several distros won't automatically set this up during install because the drivers aren't open. As soon as we get these drivers, Linux setup and support should get that much easier, because the installer will pop up a dialog that says "RADION XXXX detected, install 3D acceleration? Y/N."

      We're getting a few steps closer to the mainstreaming of Linux, which will snowball once games and other software starts to include Linux binaries/installers on the disks that you can buy retail.

    2. Re:You may be right ... by thephotoman · · Score: 2, Insightful

      Better idea: instead of popping up a dialog asking to install 3D acceleration, the installer just does it. After all, it'll be free software.

      --
      Haec merda tauri est. Ceterum censeo Carthaginem esse delendam.
    3. Re:You may be right ... by ultranova · · Score: 3, Insightful

      Better idea: instead of popping up a dialog asking to install 3D acceleration, the installer just does it. After all, it'll be free software.

      An even better idea: since a Free driver can be included in the kernel source and compiled into a module, the installer doesn't have to do anything special to enable 3D acceleration. It just installs all available kernel modules as normal and the kernel figures it out at bootup time and loads the ATI driver if appropriate.

      --

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

    4. Re:You may be right ... by Ford+Prefect · · Score: 0, Flamebait

      Better idea: instead of popping up a dialog asking to install 3D acceleration, the installer just does it. After all, it'll be free software.

      No, it'll have to ask the user - because the 3D driver will have various instabilities and have a tendency to hard-lock the machine at random intervals for some ill-defined reasons. Years will go by, and the situation won't improve - because by then, the hardware in question will be obsolete, and nobody will be bothering to improve the drivers.

      Okay, I'm being pessimistic - but something pretty similar happened after Matrox released the specifications for its 3D graphics cards. There were fully open-source drivers, but they weren't exactly high-quality. I moved on to Nvidia after that...
      --
      Tedious Bloggy Stuff - hooray?
    5. Re:You may be right ... by Chandon+Seldon · · Score: 2, Funny

      Okay, I'm being pessimistic - but something pretty similar happened after Matrox released the specifications for its 3D graphics cards. There were fully open-source drivers, but they weren't exactly high-quality. I moved on to Nvidia after that...

      Was Matrox even producing products at that point, or were you expecting one of the other six guys with old Matrox cards to support your drivers?

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    6. Re:You may be right ... by Ford+Prefect · · Score: 1

      Was Matrox even producing products at that point, or were you expecting one of the other six guys with old Matrox cards to support your drivers?

      This was in the days when the G400 was still competitive hardware - and Slashdot made a big noise about it all back then too.

      Everything started off well (I mean, that John Carmack helped write the initial G200 driver), but things kind of fizzled out once things were about 95% working for the G400. Previously, I'd always bought Matrox, but after that I was happy to move on to Nvidia - despite the new drivers for Linux being binary-only.
      --
      Tedious Bloggy Stuff - hooray?
    7. Re:You may be right ... by Anonymous Coward · · Score: 4, Funny

      The big effect will be if every Linux OEM started shipping Radeon in every box, that could be a pretty big number of lost potential sales that they weren't considered for solely based on software.

      Where's the -1 delusional mod?

    8. Re:You may be right ... by RobBebop · · Score: 1

      once games and other software starts to include Linux binaries/installers on the disks that you can buy retail.

      I'll take the source code for the next big game for free and compile it myself, then pay for the the rights to join their servers and/or pay for the hardware to setup and run my own servers.

      Quake 3: Arena source is open.... maybe one of these are appealing alternates to Halo and Half Life.

      Also... if any of these developments does ever provide a good diversion for me (and I'm not paying for the use of the developer's servers), then $20-40 from myself and each of my friends playing the game to the developer's for their troubles.

      --
      Support the 30 Hour Work Week!!!
    9. Re:You may be right ... by billcopc · · Score: 1

      Sad as it may be, I've been grossly disappointed with Matrox since 1999 or so. I don't even understand why they're still in business today. The only product of theirs that I consider noteworthy is the TripleHead2Go, and they can shove their PCI and AGP cards where the sun don't shine. Their drivers are flaky, the hardware is temperamental, and the performance just isn't there anymore. I get more mileage out of a low-end ATI or NVidia product at a third of the cost.

      --
      -Billco, Fnarg.com
    10. Re:You may be right ... by Knuckles · · Score: 2, Informative

      I don't even understand why they're still in business today.

      The world does not revolve around 3D. Matrox produce fine gear for high-quality 2D work, like medical imaging.

      --
      "When I first heard Daydream Nation it quite frankly scared the living shit out of me." -- Matthew Stearns
    11. Re:You may be right ... by cortana · · Score: 1

      I don't remember the G400 ever being competitive hardware.

    12. Re:You may be right ... by kestasjk · · Score: 1

      A non-free driver can also be loaded as a module on boot-up, this is pretty common practice in all distros that aren't strictly FOSS-only.

      --
      // MD_Update(&m,buf,j);
    13. Re:You may be right ... by leenks · · Score: 1

      But what about choice?! This is about open source - we've got to have choice otherwise what do we have? ;-)

    14. Re:You may be right ... by harrkev · · Score: 1

      Matrox produce fine gear for high-quality 2D work
      Huh? I do not understand this at all. What does "high quality" mean in terms of 2D?

      Does it mean that the 2D acceleration is awesome? With current bus speeds, you could take out all acceleration, and probably not even notice a difference on a 2D desktop (including XP).

      Does it mean that the image quality is awesome? If so, that relates strictly to the DAC installed (one single part), and is rendered obsolete by using a digital DVI interface.
      --
      "-1 Troll" is the apparently the same as "-1 I disagree with you."
    15. Re:You may be right ... by default+luser · · Score: 1

      I guess you have a poor memory. How about a refresher?

      I used my G400 MAX to play Counterstrike, Quake III, MDK2, etc. The 32-bit Direct3D performance was second to none. The OpenGL performance using the TurboGL was outstanding, and supported all OpenGL games in existence. I had a lot of faith in Matrox, until they casually ignored the powerful GeForce 256 and announced the G450. What a horrible memory.

      --

      Man is the animal that laughs.
      And occasionally whores for Karma.

    16. Re:You may be right ... by Knuckles · · Score: 1

      You can browse matrox.com yourself, can't you?

      --
      "When I first heard Daydream Nation it quite frankly scared the living shit out of me." -- Matthew Stearns
    17. Re:You may be right ... by awing0 · · Score: 1
      ...and is rendered obsolete by using a digital DVI interface.

      +1 Punny

      --
      Cthulhu Saves.
    18. Re:You may be right ... by Anonymous Coward · · Score: 0

      With current bus speeds, you could take out all acceleration, and probably not even notice a difference on a 2D desktop (including XP)


      Yes, if we all pull ideas out of our ass and make facts up on the spot, everything is possible.

      Here's an idea - how about you temporarily disable all acceleration in your drivers. Put your theory to the test, with less than 5 minutes of effort on your part. Then, instead of blindly conjecturing about what you think the world is like, you could actually find out what the world is like.

      Because I've had to disable all 2D acceleration to debug & work around 2D problems. And based on my real-world experience, with PCIe x16 cards mind you, we're still nowhere near as far along as you believe we are.
    19. Re:You may be right ... by billcopc · · Score: 1

      That's what they've been telling us ever since the Mystique came and went. The truth is, I've worked on many of their cards and I don't see the difference AT ALL. In fact, I see problems with the Matrox gear that I don't see with my "low-quality 2D" NVidia boards; problems like weird noisy pixels (animated!), imprecise rectangle blits - drag a window around and it jumps back and forth around the mouse pointer like Parkinson's disease, oh and I shouldn't forget the best glitch of all: even their best "2D" card can't drive a 27" or 30" LCD... you have to step up to the sad sad Parhelia to get anything above 1920x1200.

      It's 2007, any idiot with an FPGA can design a fast RAMDAC and rectangle accelerator. Matrox has dozens of products in its lineup, the only differences between them all are the number and type of connectors, and the multiplier applied to the price. What I don't understand is why did Number Nine die and Matrox survive ?

      --
      -Billco, Fnarg.com
    20. Re:You may be right ... by billcopc · · Score: 1

      Not so fast there buddy... it's not just about the bus, latency plays a role too. If you want to see unaccelerated 2D performance, just fire up Linux and Xorg with the plain VESA driver. It's usable, but you can definitely feel the weight of the software blitter when you drag things around and realize the window's lagging behind your mouse pointer by a few tenths of a second or more.

      Bus speeds have improved significantly over the years, but so has the data traffic imposed on them. Hey, if you want to run in 640x480 4-bit, knock yourself out. My current desktop resolution chews up around 92 times more memory than decade-old VGA, which means it needs to blit stuff at least 92 times faster for me to not notice any slowdown. Here's a hint: PCIe ain't 92 times faster than PCI or even ISA.

      --
      -Billco, Fnarg.com
    21. Re:You may be right ... by harrkev · · Score: 1

      You are absolutely right. 2D acceleration is very nice to have. But, here is another hint: 2D acceleration is not rocket science either. I design digital hardware (but not aimed at general consumers). I could come up with some pretty nice 2D acceleration all on my own. You have basic things like block fills and block copies, and that goes a long way towards speeding things up.

      --
      "-1 Troll" is the apparently the same as "-1 I disagree with you."
  21. Definitely a good move. by Mr_Mirsal · · Score: 1

    These are great news for the FOSS world, and I hope it will be for AMD too. This is one of the things the GNU/Linux desktop world was awaiting the most, and this will prove there is a real market there if AMD benefits from opening their specs.

    IMHO, effectively releasing full specs, and doing it fast is not only a matter of "looking good from the geeks point of view" but a real desire of taking the GNU/Linux (and BSDs) desktop from nvidia.

  22. No, this is normal for AMD. by Anonymous Coward · · Score: 0

    They released full specs for their hardware for ages, and even donated hardware to open source projects. Not just big name ones like linux either, openbsd got AMD64 hardware. This is why all the free operating systems had full AMD64 support very soon after the hardware was released (if not before it was released).

  23. It's It's, not Its by bilbravo · · Score: 2, Funny

    google it. :-)

    1. Re:It's It's, not Its by Anonymous Coward · · Score: 0

      you make me wish there was a mod point for pumpis asshole

    2. Re:It's It's, not Its by Anonymous Coward · · Score: 0

      I got over 900 hit's.

    3. Re:It's It's, not Its by Anonymous Coward · · Score: 2, Funny

      WHAT nine HUNDRED?!

    4. Re:It's It's, not Its by I+Like+Pudding · · Score: 4, Informative

      Specifically, google "over nine thousand" since the OP was paraphrasing. On a related note, my karma is over NINE THOUSAND!

    5. Re:It's It's, not Its by bilbravo · · Score: 1

      My bad, I thought he was just correcting the "900+" saying it should say "over 900", so I was doing the same. Joke, sorry.

    6. Re:It's It's, not Its by Anonymous Coward · · Score: 0

      Channelling the spirit of Az, are we? He'll probably want it back sooner or later.

      http://www.gaijinsmash.net/

  24. Within by jrwr00 · · Score: 1

    Within 6 months i bet nvidia does the same thing

    1. Re:Within by ewhac · · Score: 5, Interesting
      I will take that bet. Shall we say a six-pack of winner's choice of beer?

      NVidia have been stalwart protectors of their hardware designs, mostly due to historical accident. A few of the principals at NVidia used to work at Sun, where they designed the GX graphics chip. As it turns out, a version of SunOS was released with a header file describing the chip's registers. Using that -- and a logic analyzer -- a company called Weitek successfully created a functional clone of the chip that was good enough such that Sun's own drivers worked on it. This stuck in the craw of the Sun guys, and evidently vowed no such thing would happen again.

      Another historical accident was that NVidia did, in fact, have a few source code releases way back. And every time they did, so it seemed, they got hit a few weeks later with a patent infringement lawsuit, usually from SGI. NVidia solved this latter problem largely through the expedient of buying SGI.

      So, no, I don't think they're going to do it, and certainly not within six months. And yes, I would be perfectly tickled to be wrong about that.

      Schwab

    2. Re:Within by Workaphobia · · Score: 1

      Even if that's true, it seems it might still be prudent to buy from the company that made the choice to support linux first.

      --
      Evidently, the key to understanding recursion is to begin by understanding recursion. The rest is easy.
    3. Re:Within by Don_dumb · · Score: 2, Interesting

      Another historical accident was that NVidia did, in fact, have a few source code releases way back. And every time they did, so it seemed, they got hit a few weeks later with a patent infringement lawsuit, usually from SGI. NVidia solved this latter problem largely through the expedient of buying SGI. A better bet would be how long before someone hits AMD with a patent lawsuit. - I say days.
      --
      If this were really happening, what would you think?
  25. Oblig. Monty Python by SwordsmanLuke · · Score: 1
    And there was great rejoicing.

    Slashdot: Yay!

    --
    Any plan which depends on a fundamental change in human behavior is doomed from the start.
    1. Re:Oblig. Monty Python by bugnuts · · Score: 1

      Narrator: In the frozen land of nVidia, they were forced to eat 900 pages of ATI documentation.
      Slashdot: Yay!

  26. Next graphics card: Ati by A+beautiful+mind · · Score: 1

    My current card is nvidia because they had the best 3D drivers so far. That's going to change.

    Thanks AMD for taking this step!

    --
    It takes a man to suffer ignorance and smile
    Be yourself no matter what they say
    1. Re:Next graphics card: Ati by Anonymous Coward · · Score: 0

      My current card is nvidia because they had the best 3D drivers so far. That's going to change. Are you ready to bet your first born on that?

      Just because it is available doesn't mean that the drivers will become better. Maybe over time but by then nVidia have had time to fire back.
  27. Thank you ATI + AMD! by Anonymous Coward · · Score: 0

    I promise that you will get my business for as long as your drivers stay open-source. I bought an AMD CPU already and I'll buy the rest (ATI graphics card and motherboard w/ATI subsystem) once I get paid.

    May you bottom-line go through the roof, your hardware stay cool (both literally and figuratively), and your (Linux) drivers stay open source.

    - On a technical note, how long would it take for MPlayer/VLC/GSTREAMER people to have hardware-accelerated playback of DVDs and the like?

  28. juicy news! by jhwilliams · · Score: 1

    Boy oh boy, this is some juicy news. I had stopped using ATi stuff cause it blew. Maybe I'll start to see some new options open up.

  29. Too late..... by Anonymous Coward · · Score: 0

    I just bought a laptop with an nVidia card in it; ATI was too late to get their share of my $2200.

    However, for my next ~$5000 desktop, ATI will get the nod.

    Thank you, AMD.

    captcha: solemnly

    1. Re:Too late..... by Anonymous Coward · · Score: 0

      $5000 desktop? You could probably get a rackfull of servers for that much money, not just a desktop... Or are you one of those people that gold-plate their heatsinks and order their RAM directly from Apple?

    2. Re:Too late..... by pakar · · Score: 1

      $5000 is nothing if you really think about it, and it wont get you any good rack-servers... Those are usually around $2-3k at least.

      $5000 system.

      Good chassis with only passive cooling, around $1000
      CPU around $700 AMD Athlon64 FX
      RAM around $800 4Gb of some good low-latency memory or maybe 4x2Gb DDR2's
      MB around $200-$300
      Soundcard $300-$400
      HD's $800 for a few 400Gb SATA drives in a raid0+1. Or maybe a bit smaller and faster SCSI drives, but then you would need a scsi card too.
      I-RAM drive $400 Gigabytes i-ram disk so you can have a disk with real low latency, 150 for the i-ram card and 250 for the DDR memory.
      Screen 24" $1200
      Gfx-Card $2400 ATI FireGL T2 V8650 Dual-DVI 2GB... And Nvidia Quadro goes up to around 8000 for the-top-of-the-line cards.

      = around $8000-$9000

      So $5000 for a maxed out machine is not really that much...

      The prices are converted from swedish SEK without knowing the current rate, but it should give you a rough estimate...

  30. So, which cards does this cover? by Eneff · · Score: 2, Interesting

    I am not a hardware hacker, so I was wondering what cards would benefit from this first release.

  31. Holy Crap Batman !!!!!!! by sundru · · Score: 1

    Open source drivers for my X1900XT :):):) Hot diggity damn ! Gonna start a petition to port command and conquer - tiberium wars to Linux. or reopen my subscription to transgaming. -Sundru

    1. Re:Holy Crap Batman !!!!!!! by Anonymous Coward · · Score: 0

      Consider supporting Wine/CrossOver instead. That way everybody benefits. For me I find Wine is getting very noticeably better with every release. http://appdb.winehq.org/appview.php?iVersionId=7440

  32. Great! More fans of Bob Llama are pissed at Google by Anonymous Coward · · Score: 0

    >Google still tries to maintain the "we're just a benign bunch of geeks" image (an image which is eroding, as it
    >becomes more apparent that they are more akin to a lovechild of M$ and the NSA than a giant sushi eating LAN party).

    I think Google's actions towards the open source community and users worldwide has been excellent.

    You are I am sure mad at them because of their dealings with China over the treatment of Bob Llama...

    blah, blah, blah....

  33. Well hold on there by Sycraft-fu · · Score: 5, Insightful

    They've released the specs, this doesn't mean anything yet. People forget just how complex graphics cards are. Writing a driver for something like a network card or SCSI controller is fairly easy, and that's also evident from how small the drivers are. There's just little to do. 3D cards are extremely complex, hence the massive amount of documentation. It isn't like there was just some magic number that needed releasing and the OSS drivers would be perfect with full support. There's now a ton of work to be done, since it sounds like it is just specs, not code, they are releasing.

    So you'll probably want to wait and watch until the driver is ready to go and up to whatever performance and stability standards you need for your application. Switch now and you are likely to find yourself in essentially the same situation as before: ATi's binary driver, or an OSS driver that doesn't do what you want.

    It'll be some time before this information can be transformed in to a fully functional, stable, fast driver. After all, if it were so easy, ATi and nVidia would have perfect drivers out on the launch of a new card and never need to do anything but minor updates.

    1. Re:Well hold on there by Anonymous Coward · · Score: 0

      Writing a driver for something like a network card or SCSI controller is fairly easy, and that's also evident from how small the drivers are.

      As someone who has both written drivers and ported between UNIXish kernels, allow me to say "Ahahaha". Look at the e1000 or tg3 NIC drivers for an example of how complex they can get.

      3D cards are extremely complex, hence the massive amount of documentation.

      This is true however. A video card driver can be just as complex, and that's without support for 3D.

    2. Re:Well hold on there by eggnoglatte · · Score: 1

      How right you are. And the 900+ pages are just the tip of the iceberg, since they do not seem to include the documentation for the 3D part. Expect the full documentation to be sevaral thousands of pages. By the time that is all implemented and stable enough for regular use, several new generations of GPUs will have come and gone.

      This is a great first step towards open source drivers, but there is a lot of catching up to do.

    3. Re:Well hold on there by Dr.+Spork · · Score: 2, Interesting
      I think this is a good point. After all haven't we always said that ATi drivers (for windows) suck? And those were written by professionals, the best that ATi could hire, and they certainly had access to all the specs. Now, I'm convinced that if all the relevant documentation is released, the open source drivers will be better than the professional ones, but that's because the OSS community will put countless hours of grueling work into the project. It's definitely worth it. Crappy graphics support has been a curse for Linux in many ways. We will all have a good belly laugh when Windows users look at Linux and envy us for our graphics drivers.

      (And seriously, won't there be a Windows version of the open source driver? And if so, might it not surpass AMD's own Windows driver? And might this not be a part of AMD's strategy, out of recognition that everyone disses their drivers and that their coders cost them too many salaries? Probably not, but this will certainly do a lot to make people finally reconsider the undying meme that ATi cards are better but their drivers suck.)

    4. Re:Well hold on there by A+beautiful+mind · · Score: 2, Insightful

      Yeah, there might be bugs. Who cares. That is the reason why we have the code, so that I can look at the source and fix stuff I want to get working.

      --
      It takes a man to suffer ignorance and smile
      Be yourself no matter what they say
    5. Re:Well hold on there by renoX · · Score: 1

      >They've released the specs, this doesn't mean anything yet.

      No, it means a lot: an opportunity is there, but it's true the driver will take some time.

      >People forget just how complex graphics cards are.

      Why are you generalising? I remember that the Free driver for the 7900 wasn't that great..
      Let's study the reasons:
      - people didn't really care about 3D as it was mostly useful for games only: this is not true anymore as X over OpenGL is the new fad.
      - specs were limited and available under a NDA only: the scope and quality of the specs remain to be seen but AMD has made the good decision of dropping the NDA, so there's hope.
      - drivers are complex: sure the 3D part is still complex (and even more now), but most people care about having a working 2D driver first during the installation, something which isn't always working now with Linux.

      In the operation 'we will write driver for you if you open the spec', there are 76 developers who registered, I don't know how many want/can work on the AMD 2D/3D drivers, but these kind of figures tend to make me optimistic..

    6. Re:Well hold on there by Anonymous Coward · · Score: 0

      Don't forget we already have existing open-source drivers. The current code can just be compared to the specs and corrected, which is much less work than writing new drivers from scratch, which seems to be what you're implying.

      Of course this is the tip of the iceberg, but we can get much more stable drivers pretty quickly. Full functionality and speed will come.

    7. Re:Well hold on there by Sancho · · Score: 1

      It also means that other companies might follow suit, if this helps ATI's bottom line. My next graphics card will be ATI-based for this very reason.

    8. Re:Well hold on there by TheLink · · Score: 1

      Yeah, given how sucky ATI has been in recent years - problems in both windows and linux, I can imagine some ATI boss telling the ATI developers, "OK, in 3 months time we'll find out whether you'll keep your jobs".

      So the closed source drivers might just start improving dramatically or be discontinued ;).

      It's a risk though - it may turn out that the ATI hardware actually really sucks, and the software guys just can't manage to reliably workaround all the bugs in the hardware.

      In which case ATI/AMD better be ready - with OSS software the source code is visible and that means the swearing and insults are too ;).

      --
  34. 900+ you say? by rev_sanchez · · Score: 1

    The Athlon 4800+ runs at 2.4GHz so this 900+ GPU thing must be really dragging ass.

    --
    If you didn't come to party don't bother knocking on my door. Prince '1999'
    1. Re:900+ you say? by LarsG · · Score: 1

      Yeah, compared to ooxml it is a slowtard.

      --
      If J.K.R wrote Windows: Puteulanus fenestra mortalis!
    2. Re:900+ you say? by Anonymous Coward · · Score: 0
      Good morning!

      Please patch yourself with

      + #include <coffee.h>
      and recompile yourself.
    3. Re:900+ you say? by Anonymous Coward · · Score: 0
      The Athlon 4800+ runs at 2.4GHz so this 900+ GPU thing must be really dragging ass.


      Whoosh! It's friggin' 900 PAGES of documentation. Doesn't have anything to do with performance of GPU.

    4. Re:900+ you say? by X0563511 · · Score: 1

      Yea, because a chip specifically designed to do calculations for graphics is identical to a chip designed to handle anything.

      Try something some time: run something in software OpenGL and watch how puny that 2.4+ ghz cpu seems.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    5. Re:900+ you say? by Anonymous Coward · · Score: 0

      Whoosh! indeed.

  35. obvious disconnect by Anonymous Coward · · Score: 0

    How would you even know they had good image quality if they had crappy drivers? Does not compute, unless you got lucky and bought the magical cards that ran without drivers. Or is that just some rumors you heard on the internets?

    1. Re:obvious disconnect by Anonymous Coward · · Score: 2, Funny

      How would you even know they had good image quality if they had crappy drivers? Does not compute, unless you got lucky and bought the magical cards that ran without drivers. Or is that just some rumors you heard on the internets?

      What are you, a retard? Display drivers have nothing to do with image quality -- that's the result of the hardware going out to the monitor plug. A cheap board will be fuzzier and its colors will be less crisp. All the ATI cards I've had gave me sharp, clear images with vibrant color and little bleeding.
        Even if drivers affected that (which they don't) how bad do you think the drivers are? Do you think they made me blind? Or are you some kind of fucked-up idiot who can't tell the difference between "crappy drivers" and "no drivers at all"?
        Crappy drivers may be slow, or they might have bugs, but they're drivers. ATI's hardware has always been good. Their drivers, OTOH, range all the way from "pretty good" to a "poke in the eye with a sharp stick."

    2. Re:obvious disconnect by Hal_Porter · · Score: 1

      Crappy drivers may be slow, or they might have bugs, but they're drivers. ATI's hardware has always been good. Their drivers, OTOH, range all the way from "pretty good" to a "poke in the eye with a sharp stick."

      Those pokes in the eye with sharp sticks are probably why you don't notice the image quality ;-)

      Or in your case ,-)

      --
      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;
  36. Two words. by kwabbles · · Score: 1

    Hell yes.

    --
    Just disrupt the deflector shield with a tachyon burst.
  37. now I can buy one by dltaylor · · Score: 1

    I used ATI products for years, starting with a 2MB Graphics Xpression that let me run at 1152x900 (used OpenLook, so my Linux desktop at home mirrored the Sun at work), but my last card buy had to be nVidia, although I despise them for what they did to 3dfx.

    I was just about to build a new system around the Asus M2N32 WS with yet another nVidia card, but not now. It will be an ATI again.

    Vote with your dollars AND ballots.

    1. Re:now I can buy one by Champion3 · · Score: 1

      I hate nVidia for the whole 3dfx thing too. I had just bought a new 3dfx card and was left dead in the water when nVidia acquired them for their IP and shut everything down. I've been buying ATI ever since, and in light of this announcement things are looking better.

      --
      I'm going to the casino. Don't gamble.
  38. Mobile and IGP GPUs? by Epsillon · · Score: 1

    With portables on the up, are we going to get specifications for the Radeon IGP chipsets? These seem to be causing the reverse engineering folks a huge headache and I've seen no mention of the Radeon Xpress 200M RS480 (yes, I have one, so I suppose I had better declare self-interest) or the Xpress 1100 so far. OK, we have 2D with the radeon(4x) driver in Xorg 7.2. 3D would be nice, though.

    --
    Resistance is futile. Reactance buggers it up.
  39. Lets hope so by Chuck+Chunder · · Score: 1

    I bought an Intel 965 based motherboard for my new computer because Intel have open source drivers but I've been disappointed at the lack of progress with them.

    Given that 3d multimedia desktops are the new sexy which all distros seem to be getting into I'm surprised that there doesn't seem to be much progress on getting it fixed on what is (as far as I am aware) the only open-driver supported 3d hardware available (at least until AMD release their 3d specs).

    --
    Boffoonery - downloadable Comedy Benefit for Bletchley Park
    1. Re:Lets hope so by Anonymous Coward · · Score: 0

      Intel just outsources their Xorg driver development and dumps the code in an open repository. While technically open source, it is not community controlled. At this point it might as well be a highly copyrighted driver that you just happen to have the source for.

      AMD's plan is to give out the specifications and a self sustaining community will be able to form around them. It will be more true to the open source as a development model.

    2. Re:Lets hope so by Corwn+of+Amber · · Score: 1

      Nothing whatsoever prevents me from pulling the code, bugfix it 100%, add a couple of features and upload it on sourceforge under the name "Open Intel GFX Driver for $OS". Laws prevent me? How? "cvs -d blahblah" ... no law at my door yet trying to stop me...

      --
      Making laws based on opinions that stem up from false informations leads to witch hunts.
  40. Thanks AMD! and good luck to the devs! by Cprossu · · Score: 1

    Thanks AMD, and to all of the devs, communities, and testers out there who are going to put their heart, soul, time and money into making a rock solid driver set! Make sure that these specs and the others that we all hope will be released are put to the best use possible.

    I look forward to any OS drivers that are produced and will at the very least try to put my time to use by finding and reporting bugs with the hardware I have when the time comes.

  41. 3 options. by khasim · · Score: 2, Insightful

    #1. Send it to ATI and let them use it in their marketing campaign.

    #2. Send it to nVidia to show the profit that is going to their competitor from a FORMER customer.

    #3. Send it to BOTH.

    1. Re:3 options. by sameerds · · Score: 1

      That "it" refers to the parent message, I suppose. Any plans of actually sending the replaced nVidia card back to nVidia? Or maybe AMD/ATI would take it on an exchange offer.

  42. Re:I'll replace my nVidia when I see a good review by MichailS · · Score: 1

    I came here intending to post just what you did. AGP and all. ^_^

  43. Re:Sweet! Sour!! by Just+Some+Guy · · Score: 5, Insightful

    And the newer drivers still support chips that predate your GF4.

    For the record, Nvidia says otherwise.

    You all should be grateful instead of pissing in their Cheerios.

    "Thank you, oh benevolent masters, for supplying the software required to use the hardware that you gave me in exchange for money." Was that suitably deferential, or should I bend my knee more?

    --
    Dewey, what part of this looks like authorities should be involved?
  44. Something along the lines of by z0M6 · · Score: 1

    1. Get documentation 2. Write driver 3. ??? 4. PROFIT!!!

    1. Re:Something along the lines of by Ruie · · Score: 2, Funny
      No, it like this:

      1. Get documentation
      2. Have fun
      3. Have more fun
      4. Have fun and profit !

  45. This is great! by i_love_unix · · Score: 1

    I once called myself a "pragmatist" when it came to Linux video drivers. If it works, I'll use it. Now that they are releasing an open source 3D driver that works, I will definitely use it :) Kudos to AMD/ATI for doing this! I've always been an NVidia fan, but will most certainly be buying an ATI video card for my next PC unless NVidia follows suit within the next 6-8 months.

  46. Feeding the Troll by Anonymous Coward · · Score: 0

    First I would like to say that the GP AC is not this AC, just for clarification purposes.

    Secondly will state as a matter of historical proof that absolutely no one can make a statement with 100% clarity as to their intent and meaning to anyone. This becomes even more difficult in a crowd of geniuses as they will analyze and come up with more potential meanings for each word, each clause, each statement, each body language, each typo, etc. That said and with the parent's "apparent definition of an idiot", some might say the parent qualifies under their own definition, but I am not here to call the parent an idiot and from the parent's original post in this string of comments it is obvious that the parent is not an idiot.

    IMO the GP noticed the parent was leaving a potential door open and made a statement in the form of a question for clarification and to close that door. Possibly the moderator made the same accessment and modded the GP up. If Google made the parent's statement publicly, Microsoft could have a field day with them either directly, through their shills, or via both paths.

    IMO the parent's insulting troll of the GP is unjustifiable and they should consider some confidence building moves so they won't take insult so readily from others who may well not intend the perceived insult. No one who can post at Slashdot would fit the medical definition of an idiot and frankly no one should apply it to anyone who does not meet the medical definition.

    Disclaimer: Posted by a nerd labeled as an "idiot" far too many times by subgroups of the larger groups made up of jocks and rednecks, some of whom are family. Ever "married" two bales of hay in a barn? Neither had I when we moved to my mother's part of the world after moving from base to base with my father. Some of my relatives, and others, refer to people like me as "edumucated idjits". The mispronounciation they use is deliberate and not the way they normally say either word. If they hear a word from someone and perceive the word to be too "fancy" they think your conceited for using it if they understand it and consider you a "know it all" if they don't as well as taking insult from it. Communication is an area frought with pitfalls.

    1. Re:Feeding the Troll by MrNaz · · Score: 1

      I agree with everything you say. I was in a bad mood when I replied to the other guy. However, I would like to point out that the word "idiot" along with other insults are very often not literal references to the technical definitions of those words, rather a general invocation of disparaging language. I will demonstrate this concept by telling you that I think you are a poopy head.

      --
      I hate printers.
  47. Actually by CrazedWalrus · · Score: 4, Insightful

    Heh - I think you're missing the implication of his statement.

    Because Linux is less resource intensive, he's able to upgrade his distro several times on the same hardware, putting himself in the situation of having a new kernel with old hardware and old drivers that don't load in the new kernel.

    If you want to upgrade Windows, you usually wind up needing a new machine, so: new machine, new video card, new drivers, new Windows -- not a problem. Well, at least not the same problem.

    So it's not an issue of what's *wrong* with Linux, it's what's *right* with it. The problem is that this presents circumstances the hardware world isn't used to dealing with.

    1. Re:Actually by Kalriath · · Score: 2, Insightful

      If you want to upgrade Windows, you usually wind up needing a new machine, so: new machine, new video card, new drivers, new Windows -- not a problem. Well, at least not the same problem. Bullshit. I haven't had to upgrade my PC through the last three releases and it works fine. Hell, I have Vista running on a laptop with only a 1.4GHz processor and it runs fine. That's with Aero enabled (my understanding is that this would be roughly equivalent to Ubuntu with Beryl and Compiz correct? and have I spelt that right?)

      I think the main problem with Linux is that if a vendor releases a closed source driver (which some are obliged to do by legal agreements for other companies technologies) that the next minor Kernel release can break the driver. In the Windows world, it takes a major release to break drivers (i.e. Windows 4/Windows 98 to Windows 5/Windows 2000 or Windows 5.1/Windows XP to Windows 6/Windows Vista)

      --
      For a site about things like basic rights, Slashdot users sure do like to censor "dissent".
    2. Re:Actually by reaktor · · Score: 2, Interesting

      That's why Ubuntu and others do not upgrade the kernel with every new release. Only new releases of Ubuntu have a new kernel. [Ubuntu does put in patches for security, of course.] And Ubuntu makes their own package for Nvidia and ATI drivers, so it-will-work with their kernel. No need to go to www.amd.com for drivers, etc.

    3. Re:Actually by mcrbids · · Score: 2, Insightful

      Because Linux is less resource intensive, he's able to upgrade his distro several times on the same hardware, putting himself in the situation of having a new kernel with old hardware and old drivers that don't load in the new kernel.

      Hear here! I have an ANCIENT AMD K6-2/450 doing backups. It has 2.5 TB of hard disks in it, and its only purpose in life is to copy files over the network every day. It's 10 years old, and has been in continuous 24x7 duty all along. Rock stable, too - why change it when it works fine and performance isn't an issue?

      I haven't had to upgrade my PC through the last three releases and it works fine. Hell, I have Vista running on a laptop with only a 1.4GHz processor and it runs fine.

      You sir, are an amazingly patient person. I credit you for your lack of desire to do meaningful work. Wally would be proud!

      --
      I have no problem with your religion until you decide it's reason to deprive others of the truth.
    4. Re:Actually by Anonymous Coward · · Score: 1, Interesting

      Right,
      We will all take your word for this. 1 1.4GHz processor.
      My laptop for work, a dual core at 2GHz with 2 Gb crawls. It's a pristine Vista install.

      Please, my old single core laptop with 1/2 as much ram but still at 2GHz with Linux runs incredibly fast compared to it.

      Yes! I'm happy stupid close source drivers break. Break break break, I hope they break them even more often. This will teach hardware companies something that they should have learned a long time ago but somehow missed out. They're hardware companies! They gain nothing by screwing with software.
      Oh really? Service packs for windows don't break driver support? You're clearly living in an alternate reality. I've even seen nightly updates break patches, which is a lot more fun with the close-source nature of the whole thing. At least with Linux I can fix the bugs in a few hours generally.

      Thank god I only have to see windows at work and this will all be over in a few months. Never again.

    5. Re:Actually by pato101 · · Score: 1

      I think the main problem with Linux is that if a vendor releases a closed source driver (which some are obliged to do by legal agreements for other companies technologies) that the next minor Kernel release can break the driver. In the Windows world, it takes a major release to break drivers (i.e. Windows 4/Windows 98 to Windows 5/Windows 2000 or Windows 5.1/Windows XP to Windows 6/Windows Vista)
      You are wrong. The Closed-source nvidia binary comes with the open-source kernel module which you can compile with your kernel headers so it adapts to your current status. Thus, the same nvidia-driver can be installed among all the ubuntus (for instance) and first ubuntu release (Warty?) is about the same time XP appeared while current ubuntu (Feisty) was released about the same time than Vista. Now try to install XP drivers on Vista...

      Goto nvidia downloads and you'll see that there is only one driver for *all* the linux distributions/flavours/whatever.

      Of course, deep changes in kernel break this model and if you conserve the current driver and try to install in future releases perhaps it does no longer work and that's what people are complaining about, in the sense that your card can get unsupported in the future by the next driver releases. However the approach is elastic enough so the linux driver spanlife seems better than the Windows one.

    6. Re:Actually by marcosdumay · · Score: 1

      "Bullshit. I haven't had to upgrade my PC through the last three releases ... laptop with only a 1.4GHz processor..."

      I wonder where did you get that 1.4GHz processor at 98/99?

    7. Re:Actually by level_headed_midwest · · Score: 1

      Just giving the CPU speed is useless. A 1.4 GHz P4 is much different from an Athlon 1400 or a 1.4 PIII-S, a 1.4 Pentium M, and a 1.4 GHz Core or Core 2 Duo (there's no 1.4 C2D, but you see my point.) My laptop runs a 1.06 GHz chip- a brand-new Core 2 Duo U7500. It has a low clock speed but it works just fine with modern software as it is a modern piece of hardware. I am guessing the OP has a chip that's a tad newer than he's leading us to believe and is doing so for effect.

      --
      Just "gittin-r-done," day after day.
    8. Re:Actually by Anonymous Coward · · Score: 0

      Maybe you should read more closely. He didn't say that his computer that hasn't needed an upgrade since three windows generations is his notebook. And, seeing how it is unlikely that it is, it's not.

    9. Re:Actually by Phroggy · · Score: 1

      If you want to upgrade Windows, you usually wind up needing a new machine, so: new machine, new video card, new drivers, new Windows -- not a problem. Well, at least not the same problem. Thanks for clarifying that; I was gonna have to correct you. NVidia's Vista drivers are certainly not problem-free. The latest problem I'm having is that if I have two displays in clone mode (same image on two displays, a projector and a CRT), it will consistently reset to single display mode (so the second display no longer works) every time the energy saver kicks in. I've worked around this for the moment by disabling the energy saver feature, and just have to hit the power button on the CRT when I'm done using it.
      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
    10. Re:Actually by Kalriath · · Score: 1

      You're correct. The laptop has only had one generation of Windows: Vista was pre-installed. I have had to upgrade the RAM from 512MB to 1GB, but since then it runs the same as the XP I still run on my desktop (which, incidentally, is a 3.2GHz P4 800MHz FSB Prescott with 1GB RAM) because - well, for a start XP isn't that bad, and second because Nvidia are bloody incompetent at making Vista drivers.

      --
      For a site about things like basic rights, Slashdot users sure do like to censor "dissent".
    11. Re:Actually by Kalriath · · Score: 1

      Personally, I have no problems with Vista (apart from bloody annoying features, but everyone has that). It runs fine on my, yes, 1.4GHz processor with 1GB RAM (laptop, AMD processor).

      It's my belief that you are talking shit, and at least I'm willing to sign my name to my statement.

      --
      For a site about things like basic rights, Slashdot users sure do like to censor "dissent".
    12. Re:Actually by Kalriath · · Score: 1

      No, it's a 1.4GHz. As I said though, it is a notebook and as such the number doesn't mean exactly the same thing - you are right there. It's a Sempron Mobile 1700+ I believe, which clocks in (according to the BIOS and Windows) at 1.4GHz. If you must look at the specs, it's an Acer 3054WXMI (though I've upped the RAM to 1GB).

      --
      For a site about things like basic rights, Slashdot users sure do like to censor "dissent".
    13. Re:Actually by marcosdumay · · Score: 1

      There were simply no PC microprocessors running at 1.4GHz by 98/99.

      The GP is exagerating things a bit on its post.

  48. huh by goarilla · · Score: 1

    well well i never tought i would live to see this day
    hell actually froze over today

  49. Re:FUCKING by Anonymous Coward · · Score: 0

    Needs much moar. I suggest jihad against the front page till Taco blocks teh 20721 carpfolder.

  50. May be best for HTPC by Dr_Barnowl · · Score: 3, Interesting

    ATI have historically always had excellent features on their cards for supporting media playback. The downside was that accessing them in Linux has always been much harder than using the equivalent features on nvidia hardware.

    If these specs allow a good stable XVMC driver to be written for ATI hardware, ATI could become the top choice for Linux media centre boxes.

    1. Re:May be best for HTPC by TheRaven64 · · Score: 1

      They seem to provide settings for video overlay. I don't think they are enough for a full XvMC implementation (I didn't see anything related to iDCT etc), but they are a good start.

      --
      I am TheRaven on Soylent News
  51. bittorrent by Anonymous Coward · · Score: 0

    here is a torrent for the files: http://thepiratebay.org/tor/3804798

    1. Re:bittorrent by Rod+Beauvex · · Score: 1

      Are you sure it's ok to use TPB for non-illegal purpouses? :p

  52. Yes, but... by gillbates · · Score: 1

    For those of us who have an ATI X1250, that's not much good.

    I'm reading these, stuck wondering to which products they actually apply (as in, Marketing Names(TM)). It would be nice if they mentioned the products by name. Also, there's no high level overview, so you just have to guess how the registers work.

    Hopefully someone can provide a link to more documentation, because this is just trickle compared to what the OEMs must get from ATI. Normally, you'd get a datasheet, a register reference, a technical reference manual, and programmer's guide. ATI provides one of these, and people are ecstatic? This is hardly open.

    --
    The society for a thought-free internet welcomes you.
    1. Re:Yes, but... by ThisNukes4u · · Score: 4, Informative

      Specs for r500(the X1k series) are supposedly in the pipleline. For matching codenames->marketing names, I recommend http://en.wikipedia.org/wiki/Comparison_of_ATI_Graphics_Processing_Units And the reason why they are so "deficient" is they have a team of a few engineers helping the X.org folks write the drivers. They have also said that they will be providing code snippets in the future to help clear up unclear parts of the spec. This is just a teaser release, not all we're getting.

      --
      thisnukes4u.net
  53. Quick, everybody look away, its a trick by bug1 · · Score: 1

    1. It takes years for a corporation to do something related to a press release.
    2. Press release was only released weeks ago.
    3. FAKE

    Im going to stay in JAVA (sum microsystems) world where corporate decisions happen at a snails pace and impeded the companies operations for years (case in point open sourcing of java)

    Sarcasm Inside

    1. Re:Quick, everybody look away, its a trick by Anonymous Coward · · Score: 0

      That's ridiculous.

      The Java spec has been open from day 1, virtual machine, language and library.

      What took a long time what open-sourcing the code, which I assume was full of 3rd party code (hell, they're not even done removing it yet). I
      Where's that link to ATI's Windows driver source code, again?

    2. Re:Quick, everybody look away, its a trick by bug1 · · Score: 1

      Java spec has been open as in look but dont touch, remember how microsoft got their arses kicked for modifying the spec... but anyway, its pretty hard to have a closed spec for a languange and expect people to adopt it, they didnt have a choice.

      Sun did have a choice of open sourcing their java _code_ and rumours went around for about five years before they actually did anything.

  54. I'm not in the market now...but in a year or so... by HiThere · · Score: 2, Interesting

    The last time I bought a computer, I went with AMD because I was mad at Intel. I still am, though less so. It fades over time.

    Last year I did an evaluation, and Intel came out on top ... solely because they had an open source video driver. This will soon eliminate that benefit.

    N.B.: For me to choose Intel it must be 5% better than the competition. This is due to various corporate actions that I dislike. (Two years ago it was 10%...I use a time decaying function.) If they were up against a competitor that didn't support DRM, they'd need to be 50% better, but I don't see one, so that part of the playing field is level.

    --

    I think we've pushed this "anyone can grow up to be president" thing too far.
  55. My next cards will be ATI by YA_Python_dev · · Score: 2, Interesting

    I have been buying nVidia video cards for forever for their... let's say less bad support for GNU/Linux and I recomended them to my Windows-using friends. But this changes everything: from now on I know that I can buy ATI cards and be sure that they'll work and have good software support. Thanks AMD, thanks ATI: you have made a new loyal customer today.

    --
    There's a hidden treasure in Python 3.x: __prepare__()
    1. Re:My next cards will be ATI by geekoid · · Score: 1

      "I can buy ATI cards and be sure that they'll work and have good software support. "

      How do you know that? You can postulate that will happen, but video card drivers are hard and someone with the know how, time, and rive has to write them, and keep them updated. Just because OS provides the potential, doesn't mean it will happen. Hell, the world could use good OS Bank teller software as well, but that isn't really happening, is it?
      Don't get me wrong, I am sure there will be a dozen drivers in a week, but would you trust putting them on someone elses computer that will cause you to be called every time it crashes?

      Personally, I don't care of the God of coding video drivers appears on earth and personally write the OS drivers. ATI lied to me, and cost a friend his company because of that lie.
      ATI can burn in hell and I'll bring marshmallows.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    2. Re:My next cards will be ATI by B5_geek · · Score: 1

      My thoughts exactly.

      Another future customer AMD, good job.

      --
      "The price good men pay for indifference to public affairs is to be ruled by evil men." ~Plato (427-347 BC)
    3. Re:My next cards will be ATI by Anonymous Coward · · Score: 0

      Personally, I don't care of the God of coding video drivers appears on earth and personally write the OS drivers. ATI lied to me, and cost a friend his company because of that lie. ATI can burn in hell and I'll bring marshmallows.

      Details please.

    4. Re:My next cards will be ATI by Anonymous Coward · · Score: 0

      Ditto, you can't just give vague details like that and not give us the whole sordid story!

    5. Re:My next cards will be ATI by Elf-friend · · Score: 2, Informative

      ...video card drivers are hard and someone with the know how, time, and rive has to write them, and keep them updated. Just because OS provides the potential, doesn't mean it will happen.

      As I understand it, the X.org DRI (Direct Rendering Infrastructure) project are doing just that. In fact, they have been doing just that for ATI cards (among others) for some time, but progress has been slow, because they've needed to reverse-engineer everything (they previously could get specs with an NDA, but not since the 9200 cards). As a result, the DRI Radeon drivers currently only work with older cards (up through X850, IIRC), and provide little or no 3D acceleration for all but the oldest cards. Even so, the general consensus has been that their drivers are superior to the proprietary, FireGL drivers that ATI provide for Linux (and they work on *BSD, which the FireGL drivers do not[1]). With this information, they should be able to make steady progress on providing support for recent ATI cards under X.

      [1] Not natively, at least. Last winter, I believe, they were finally made to work under FreeBSD with Linux binary compatibility.
  56. High end is good but... by ceeam · · Score: 2, Insightful

    ... what about all those X1100 and similar in all these notebooks? And pardon my ignorance - are there already good, reliable, full-featured drivers for the likes of 9800 etc?

    1. Re:High end is good but... by Lehk228 · · Score: 1

      they are probably starting with the current high end so they can get some really impressive looking results to push development for the rest of the specs when they release them

      --
      Snowden and Manning are heroes.
  57. Re:I'll replace my nVidia when I see a good review by Crizp · · Score: 1

    Not many but at least for last-gen cards you can get a decent one. As an x1650 Pro AGP owner I'm happy to hear these news :)

    The tags to this story are awesome also.

  58. Re:I'll replace my nVidia when I see a good review by Anonymous Coward · · Score: 0

    Actually, mostly targetting the gamer market, there are still newish AGP cards out there - basically, PCI-express gfx chips with a PCIe to AGP8x bridge chip. Works quite well. I've got one myself (because I delayed my usual upgrade cycle due to AMD's Barcelona delay (I want one, best memory bandwidth...), but wanted to play with OpenGL 2.1 features).

  59. Teaser indeed... by gillbates · · Score: 2, Interesting

    The thing that gets me is that Wikipedia seems to know more about the ATI chips than ATI. Of course this can't actually be the case, but I think it is somewhat telling that ATI is not the authoritative reference for even their own hardware. There seems to be some uneasiness with releasing the full product specs, which suggests to me that they don't have a real committment to openness.

    Well, if I can't get specs, my next video card will be an nVidia. Why should I suffer because my HW vendor wants to hide something from me? Do they really believe that non-functional hardware gains them any marketshare?

    With Windows hopelessly insecure, my only real option is to either buy a Mac, which is too expensive for my taste, or to use Linux. Which means that if ATI doesn't provide the documentation that I - or somebody - needs to write open drivers, I'm just not going to buy their HW. Period. That super-secret, proprietary graphics pipeline won't sell ATI cards if no one can use it. Do they really think that I'm going to run Windows just to get video to work?

    --
    The society for a thought-free internet welcomes you.
    1. Re:Teaser indeed... by ThisNukes4u · · Score: 1

      I don't know what you're getting at, but AMD/ATi has committed to releasing the specs that will allow open drivers to be written. This was just the initial release of 2d specs for a limited range of cards. More information is coming.

      --
      thisnukes4u.net
    2. Re:Teaser indeed... by fabs64 · · Score: 4, Insightful

      Well, if I can't get specs, my next video card will be an nVidia. Why should I suffer because my HW vendor wants to hide something from me? Do they really believe that non-functional hardware gains them any marketshare?

      Now how does this make a lick of sense? nVidia haven't released ANY specs.
      Also, I'd imagine that 2d rendering is reasonably similar across chipsets, but I admit I'm just guessing there.

  60. It's about damn time! by Anonymous Coward · · Score: 0

    Seriously, what the hell took them so long? Still, I've got to give them credit for doing the right thing - even if it took them a while to do.

  61. Holy cow by Anonymous Coward · · Score: 0

    If it's for real, I guess I'm going to switch over to being an ATI customer.

  62. 900 pages? by dpilot · · Score: 1

    > Come off it... that's not even enough for an Office document standard.
    >
    > Worthless!

    You said it. It's ONLY 900 pages, nowhere near the OOXML 6000+. They didn't even include documentation for every legacy chip, much less the unannounced next-generation ones that they've no doubt got ready for first tape-out.

    Plus there's no Swahili translation, either.

    It's too little, too late, may as well stick with the old closed-source stuff that "works."

    --
    The living have better things to do than to continue hating the dead.
  63. Re:I'll replace my nVidia when I see a good review by red_dragon · · Score: 4, Funny

    And the Windows Vista driver sucks, somewhat hilariously.

    It wasn't too long ago that I was at Microsoft's Philadelphia offices for an Exchange 2007 presentation. The first thing that they wanted to show was a short video on a projection screen -- what they actually showed the audience was a Vista laptop with ATI graphics choking half way through a two-minute video and then puking an error message saying that the video driver crashed and was being restarted. And some guy behind me said "Twelve years later and they still can't get the presentation right."

    --
    In Soviet Russia, Jesus asks: "What Would You Do?"
  64. Re:I'll replace my nVidia when I see a good review by ross.w · · Score: 2, Informative

    They won't work in your SLI rig. Only NVidia cards support SLI. ATI have a different system that uses a different motherboard, so you'd have to replace that too, if you want SLI type performance.

    --
    If my call is important, why am I talking to a recording?
  65. print baby print!! by downix · · Score: 1

    My printer is working overtime tonight....

    --
    Karma Whoring for Fun and Profit.
  66. fantastic news by voidy · · Score: 1

    I have always given kudos to nvidia for producing a very good driver for linux [good as in, performance and stability wise], but this really changes everything. i don't even know anything about ati cards, ive not really needed to in the last 10 years since starting to use linux full time, but if ati can give me more bang for my buck, then they will definitely be on the cards next time i buy a card. i bought a 7950gt last month, so it should be a while, and by then. the drivers produced may be good enough by then :]

    --
    I do not fear computers. I fear the lack of them. Isaac Asimov
  67. Great-I got were I am by stepping on the masses. by Anonymous Coward · · Score: 0

    "It's about fucking time that companies realize the trickle-down effect of abusing nerds."

    Pfft! It's not abusing nerds that got ATI were it is presently. It's good old fashion competition, business fuck-ups, and a deep-pocket buyer.

    "Who do the ignorant masses go to when they need advice? Their nerdy friend.."

    No, they go to you to be insulted. They go to wherever advertising tells them to go.

    "ATI lost market share for almost the exact reason that IE did (albeit to less extent)."

    More like the reason 3DFX did.

  68. Re: closed nVidia support actually bad by Psykechan · · Score: 3, Insightful

    The issue with a closed driver for the nVidia cards that actually performs somewhat well is actually a detriment for the community at large. It causes some people who would be interested in making a better open driver to just suck it up and use the existing closed driver because it's easier even though it has many problems.

    I am very thankful that AMD has released specs. Until nVidia follows suit there should be no real reason to buy nVidia cards. This means that they will be forced to eventually release specs and those of us who had no support from nVidia will finally get a working driver.

    As an nVidia customer, all I can say is Thank You AMD!

  69. Yay for Tibbles! by NotZed · · Score: 1

    Was a bit sceptical when he went to work for ATI having been a long-time 'open saucer' but its good to see some good came of it in the end ...

    Now if only Ubuntu on my thinkpad had 3d drivers which worked ...

    --
    _ // `Thinking is an exercise to which all too few brains
    \\/ are accustomed' - First Lensman
  70. A little confused. by sc0ob5 · · Score: 1

    So what are the cards that will have open source 2D and 3D acceleration? I have an x1950pro and I am sick of crappy linux drivers. Never the less a good start by AMD to release some of their specs but they should release everything.

  71. Suddenly... by Repossessed · · Score: 1

    I find myself in the market for a new graphics card, what's ATI got in the low range x16 slot type that's one of these R500 or R600 cards?

    --
    Liberte, Egalite, Fraternite (TM)
  72. Re:I'll replace my nVidia when I see a good review by microbee · · Score: 2, Insightful

    Well, there is a real market.

    Since DELL has been shipping pre-installed Linux PCs, they will eventually favor ATI if it performs better than nVidia due to higher quality drivers.

  73. Good to hear by LinuxWhore · · Score: 1

    I just got home today to find that my Nvidia card on my MythTV system has 3 blown caps (probably through no fault of NVidia). I'll probably be picking up an ATI card to replace it. It's well worth it to have the piece of mind that my card will be supported in the future.

    --

    I am MuchTall
  74. Re:Sweet! Sour!! by anagama · · Score: 2, Funny

    "Thank you, oh benevolent masters, for supplying the software required to use the hardware that you gave me in exchange for money."

    I wish I had mod points -- the question would be funny or insightful though -- it's both.

    With respect to your previous comment, I upgraded my system last weekend and I didn't really get $500 worth of improvement. My old motherboard was on the verge of fritzing though so it had to be done (Athlon XP 2200+ system), and even though the various parts are maybe 4 years old, nothing fits in the new motherboards anymore. Thank goodness I had an old PCI IDE card in the closet -- none of my HDs are SATA.

    I'm proabably one of the few people who went out specifically looking for a GMA950 motherboard -- I was impressed with how well Fusion ran on my macbook with the open source drivers compared to how it worked on my Desktop/nVidia system. I still am impressed with that on my new Desktop, I'm just dissapointed in the ridiculously long BIOS startup time which negates the quicker boot time (from the grub prompt). If I could have found the GMA950 on an add-in card, I would have bought that and stuck with AMD processors. At least I'll still get to help out AMD and buy a video card once a bit of driver work gets done.

    As for the old board, which is basically a complete system sans drive (well, I have a couple unused 40gb drives in the closet), powersupply (one of those in the closet too, though I can't remember if it works), and case (in the closet, no powersupply, small case requires specially sized PS), I decided to try my hand at replacing capacitors -- 3 had leaked and one was bulging. If I get lucky, I'll have that old machine back for other uses.
    --
    What changed under Obama? Nothing Good
  75. I wish they had done this last year ... by Dragon+Bait · · Score: 1

    At work I bought over 50 desktops during the course of last year, all with nVidia cards because they were supposed to support Linux better. Sigh. Well, at least I know for my next purchases.

  76. I want Theo's opinion by turing_m · · Score: 1

    I'd love to get a comment from Theo de Raadt on the subject. This (with 3D drivers) is what he appears to be after, it would be good to hear confirmation from him. Maybe he's holding out the praise until they release the 3D specs.

    --
    If I have seen further it is by stealing the Intellectual Property of giants.
    1. Re:I want Theo's opinion by TheRaven64 · · Score: 1

      I doubt Theo is much interested in the 3D drivers, since OpenBSD doesn't have a DRI implementation yet. I only skimmed the 2D specs, but they seemed to be to the level of detail required for a driver, so it sounds like they're exactly what Theo wants.

      --
      I am TheRaven on Soylent News
  77. Too bad you can't share the specs. by jbn-o · · Score: 1

    Speaking of redistribution rights, consider the following quote from page two of the two released docs:

    No license, whether express, implied, arising by estoppel, or otherwise, to any intellectual property rights are granted by this publication.

    If we assume that the propaganda term "intellectual property" is to include copyright, it would seem that distributing copies of these specifications non-commercially and verbatim is disallowed. One wonders what else you're not allowed to do with these specs, make derivative works perhaps? After all, "no license" is granted.

    1. Re:Too bad you can't share the specs. by marcosdumay · · Score: 1

      That is the problem with the "intelectual property" term. Stopel normaly refers to patents, and such a note on a specification does normaly also refer to patents.

      But IANAL.

  78. Re:I'll replace my nVidia when I see a good review by Anonymous Coward · · Score: 0

    If (IF) the driver works with the latest ATI card, then I will dump the Nvidia card (NVIDIA TOO), and go ATI. From now till then, Nvidia is still the best bet. When you prove otherwise (and remember I'm hopeful that you do), then I'll switch. Right now Nvidia is the best solution, but its not perfect. Their driver releases tick at 6 month intervals, and every so often changes in the kernel wreck the driver. Some argue "so use an older kernel", but I'd rather have a very very small binary (if I had to have one at all), and have the rest as software that changes with the kernel. Better than that would be a full in-kernel driver. I would dump Nvidia in a heartbeat.

  79. Xorg pwnd by Slashdot by Anonymous Coward · · Score: 0

    SLASHDOTTED!!!

    LMAO

    can't access the website

  80. As a FreeBSD/amd64 user... by Anonymous Coward · · Score: 0

    ...all I can say is: Thank you AMD!

    Fuck NVIDIA in the ass!!!!!

    Glass

    1. Re:As a FreeBSD/amd64 user... by Anonymous Coward · · Score: 0

      I agree! ...But to both parties involved and screw nVidia and it's nice to see things like this, especially for real FOSS such as OpenBSD...

  81. Re:Sweet! Sour!! by Anonymous Coward · · Score: 0

    And the newer drivers still support chips that predate your GF4.
    For the record, Nvidia says otherwise.

    Reading comprehension much? From your link:
    Below are the legacy GPUs that are no longer supported in the unified driver. These GPUs will continue to be maintained through the special legacy NVIDIA GPU driver releases.
  82. Re:Sweet! Sour!! by Jthon · · Score: 1

    Or you might notice that they plan to support those with the legacy driver which can be upgraded to compile against newer kernels. They just aren't adding features or fixing bugs for those older cards.

  83. But it can finally be done by TeXMaster · · Score: 2, Insightful
    Of course it's going to take time, but the crucial point here is that it can be done with officially released specs, without having to waste time reverse engineering and wild-guessing how things should be done.

    Consider this: I'm actually surprised how far nouveau development already went, without any specs and starting from the obfuscated nv driver. How much further could they be now if they had the specs and didn't have to waste uncountable hours tracing register changes and second-guessing their use?

    --
    "I'm never quite so stupid as when I'm being smart" (Linus van Pelt)
  84. Error. by rew · · Score: 1

    There is an error on page 175. It says: "others - reserved" in the if (D1GRPH_DEPTH = 0x2) section. All options are defined, so there are no others that can be reserved.

  85. Patent Infringement on the horizon? by Brit_in_the_USA · · Score: 1

    I have often heard in open vs. closed graphics driver binaries the reasons that manufactures keep the source closed is that it contains a lot of proprietary technology and a lot of licenced / patented technology.

    Now I realise that ATI/AMB have only released specs, but at some point in an open source driver development will someone "re-invent the wheel" and come up with a technique that is already patented.

    What will happen in that scenario? Will open source drivers always be held back in this respect?

    Do you think AMD wiould apply for permission to use the tehcnoologies for the open source drivers to continue support for the open source driver effort?

    Also the opposite scenario, what happens if an open source coder comes up with a valuable new technique for graphics drivers and graphics display that no one has patented before? Having an open source driver faster / more useful than a closed source driver will defintly be an interesting scenario as to use it the the AMD's and nVidia's would need to open up everything correct?, which they are supposed unable to do...

    What are your thoughts on this?

    1. Re:Patent Infringement on the horizon? by Anonymous Coward · · Score: 0

      He might find a job then.

    2. Re:Patent Infringement on the horizon? by Anonymous Coward · · Score: 0

      I think the reason why it took them this long to release the specs is that they were negotiating with the companies they licesned the patents from.

  86. Open source Windows drivers? by Andreaskem · · Score: 1

    Those specs could probably be used to build open source Windows drivers, too. There should be a few very interesting possibilities for Windows users as well if someone creates such a driver, I think.

  87. Hardware Video Acceleration by Anonymous Coward · · Score: 0

    Does anyone know if there will be any specification for this as well ?

    Related :

    http://en.wikipedia.org/wiki/AVIVO
    http://en.wikipedia.org/wiki/Unified_Video_Decoder

    Thank you.

  88. Re:I'll replace my nVidia when I see a good review by Nurgled · · Score: 2, Interesting

    Amusingly, when I bought my Ubuntu PC from Dell's UK site a few weeks back the graphics card section had a giant ATi banner above it but only offered an NVidia card as an option. I assume that this is because right now NVidia's linux drivers are better, though neither are open source. Hopefully this'll change soon.

    (Interestingly, the system shipped without NVidia's drivers installed, so I had to explicitly install NVidia's driver using the Restricted Driver Manager. I suppose you could argue that NVidia's driver has no business on a system being sold as an "Open Source" computer, but this is an annoying extra barrier for the potential non-technical user.)

  89. So "not supporting" is right by Anonymous Coward · · Score: 0

    If they aren't bugfixing, in what way are NVidia supporting the older cards?

    NVidia drivers are blobs. They may or may not work with any future kernel and the developers of the kernel don't know when this will happen because the NVidia driver is a blob. They may be making assumptions about processor registers that won't work in the future. They may be assuming certain calls that won't exist. NVidia drivers even had a remote expoloit. If another is found, you aren't supported.

    "Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.

    It's been 1 hour 6 minutes since you last successfully posted a comment"

  90. free3d.org by Anonymous Coward · · Score: 0

    This means http://free3d.org/ will have updates soon :-)

    Also, I've been running an ATI RV380 [Radeon X600] and a Radeon 9250 for a while now. They're not crazy gamer cards, but they fit my needs. As long as I can play Torcs, use Blender, and edit/watch some movies I'm happy. I've had Torcs crash because of unimplemented features in the r300 driver, but other than that everything has been very stable.

  91. Can desktop linux ever match windows in gui speed? by walterbyrd · · Score: 1

    Debian is my primary desktop. I use, and advocate, linux all the time. But, I must admit, windows seems to have a crisper, snappier, gui. I have dual booted several PCs, with several different versions of linux, and several different WM/DEs, but windows is always noticably faster.

    I think part of it may be linux, but I'm guessing that it mostly because of X11.

  92. AMD 8.41.7 Display Driver Also Released by Boinger69 · · Score: 1

    For those of us who really dont care about binary blob vs FOSS when you just want drivers that work:
    http://ati.amd.com/support/drivers/linux/linux-radeonhd.html

    The release notes say it is not intended for x1900 or lower, but be sure I'll be trying it on my T41p anyway.

    Compiz Fusion awaits!

    1. Re:AMD 8.41.7 Display Driver Also Released by Boinger69 · · Score: 1

      And compiz fusion waits for 8.42 since this release still lacks AIGLX and now FireGL support too.

      D'oh!

  93. conspiracy theory xxx by __aalwyc6372 · · Score: 1

    i think it's a semi-smart move. they've lost market share to nvidia and probably won't be able to keep up with them technology wise. so it's their way to get some free look at some genuine open source drivers. using the community to get back into business. i think it's a little similar to what IBM does: being kicked in the ass by it's "little child" microsoft, they now try to get back at them through a movement that can't be stopped or buyed or whatever. they're all just using the open source community to their ends in my opinion.

    what they don't seem to realize, is, that in the end they all won't matter. on the other hand, for the next few years it may work well for them and who cares, what's happening to their "old company", when they are already in pension, but that's another story...

    1. Re:conspiracy theory xxx by Elf-friend · · Score: 1

      Personally, I think it's more a matter of AMD realizing that a large percentage of desktop *nix users have have traditionally preferred AMD to Intel, and not wanting to damage the relationship will those users by being seen as uncooperative (in fact, I've gotten the impression that a lot of *nix users have been expecting nothing less than this of AMD ever since the merger was announced). By making ATI cards functional on *nix, the position of AMD in the chipset market is also improved. Heretofore, the choice between CrossFire and SLI support on motherboards has been a no-brainer - CrossFire cards not only aren't functional under X, but prevent X from loading unless they are physically removed from the slot. This has meant *nix users weren't choosing CrossFire boards, which represents nearly all AMD-chipset boards. If CrossFire becomes usable, however, AMD's chipsets will benefit. All-AMD systems become viable, which can only be good for AMD. Furthermore, the OSS drivers represent a huge coup over nVIDIA's binary-only releases.

  94. No they haven't by Andy+Dodd · · Score: 3, Insightful

    "Expect more documentation (and 3D specifications) to arrive shortly."

    In short, we have 2D documentation but no 3D documentation. It's been this way for years, nothing is different.

    The last time someone (Matrox) said "3D specifications to arrive shortly", a whole bunch of suckers (including myself) bought cards and got shafted because the promised specifications were never released. My G200 was replaced by a Riva TNT2 within six months and I haven't left NVidia since then.

    Others promise open specifications and fail to release them fully, resulting in cards that are paperweights.

    NVidia doesn't promise open specifications, but at least they deliver solid drivers that work (and work well).

    --
    retrorocket.o not found, launch anyway?
  95. Re: closed nVidia support actually bad by Andy+Dodd · · Score: 1

    "Until nVidia follows suit there should be no real reason to buy nVidia cards."

    Yes there is still real reason to buy NVidia cards. They have working functionality, ATI has only promised (but not yet delivered) 3D documentation. The last company to do that (Matrox back in 1998) never delivered. I'll believe it when I see it.

    --
    retrorocket.o not found, launch anyway?
  96. Re:9000 karma by CodeShark · · Score: 1

    And how dear sir, do you calculate that? To my knowledge /. doesn't publish the karma numbers, and if it is that high and you've tracked it their, you have way way way too much time on your hands.

    --
    ...Open Source isn't the only answer -- but it's almost always a better value than the alternatives...
  97. Re: closed nVidia support actually bad by pak9rabid · · Score: 1

    "Until nVidia follows suit there should be no real reason to buy nVidia cards"

    I can think of a reason...huge performance gains perhaps?

    But seriously, if...i mean when the open source community produces a fast, quality driver for ATI cards that exploits the GPU's full potential, I will definately be reconsidering ATI as a viable solution. Until then, it's nVidia.

  98. Read the distribution license by Anonymous Coward · · Score: 0

    In cases it cannot be done. NVidia did, for a long time, FORBID the inclusion of the download driver binaries and installer. You HAD to get it from nvidia.com.

    Under GPL, you cannot add this restriction.

    "Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.

    It's been 53 minutes since you last successfully posted a comment"

  99. Re:I'll replace my nVidia when I see a good review by Sylvak · · Score: 1

    I think he was referring to the fact that some governments are pushing open source, which means that when the time comes for them to make the decisions of which video card to use in their corporate desktop standard, this move could influence their decisions. When you think about how much hardware governments buy, you can see how this could have a positive impact for ATI/AMD. Plus, 3D is not so important for most government workers, so pushing out a 2D open source right away seems like a good move.

  100. The real problem. by Outland+Traveller · · Score: 1

    This appears to be the biggest obstacle to open source drivers. The current US patent laws have created an environment where any sufficiently complex, or even non-complex piece of software can be construed to violate someone else's bogus software patent.

    There's a lot of risk attached to sharing source code. It allows hostile entities a free opportunity to pursue their bogus software patent litigation.

    On the positive side, we're starting to see more ways to mitigate this risk, such as creative licensing (GPL3), creative code ownership arrangements, and not but not least pressure to reform the US patent system.