Slashdot Mirror


Hands On With Nvidia's New GTX 280 Card

notdagreatbrain writes "Maximum PC magazine has early benchmarks on Nvidia's newest GPU architecture — the GTX 200 series. Benchmarks on the smokin' fast processor reveal a graphics card that can finally tame Crysis at 1900x1200. 'The GTX 280 delivered real-world benchmark numbers nearly 50 percent faster than a single GeForce 9800 GTX running on Windows XP, and it was 23 percent faster than that card running on Vista. In fact, it looks as though a single GTX 280 will be comparable to — and in some cases beat — two 9800 GTX cards running in SLI, a fact that explains why Nvidia expects the 9800 GX2 to fade from the scene rather quickly.'"

212 comments

  1. Yeah but... by Anonymous Coward · · Score: 2, Funny

    Can it play Duke Nukem forever?

    1. Re:Yeah but... by Anonymous Coward · · Score: 2, Insightful

      Can Linux run it?

      Same answer as all cool new hardware: NO!

    2. Re:Yeah but... by the_humeister · · Score: 0

      If it has branch instructions, perhaps it could.

    3. Re:Yeah but... by Devin+Jeanpierre · · Score: 2, Insightful

      Same answer as all cool new hardware: NO!
      Easy counter-example would be any new CPU architecture, which is generally adopted by Linux faster than the competition (especially Windows, which is probably what you're comparing Linux to, given the context). AMD64 (and Itanium 2, for that matter) is an example. While Linux can be slow to get support for some things, that's certainly not true for all cool new hardware. What about the PS3? Pandora? Heck, some cool hardware Linux supports would be impossible for a great deal of other OSes, especially Windows (especially if I mean Anything/Linux, and not GNU/Linux).
      --
      -Devin Jeanpierre
    4. Re:Yeah but... by masterzora · · Score: 2, Informative

      Actually, NVidia's pretty good about getting Linux drivers for new cards out relatively quickly.

      --
      Remember, open source is free as in speech, not free as in bear.
    5. Re:Yeah but... by taupter · · Score: 1

      YMMV. I personally still feel Nvidia drivers aren't even on par with the Intel driver.
      Unaccelerated console framebuffer if you use the proprietary drivers, memory leaks, lockups, crashes and overall instability compared to the free nv driver, problems with notebook screens, shader bugs and bad performance on newer GPUs all of these made me unwilling to spend more money on Nvidia hardware. Maybe it's time for Intel (even if so much slower) or AMD/ATi (with good closed-source drivers, and flourishing new open source drivers coming along with the specs).
      Nvidia is doing something to support Linux, and we need to praise them for it, but right now it's injustifiable to buy Nvidia to put in a Linux machine. Sadly. We need better drivers, framebuffer acceleration, kernel-based resolution mode change, less memleaks, and, I hope someday, free-as-in-speech 3D drivers, or at least the GUPs' specs so we can implement it ourselves.
      What bothers me is the fact Nvidia didn't set things straight, it wasn't the leader in Linux support, and it's not even being the follower, as some things we take for granted from other manufacturers don't come from Nvidia.

    6. Re:Yeah but... by masterzora · · Score: 1

      tl;dr version: ATI & Intel have better drivers, but NVidia does better in making sure all of the latest products are supported, reinforcing my point while not harming the parent's.

      I'll grant you that they're buggy and not on par with the Intel drivers, but you'll also note that wasn't my point. The real point is that the drivers exist and that they do work. I'm running 3D just fine in Linux right now on my NVidia card.

      In fact, having recently switched from an ATI to an NVidia on my box (under AMD64 Lenny, for the record), I was happy to find that my NVidia card was actually _supported_. My old ATI (an x1550. Not a powerhouse by any means, but the GPU was where I skimped when I put this box together and ran short on cash) wasn't supported by any of the 3D drivers. My more recently acquired NVidia (a 7900 GT, handed down to me by a friend who upgraded to the 9800 gx2) card is supported by the NVidia drivers. I'm sure if the ATI card was supported, it would be much better, but unfortunately for me, this wasn't the case.

      --
      Remember, open source is free as in speech, not free as in bear.
  2. The this incredible!! by Zosden · · Score: 1, Funny

    Before the only computer that could do this at even medium graphics was Link

  3. Anandtech and TechReport reviews by Anonymous Coward · · Score: 1, Informative
    1. Re:Anandtech and TechReport reviews by TheLinuxSRC · · Score: 3, Informative

      The 9800GX2 may be cheaper but it most certainly is not faster, even considering your links. From Anandtech, the charts show a significant speed increase with the new hardware.

      In fact, from the article:
      The GTX 280 delivered real-world benchmark numbers nearly 50 percent faster than a single GeForce 9800 GTX running on Windows XP, and it was 23-percent faster than that card running on Vista. In fact, it looks as though a single GTX 280 will be comparable to--and in some cases beat--two 9800 GTX cards running in SLI, a fact that explains why Nvidia expects the 9800 GX2 to fade from the scene rather quickly.

      Which leads me to the question, are you trolling?

    2. Re:Anandtech and TechReport reviews by Anonymous Coward · · Score: 0

      Conclusion: You can copy and paste! Congratulations, way to not provide any of your own information.

    3. Re:Anandtech and TechReport reviews by Gnavpot · · Score: 0

      From your quote:
      "will be comparable to--and in some cases beat--"


      Is that what you call "a significant speed increase with the new hardware" (comparing a GTX 280 to a 9800 GX2)?
    4. Re:Anandtech and TechReport reviews by BronsCon · · Score: 2, Informative

      Actually, they're comparing the GTX 280 to TWO 9800s in SLI configuration. RTFS FFS

      --
      APK quotes people (including myself) without context and should not be trusted. Just thought you should know.
    5. Re:Anandtech and TechReport reviews by blueZ3 · · Score: 1

      I think you missed the last half, which says that a SINGLE 280 is comparable to and in some cases beat TWO 9800s

      --
      Interested in a Flash-based MAME front end? Visit mame.danzbb.com
    6. Re:Anandtech and TechReport reviews by mordenkhai · · Score: 1

      The chart you linked to is 3DMark, in which the new GTX 280 does in fact top the list. However if you slide a few pages to the games section its a very different picture. http://techreport.com/articles.x/14934/10 Call of Duty, which is topped by the 9800GX2
      http://techreport.com/articles.x/14934/11 Half Life 2, is alot closer and lets face it both cards should be rolling through HL2 w/o any issues, however the GX2 is again on the top, by a small to tiny margin
      http://techreport.com/articles.x/14934/12 ET:QW finally a real split, the GX2 tops out the 1920x1200 resolution and below, but the GTX280 takes the crown come high end 2560x1600
      http://techreport.com/articles.x/14934/13This Chart shows Crysis and Assasins Creed info, both of which the 9800GX2 tops the average with a 10-20% margin.
      So it isnt as clear cut as you make it sound at all. Everyone has their own opinion on which is more important, for me, I'll take game performance over numbers crunching, though I do use 3DMark as well.

    7. Re:Anandtech and TechReport reviews by Kamidari · · Score: 4, Informative

      Yes, they're comparing it to two 9800GTXs, which is what a 9800GX2 is: Two 9800GTXs on one board. RTFS indeed. It seems like a case of give and take. The GTX280 is more expensive, but is a single GPU solution, which tends to be more stable. The 9800GX2 is cheaper, runs about as fast, but is a dual GPU unit, so you might have a few more "issues" to deal with in your game playing adventures.

    8. Re:Anandtech and TechReport reviews by rgviza · · Score: 1

      Right, that's why at this link:
      (Posted 05/1/08 at 04:20:57PM | by Michael Brown)

      http://www.maximumpc.com/article/gigabyte_geforce_9800_x2

      They show Crysis getting 41.4FPS on a Gigabyte GeForce 9800 X2

      However from the linked article in the op (which is the same freaking magazine):
      Crysis GeForce 9800 GTX: 11.7 FPS, 2x Geforce 9800 GTX: 12.8?

      WTF? They think we were born yesterday? They either lied for the first article or they are lying now.

      I smell payola and Maximum PC is not to be trusted.

      They are either:
      A. falsely claiming that the 9800 GTX is the fastest current nvidia processor (what about the GX2?), or
      B. Comparing the GTX280 to a lemon.

      This benchmark/comparison is worth almost as much as that Brooklyn Bridge deed I have in my wallet... Wow I just sneezed and it sounded a little like "HBULLSHIT!"

      This article is suspect at best. I don't own a 9800GTX (I have a GTS 640 from last year) but if I did I'd take this article with a grain of salt the size of a standard rubik's cube.

      -Viz

      --
      Don't kid yourself. It's the size of the regexp AND how you use it that counts.
    9. Re:Anandtech and TechReport reviews by MrDiablerie · · Score: 1

      Read the articles over again. GTX 280 blows away the 9800 series in performance.

    10. Re:Anandtech and TechReport reviews by Anonymous Coward · · Score: 0

      Shhh, don't you realize all the work this guy had to do in convincing his wife they needed that 9800 SLI watercooled system they just bought instead of a house payment???

      It's already obsolete and the UPS truck probably hasn't even arrived yet.

    11. Re:Anandtech and TechReport reviews by Anonymous Coward · · Score: 0

      Right, that's why at this link:
      (Posted 05/1/08 at 04:20:57PM | by Michael Brown)
      They show Crysis getting 41.4FPS on a Gigabyte GeForce 9800 X2 We've given up benchmarking Crysis at 1920x1200, and no card has yet delivered what we'd call acceptable performance even with the 1280x720, 2x AA settings we've been usingâ"until now: The 9800 GX2 pumped out 41.4 frames per second.

      However from the linked article in the op (which is the same freaking magazine):
      Crysis GeForce 9800 GTX: 11.7 FPS, 2x Geforce 9800 GTX: 12.8? We're especially pleased with the performance delta we observed with Crysis: Even with the resolution at 1920x1200, 4x antialiasing enabled, and all the game's other quality settings on high, our engineering sample delivered the game at more than 30 frames per second running DirectX 9. Games still run slower on Vista, however; Crysis, for example, shed about eight frames per second running DirectX 10, but it was still twice as fast as a single 9800 GTX. And remember, we tested an engineering sample running on pre-release drivers.
    12. Re:Anandtech and TechReport reviews by masterzora · · Score: 1

      I think you missed what a 9800 GTX2 is (ie, 2 9800s).

      --
      Remember, open source is free as in speech, not free as in bear.
    13. Re:Anandtech and TechReport reviews by GeffDE · · Score: 1

      A very good conclusion figuring that the 9800GX2 is a mature product while the GT200 was an engineering sample and running on pre-beta drivers. ...The sarcasm...it burns my eyes...

      --
      It has been a nervous year, with people beginning to feel like Christian Scientists with appendicitis.
    14. Re:Anandtech and TechReport reviews by AmiMoJo · · Score: 1

      Looks like they are finally listening re power consumption: http://anandtech.com/video/showdoc.aspx?i=3334&p=9

      Still some way to go to catch up with ATI though. Unfortunately all these benchmarks are largely academic until we have matching 4850/4870 ones.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    15. Re:Anandtech and TechReport reviews by PIBM · · Score: 1

      It's also very depending on the resolution you are using. If you are playing on a 30" monitor, the 1GB of ram (for 1 gpu) will make a very big difference in itself in 2560x1600 @ 4x. In some cases the 9800GX2 (also 1gb, but 512mb effective by GPU) can't even run the game.

    16. Re:Anandtech and TechReport reviews by Nicolay77 · · Score: 2, Informative

      Those game benchmarks are probably CPU bound.

      So an increase in CPU speed would probably benefit more the GTX 280 than the 9800GX2.

      Of course, we can't know if my theory is true unless we test it, but seems logical to me.

      --
      We are Turing O-Machines. The Oracle is out there.
    17. Re:Anandtech and TechReport reviews by MooseMuffin · · Score: 1

      Are you blind? Your link shows the gx2 beating the 280 in nearly every actual game.

    18. Re:Anandtech and TechReport reviews by Gnavpot · · Score: 1

      RTFS FFS

      RAUTFS FFS!

      Someone claimed that a 9800 GX2 (which is identical to two 9800GTX in SLI) is a better purchase.

      Some other one responded that the GTX has a significant speed increase over the old card and then went on citing this:
      In fact, it looks as though a single GTX 280 will be comparable to--and in some cases beat--two 9800 GTX cards running in SLI, a fact that explains why Nvidia expects the 9800 GX2 to fade from the scene rather quickly.

      So in short, the summary says that a GTX 280 will be comparable to - and in some cases beat - a 9800 GX2. Would you call that a significant speed increase?

      And for pointing that out I got modded down. The stupidity on Slashdot has grown to new heights.

    19. Re:Anandtech and TechReport reviews by Anonymous Coward · · Score: 0

      so you might have a few more "issues" to deal with in your game playing adventures. WTF there FUDster? Why would a word like issues need to be in quotes? If there are "issues" with the GX2 cards post some forum links or something. Don't go making accusations with without something to back them up.

    20. Re:Anandtech and TechReport reviews by BronsCon · · Score: 1

      That really depends which cases you see the GTX 280 performing better in. If they're all cases in which the 9800 GX2 is running at its peak and the GTX 280 is, perhaps, not, I'd be led to believe that we haven't seen the full potential of the 280.

      I'm not a gamer (in fact, I make very light use of 3D acceleration) and I know that any increase in speed is significant.

      At any rate, the article claims something different than what you seem to claim. It compares the GTX 280 to two 9800GTX's in SLI, which would perform differently than the 9800 GX2.

      Again, not a gamer. I guess that means I have more free time to actually learn about how the hardware works and educate others via Slashdot.

      Now, I'll sit back and let the grammar nazis point out the error in the first line, which I'm going to be too lazy to go back and correct.

      --
      APK quotes people (including myself) without context and should not be trusted. Just thought you should know.
    21. Re:Anandtech and TechReport reviews by Anonymous Coward · · Score: 0

      "Yes, they're comparing it to two 9800GTXs, which is what a 9800GX2 is: Two 9800GTXs on one board."

      Umm, no its not. It's two _lower clocked_ 9800GTXs on the one board. 2x 9800GTX is faster than a 9800GX2. This is due to thermal considerations on the GX2.

    22. Re:Anandtech and TechReport reviews by Anonymous Coward · · Score: 0

      9800 GX2 is two 8600's put together, it does not have G92 processors. 9800 GTX's in SLi out perform a GX2, and three 9800 GTX's in tri-SLi out perform two GX2's in quad SLi. That being said, I tink I'm going to step up to the 260 from my 9800 GTX.

    23. Re:Anandtech and TechReport reviews by Guspaz · · Score: 1

      I think you're confused about a great many things. The 9800 GX2 (not GTX2) is equivalent to two 9800s, not two 9800 GTX.

      Since the GTX 280 is consistently slower than a single 9800 GX2, and since two 9800 GTX are faster than a single 9800 GX2, it's quite obvious that the GTX 280 is slower than two 9800 GTX.

      In fact, Anandtech's results show that the GTX 280 is slower than two 8800 GTs in most tests. Considering that you can buy two 8800GTs for a bit more than half the projected cost of the GTX 280, that really doesn't bear well for it.

  4. Power vs Intel by SolidAltar · · Score: 4, Interesting
    Let me say I do not know anything about chip design but I have a question -

    How is Nvidia able to year after year make these amazing advances in power while Intel makes (although great) only modest advances?

    As I said I do not know anything about chip design so please correct me on any points.

    1. Re:Power vs Intel by the_humeister · · Score: 5, Informative

      Because graphics operations are embarrassingly parallel whereas regular programs arn't.

    2. Re:Power vs Intel by Deltaspectre · · Score: 1, Insightful

      It's not exactly raw power, I seem to remember an article explaining the differences between the 7x00 gen and the 8x00 gen, where they fundamentally changed how the graphics card dealed with "stuff", can't remember what magazine it was in though.

      I don't think intel has quite that flexibility because of their commitment to backwards compatability, while nVidia can just push out new drivers.

      --
      My UID is prime... is yours?
    3. Re:Power vs Intel by gnick · · Score: 1

      How is Nvidia able to year after year make these amazing advances in power while Intel makes (although great) only modest advances? There is more room for improvement in the graphics card/GPU arena than in the CPU arena. Since the market is so much larger surrounding CPUs, more research has been done and the chips are closer to "perfectly" using available technology and continually expanding the realm of what technology is available.

      And I'll echo the_humeister's statement that graphics operations are much more easily done in parallel than generic computing. You can throw processors/cores at the problem pretty easily and continue to see improvement.
      --
      He's getting rather old, but he's a good mouse.
    4. Re:Power vs Intel by corsec67 · · Score: 2, Informative

      I think one huge thing is that graphics is a hugely parallelizable task. The operations aren't very complex, so they can just keep cramming more and more processing units onto the chip.

      Intel and AMD are having issues getting over 4 cores per die right now, while this card "... packs 240 tiny processing cores into this space, plus 32 raster-operation processors".

      --
      If I have nothing to hide, don't search me
    5. Re:Power vs Intel by cliffski · · Score: 4, Interesting

      As a game dev, and from what I see, I'm assuming its a stability thing.
      Intel's chips have to WORK. and I mean WORK ALL THE TIME. getting a single calculation wrong is mega mega hell. remember the pentium calculation bug?
      People will calculate invoices and bank statements with that intel chip. It might control airplanes or god knows what. It needs to be foolproof and highly reliable.

      Graphics chips draw pretty pictures on the screen.

      It's a different ballgame. As a game dev, my 100% priority for any new chips is that they ship them with stable, tested drivers that are backwards compatible, not just great with directX 10 and 11.
      If someone wrote code that adhered correctly to the directx spec on version 5 or even 2, the new cards should render that code faithfully. Generally, they don't, and we have to explain to gamers why their spangly new video card is actually part of the problem in some situations :(

      --
      DRM-free indie games for the PC and Mac: Positech Games
    6. Re:Power vs Intel by mrchaotica · · Score: 3, Interesting

      People will calculate invoices and bank statements with that intel chip. It might control airplanes or god knows what. It needs to be foolproof and highly reliable.

      Graphics chips draw pretty pictures on the screen.

      Nvidia is increasingly marketing its chips as "stream processors," rather than "graphics processors." They are becoming increasingly used for scientific computation, where reliability and accuracy are just as important as in the general-purpose case (which reminds me, I need to check if they support double-precision and IEEE 754 yet). It could be the case in a few years that the structural analysis for the building you'll be in might be done by a program running on one of these chips.

      --

      "[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz

    7. Re:Power vs Intel by AmiMoJo · · Score: 2, Informative

      Also, graphics processors are evolving quickly, where as CPUs have had basically the same instruction set for 30 years now.

      For example, with the 8000 series pixel shaders had become very important in modern games, so the cards were optimised for pixel shading performance much more than the 7000 series was. There is simply no equivalent for CPUs - even stuff like SSE extensions is really just trying to do the same stuff in a more parallel way, it isn't a radically new way of doing things.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    8. Re:Power vs Intel by Lord+Ender · · Score: 1, Troll

      Why is parallelism embarrassing?

      --
      A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
    9. Re:Power vs Intel by mikael · · Score: 1

      A GPU is a large number of high-speed floating point units (stream processors), parallelised memory (video ram) and some clever memory caching (texture units). In earlier GPU's, integer operations were just a special case of floating point calculations. The later GPU's have seperate logic pathways for integer calculations

      A CPU consists of a single pipelined processor with all sorts of tricks to optimise performance. The Intel P6 article at wikipedia gives some explanation of these:

      * Speculative execution and out-of-order completion (called "dynamic execution" by Intel), which required new retire units in the execution core. This lessened pipeline stalls, and in part enabled greater speed-scaling of the Pentium Pro and successive generations of CPUs.
              * Superpipelining, which increased from Pentium's 5-stage pipeline to 14 of the Pentium Pro, and eventually morphed into the 10-stage pipeline of the Pentium III, and the 12- to 14-stage pipeline of the Pentium M.
              * Integrated L2 cache that runs at the full speed of the processing core, instead of the earlier designs of off-die (on motherboard) cache, which runs at a fraction of the CPU frequency.
              * Wider 36-bit physical address bus to support more than 4 GiB of physical memory (the linear address space of a process was still limited to 4 GiB).
              * Register renaming, which enabled more efficient execution of multiple instructions in the pipeline.


      Even if you were to run a simple hello world on a CPU, this would invoke the use of virtual memory to implement dynamic library linking, cache pages to be loaded for the executable, and all sorts of API calls to get through the windowing system and kernel IO library. A GPU just has to do some vector maths on data read from one block of memory to another.

      --
      Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
    10. Re:Power vs Intel by p0tat03 · · Score: 3, Informative

      Precisely. This is something that can be solved by simply throwing more transistors in. Their biggest challenge is probably power and heat, not architecture.

      Not to mention that "programs" on GPUs are ridiculously simple compared to something on a general purpose CPU. Next time you write a shader, try branching (i.e. if, else), your shader will slow to a relative crawl.

    11. Re:Power vs Intel by __aabvlw4075 · · Score: 1

      I don't think the question was about graphics processors vs. general purpose processors. I assumed it was about Nvidia graphics processors vs. Intel graphics processors.

    12. Re:Power vs Intel by Sycraft-fu · · Score: 1

      The new nVidia chips (the 280 and 260) do support double precision. I'd have to guess this is mainly targeted at general purpose computation as I don't see a whole lot of application for video (32-bit floating point per colour channel is plenty for giving smooth output in 8 and 10 bit spaces).

    13. Re:Power vs Intel by AcidPenguin9873 · · Score: 1

      Right on. The CPU makers do have two ways to "throw more transistors in" which can help: more cache, and more cores. But there are problems with both of these techniques.

      The problem with more cache is that it only scales so well in terms of performance, power, and area. With a larger cache, you get:

      • larger die size, which means fewer die per wafer which increases manufacturing cost
      • any particular die is now more likely it is to have a defect, lowering yields and thus further increasing cost
      • leakage power increases, so now your 64MB of cache burns 2-3W just for being there
      • bigger cache == longer latency to access it, so you get diminishing returns and eventually your cache will start to perform more like a mildly faster DRAM

      Finally, more cache doesn't give you a linear speedup (meaning, doubling the cache doesn't give you 2x the performance), because it only helps the latency of loads and stores, it doesn't help you get through massive datasets like truly parallelizing your application and running it on more cores does.

      Now what about more cores? More cores are similar in technique to "more pixel shaders" or "more vertex shaders" because they are directly helping with the problem that your CPUs are working on. But like Parent and Grandparent posts said, most times, the problems CPUs are working on are not easily parallelizable. Not only is it hard to write parallel code, dealing with concurrency and locking properly, but with lots of parallel problems there exists only so much inherent parallelism in the problem or algorithm itself. Even if you parallelized everything that could be parallelized, then the serial parts of your problem will start to dominate your computation time. See Amdahl's Law.

    14. Re:Power vs Intel by Anonymous Coward · · Score: 1, Interesting

      These latest ones, the ones the article is about, support double-precision. Not so clear if they support full IEEE 754: When I messed with scientific computation on GPUs last, the single most major problem wasn't the precision (there are techniques for making the most of single precision after all), it was the fact they silently carried on through fp exception conditions that in a scientific context mean "you have screwed up. Stop now.". They would mean a barely-noticeable graphical glitch at most in mere graphical output applications, even in game-physics the error wouldn't matter, maybe a crate would bounce imperceptibly wrongly or slightly intersect a wall or whatever.

      But it matters a lot in grownup science land. :-)

  5. Power Consumption by squoozer · · Score: 4, Interesting

    Something that has always concerned me (more as I play games less often now) is how much power these cards draw when they aren't pumping out a zillion triangles a second playing DNF.

    Most of the time (90%+ probably) I'm just doing very simple desktop type things. While it's obvious from the heat output that these cards aren't running flat out when redrawing a desktop surely they must be using significatnly more power than a simple graphics card that could perform the same role. Does anyone have any figures showing how much power is being wasted?

    Perhaps we should have two graphics cards in the the system now - one that just does desktop type things and one for when real power is required. I would have thought it would be fairly simple to design a motherboard such that it had an internal only slot to accept the latest and greatest 3D accelerator card that suplimented an on board dumb-as-a-brick graphics card.

    --
    I used to have a better sig but it broke.
    1. Re:Power Consumption by SolidAltar · · Score: 1

      From the article:

      "However, Nvidia claims idle power draw for the GT200 of only 25W, down from 64W in the G80."

    2. Re:Power Consumption by SolidAltar · · Score: 4, Informative
      More detail (sorry):

      You may be wondering, with a chip this large, about power consumptionâ"as in: Will the lights flicker when I fire up Call of Duty 4? The chip's max thermal design power, or TDP, is 236W, which is considerable. However, Nvidia claims idle power draw for the GT200 of only 25W, down from 64W in the G80. They even say GT200's idle power draw is similar to AMD's righteously frugal RV670 GPU. We shall see about that, but how did they accomplish such a thing? GeForce GPUs have many clock domains, as evidenced by the fact that the GPU core and shader clock speeds diverge. Tamasi said Nvidia implemented dynamic power and frequency scaling throughout the chip, with multiple units able to scale independently. He characterized G80 as an "on or off" affair, whereas GT200's power use scales more linearly with demand. Even in a 3D game or application, he hinted, the GT200 might use much less power than its TDP maximum. Much like a CPU, GT200 has multiple power states with algorithmic determination of the proper state, and those P-states include a new, presumably relatively low-power state for video decoding and playback. Also, GT200-based cards will be compatible with Nvidia's HybridPower scheme, so they can be deactivated entirely in favor of a chipset-based GPU when they're not needed.

    3. Re:Power Consumption by CastrTroy · · Score: 4, Informative

      This is how graphics cards used to work. You would plug a VGA cable from your standard 2D graphics card to your, for example, Voodoo II card, and the Voodoo II card would go out to the monitor. You could just have the 3D card working in passthrough mode when not doing 3D stuff. Something like this could work on a single board though. There's no reason you couldn't power down entire sections of the graphics card that you aren't using. Most video cards support changing the clock speed on the card. I'm wondering if this is a problem at all, with any real effects, or whether it's just speculation based on the poster assuming what might happen. Anybody have any real numbers for wattage drained based on idle/full workload for these large cards?

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    4. Re:Power Consumption by Barny · · Score: 1

      This is why NV came up with their new trick, build an integrated video adapter into all boards and let the high end cards use the pci-e 2.0 bus to move the framebuffer over to that when playing games, then when just doing normal windows tasks use the SM bus to turn off these electric heaters.

      Works in vista only though, and of course, that OS is still showing signs of flop in the games area, despite DX10 and SP1.

      --
      ...
      /me sighs
    5. Re:Power Consumption by Xelios · · Score: 1

      ATI's last generation of cards had a feature called PowerPlay, which gears the card down when its not being heavily used. The 4800 series will have the same feature and judging from TFA Nvidia's doing something similar with the GT200.

      --
      Murphey's fighting Occam, and we're in the stands.
    6. Re:Power Consumption by wagnerrp · · Score: 1

      My old 6800GS already had this feature. It idles at 350MHz on the desktop, but ramps up to 485MHz when you open a 3D application.

    7. Re:Power Consumption by kipman725 · · Score: 1

      what like the orignal Voodoo accelerator boards? (completly amazing bits of hardware btw)... it resutlts in problems with programs that only expects one graphics card and don't allow selecting which one to do the acceleration on.

    8. Re:Power Consumption by eharvill · · Score: 1

      Perhaps we should have two graphics cards in the the system now - one that just does desktop type things and one for when real power is required. I would have thought it would be fairly simple to design a motherboard such that it had an internal only slot to accept the latest and greatest 3D accelerator card that suplimented an on board dumb-as-a-brick graphics card.

      Heck, I wouldn't mind having a "turbo" button, ala the XTs and 286s to handle what you just described...
      --
      At night I drink myself to sleep and pretend I don't care that you're not here with me
    9. Re:Power Consumption by willy_me · · Score: 1

      But all modern operating systems support 3D accelerated displays. MacOS has their Quartz Extreme, Windows has Aero (I think), and even Linux has one (although the name escapes me.) Your solution would have worked well 5 years ago but times have changed. The 3D component is now always on.

    10. Re:Power Consumption by CastrTroy · · Score: 2, Informative

      However, not all that power is needed for running the 3D desktops. I can run Compiz (linux 3D desktop) on my Intel GMA 950 without a single slowdown. With all the standard 3D eyecandy turned on. So you wouldn't need to run an nVidia 8800 at full clock speed to render the desktop effects. Also, Windows Vista and Linux both support turning off the 3D effects and running in full 2D mode. I'm sure Mac OS supports the same, although I've never looked into it, so it's hard to say for sure. Especially since MacOS has such a limited number of computers that it is supported on.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    11. Re:Power Consumption by Anonymous Coward · · Score: 0

      Nvidia thought of this already, they came out with something called "Hybrid Power" which is supported on newer motherboards. It does exactly what you have stated above. In some cases, it can even use BOTH the on-board chip and the discrete card for high performance. Too bad you didn't patent your idea earlier :)

    12. Re:Power Consumption by Anonymous Coward · · Score: 1, Funny

      Sony does this with a lot of their VAIO laptops. But you have to install 17 rootkits for it to work.

    13. Re:Power Consumption by bakedpatato · · Score: 1

      Nvidia and ATi have created something that (kinda) fits your description: Hybrid SLi and Hybrid Graphics, respectively. Unfortunately, only the low end cards(8400GS/8500GT and the 2400/3400 series for Nvidia and ATi)are supported as the "discrete" part of the mix.

    14. Re:Power Consumption by willy_me · · Score: 1

      I agree, my point was that it is better to have a 3D accelerator that can scale down (as the one in the article) then to completely bypass the 3D accelerator (as the Voodoo II did.) This is because most desktops will have the 3D accelerator activated at all times regardless of whether or not it is being used extensively.

    15. Re:Power Consumption by drinkypoo · · Score: 1

      Even current nVidia cards have power-saving. My QuadroFX1500 has PowerMizer, which will drop the GPU clock down as far as 100MHz (memory clock too) even while pushing polygons. For instance I've run an xscreensaver (blocktube with all options on, low speed though) in xwinwrap on my 1680x1050 desktop while xscreensaver-demo ran the same saver again in a small window and the GPU was toggling between the low and medium speeds. The new ones (as per sibling comment by SolidAltar) do much much more, but even the current GPUs significantly cut their power consumption. I can REALLY tell the immediate difference because with a Quadro in a laptop you get to FEEL it. Not to mention this thing has like umpteen fans and I monitor eight thermal sensors... including the hard disk and the GPU.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    16. Re:Power Consumption by anss123 · · Score: 1

      Quartz Extreme is off by default up to 10.4 (don't know about 10.5), so MacOSX can run without it.

    17. Re:Power Consumption by Toll_Free · · Score: 1

      You mean like the ORIGINAL 3D card, the Voodoo?

      Still have mine :) And the S3 Virge I used I used as my 2D card.

      --Toll_Free

    18. Re:Power Consumption by willy_me · · Score: 2, Interesting

      Ok, I just went over your original posting again and wanted to address the issues you spoke of.

      There's no reason you couldn't power down entire sections of the graphics card that you aren't using.

      There is a price to pay for powering up/down sections of a chip. It takes time and power. One needs to limit the number of times this is done. But the basic idea is good and I believe this is exactly what the card in question does. However, it is not as simple as implied.

      Most video cards support changing the clock speed on the card.

      This is a good solution because it does not incur a time or power penalty unlike the previous suggestion. However, it still results is sections of the chip being left on when not in use. In addition, with leakage current being such a big deal in modern transistors, you really can't clock it down too much. One risks loosing state in the dynamic elements of the chip. (At least this is how I understand it - but it is not my field, others here would know for sure.)

      I am willing to bet that the chip in question utilizes both of the techniques you suggested. So they were good suggestions. However, I still don't believe that the 3D accelerator would ever be "turned off" completely on a modern desktop computer. So implementing such techniques is essential as one can not rely on a Voodoo II type solution.

    19. Re:Power Consumption by wolf12886 · · Score: 1

      You'd think that would always be the case, yet my 8800GTS consistently idles at 57 degrees C with the fan at 100%, while my overclocked q6600 idles at 37, god knows what the 8800 is drawing to do that...

    20. Re:Power Consumption by miknix · · Score: 1

      Something that has always concerned me (more as I play games less often now) is how much power these cards draw when they aren't pumping out a zillion triangles a second playing DNF. When I'm working on simple desktop things (even using beryl), my Geforce 8400 Linux driver puts the graphic card in 2D mode. The driver also has a "Powermizer" feature which dynamically changes GPU and MEM frequencies on the fly. This is all activated by default.

      I even wrote a desktop "gadget" for showing me the nVidia GPU frequency and temperature on the fly on Linux.

      Have a look at DigitalMon on
      http://adesklets.sourceforge.net/desklets.html

      Screenshot is here:
      http://adesklets.sourceforge.net/images/DigitalMon_thumb.png
    21. Re:Power Consumption by apoc.famine · · Score: 1

      As I went from a 300 Watt power supply to a 450, to a 500 to a 650, this started to worry me to. Slap in a few drives, two DVD/CD burners, bigass video card, more cores, faster bus speeds, and the wattage adds up fast.

      After a several years of this, I took the $300 I managed to get back from the IRS and bought an Asus EEE PC. It does enough basic desktop/internet stuff decently well enough that I rarely fire up the massive desktop in the other room EXCEPT when I need that sort of horsepower.

      Home from work, sipping a gin and tonic, and posting to slashdot - no need to fire up a desktop for that. SSHing into my server to add a theme to the PHPBB3 forum we're setting up - no need for a desktop to do that. Editing css themes for pages there? Again, a basic linux desktop and shell (Xubuntu, if you're wondering) works fine. Sure, the keyboard is tiny, but I don't really care if my wpm drops to about 20 while posting to slashdot, or editing css files on a remote server. This isn't work - it's leisure time. And I don't schedule things by the minute, or even by the hour.

      My desktop get very little use now. And really, my recliner is more comfortable than the office chair is, so I don't miss it all that much. And this is even more energy efficient than powering down the video card. I think the EEE-PC draws LESS power than this video card does on STANDBY. Of course that's less than just the video card - one can't ignore the additional overhead from the PSU inefficiency, drives kept spinning, processor cycles, fans, etc.

      --
      Velociraptor = Distiraptor / Timeraptor
    22. Re:Power Consumption by EvilIdler · · Score: 1

      That's still a lot (although better than the 150W idle I've read about with some cards). It draws more power idle, the card alone, than some small computers :)

    23. Re:Power Consumption by EvilIdler · · Score: 1

      Mac Mini is based on these bog-slow Intel integrated chipsets, and it's nice and fast at its desktop effects. Haven't found a way to switch them off, either, but I never had the need.

    24. Re:Power Consumption by GleeBot · · Score: 1

      AMD/ATI's latest and greatest is well-known for having particularly good idle power consumption. You might want to check into those numbers.

      Incidentally, the idea you suggested--having a low-power and high-power part--has already been used for many years in both general purpose and graphics processors. Besides being able to clock up/down, individual functional units of processors are shut down when they're not in use.

      (In fact, this is a major piece of chip design technology that AMD brought to the table when they absorbed ATI. Actual synergy in action.)

      In contrast, the solution you're proposing (multiple video cards) is going to be less efficient, because you're duplicating a lot more than just the components you want to disable when you're just rendering a desktop. Power regulators, RAMDACs, etc.

    25. Re:Power Consumption by Anonymous Coward · · Score: 0

      It's fairly trivial to underclock nVidia cards when running in 2D mode with the utility NiBiTor. I have an 8600GTS OC2 with the following clock speeds:

      GPU: 720MHz
      Shaders: 1547MHz
      RAM: 1110MHz

      All at 1.3V. I used a software utility (Powerstrip) to find the lowest stable settings and ended up with this:

      GPU:144MHz
      Shaders:324MHz
      RAM:200MHz

      At 1.1V. Applied these settings to the BIOS (Also changed the boot clocks to the lower values) and off I went.

      While I don't have equipment to monitor power usage, the card does run 8C colder in 2D mode now.

    26. Re:Power Consumption by Jarik_Tentsu · · Score: 1

      I'm not sure about video cards, but most high-end mobos these days come with various things which both lower heat production and power consumption.

      For instance I have the Gigabyte EX38-DQ6 and the main function that matter are:
      - EIST function - Multiplier runs at 6x while idle, then goes to 9x when under load.
      - Auto voltage function - Mobo controls the voltage dynamically based on how much it thinks the CPU needs to be stable. The only thing is, this always seems to be *higher* than what I've tested stable. =/
      - Dynamic Energy Management - Probably the fanciest feature of this mobo. I don't use it 'cause I'm an overclocker, but it's a software controlled variable power thing. My mobo has 10 phase power, and it can turn them off based on CPU usage. To make things even more smooth running you can put on dynamic FSB, which once again, changes it based on FSB. It tells you exactly how much energy (well, it lables it as power) you've saved.

      Considering that software based overclocking software for graphics cards is hugely abundant, I'm sure there's one that can dynamically change teh clock speed and voltage based on usage.

      ~Jarik

    27. Re:Power Consumption by Glock27 · · Score: 1

      If you read the article (shocking idea I know) you'll see NVIDIA is working on a solution where a more efficient on-board or on-chip GPU can take over when heavy 3D lifting isn't needed. A 'greener' GPU architecture, in other words.

      --
      Galileo: "The Earth revolves around the Sun!"
      Score: -1 100% Flamebait
    28. Re:Power Consumption by Anonymous Coward · · Score: 0

      Quartz Extreme (acceleration of window composition) has been on by default since 10.2 (the first version with support for it).

      Quartz 2D Extreme aka QuartzGL (acceleration of Quartz 2D graphics primitives) is the part which debuted in 10.4, defaulted to off. In 10.5 it's on by default, but individual applications have to use an API call to request it. Several of the bundled applications do turn it on. One example is Preview.app -- try scrolling through a large and complex PDF document on an Intel Mac with a fast GPU on 10.5, it's very fast.

  6. MY Space Heater! by Zymergy · · Score: 1

    I for one, Now plan on purchasing a new space heater soon for my box (NOTE: Nvidia GT200 has a TDP of 236W!)... so long as I can FINALLY have Crysis playable at resolution!

    -Another good article on the GTX280 (GT200 GPU) at TR: http://www.techreport.com/articles.x/14934

    1. Re:MY Space Heater! by Firehed · · Score: 2, Funny

      236W? I doubt you'll have any problems kicking ass on the frozen levels.

      --
      How are sites slashdotted when nobody reads TFAs?
    2. Re:MY Space Heater! by mikael · · Score: 1

      Dr. Batenburg is going to have to buy four of these cards to upgrade his Fastra

      --
      Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
    3. Re:MY Space Heater! by Zymergy · · Score: 1

      Uh, ...he's gonna need to get a bigger Boat.., er. I mean Case and PSU...(maybe)
      Plus the Doc will likely need to remove the other side of his current case as well (which it appears he cannot)... Better yet, Doc could juet get a bigger and better case.

      I suggest this one from NZXT: http://www.techreport.com/gallery/index.x?id=14909&image=28934 True, it does have excessive blue LEDs (Why?) but it is still a nice case.
      It also has room for DUAL PSUs so he can run 2 of those big 1500W PSUs (Most PSUs have better efficiencies at less than rated loads and a longer MTBF.)

  7. did i read the same review? by Anonymous Coward · · Score: 1, Interesting

    The article talks about the new card smoking ATI and showing 50% improvement yet the benchmark chart at the end of the article shows only a couple games got 50% fps boosts and ATI still outperformed it in Crysis with a card that is available today.

  8. Vista cuts performance... by corsec67 · · Score: 0, Troll

    The GTX 280 delivered real-world benchmark numbers nearly 50 percent faster than a single GeForce 9800 GTX running on Windows XP, and it was 23-percent faster than that card running on Vista.


    Why would Vista make the performance gains so much less? I could see XP running say 20% better with both cards, but why does Vista penalize the new card so much?

    Digital Restrictions Management strikes again, I guess...

    Vista: where do we want you to go today?
    --
    If I have nothing to hide, don't search me
    1. Re:Vista cuts performance... by Anonymous Coward · · Score: 0

      where do we allow you to go today

    2. Re:Vista cuts performance... by pdusen · · Score: 2, Insightful

      Actually, it's likely just a less developed Vista driver, like most performance problems people report with Vista (and by report, I mean actually experience and document, not the random anti-Vista FUD it's so popular to spout these days.)

    3. Re:Vista cuts performance... by Colonel+Korn · · Score: 5, Interesting


      Why would Vista make the performance gains so much less? I could see XP running say 20% better with both cards, but why does Vista penalize the new card so much?

      Digital Restrictions Management strikes again, I guess...

      Vista: where do we want you to go today?
      TFA has some very weird numbers compared to Anandtech and Tomshardware and all the other real review sites that actually tell you all the details of their testing. The 280 looks more like it's 50-75% faster than the 9800GTX in most reviews, and most of those are done in Vista. Framerate in XP vs. Vista is completely even on a 9800 GTX with current drivers (the Vista slowdown went away a long time ago), except on Oblivion where Vista is about 20% faster for no apparent reason, but maybe the drivers Maximum PC used weren't the same as those used by the serious review sites, or maybe they have something wrong with their Vista install.
      --
      "I zero-index my hamsters" - Willtor (147206)
    4. Re:Vista cuts performance... by Z34107 · · Score: 2, Interesting

      There is no "Vista DRM." That little copy protection stuff lets you play Blu-Ray discs.

      You can rip CDs, DVDs, and pirate t3h internetz if you want. I do so on a daily basis on my Vista x64 machine.

      Now, if OS support for DRM bothers you, take it up with the studios that require it. Not playing DVDs is not an option.

      --
      DATABASE WOW WOW
    5. Re:Vista cuts performance... by cduffy · · Score: 1

      Smaller companies can get away with saying "the studios require it". Microsoft can't.

    6. Re:Vista cuts performance... by drsmithy · · Score: 1

      Smaller companies can get away with saying "the studios require it". Microsoft can't.

      Why ? Microsoft are insignificant players in content creation and delivery marketplace.

    7. Re:Vista cuts performance... by Firehed · · Score: 1

      Vista's been out, what, 18 months? You think they'd get on that driver thing, especially as it's nVidia's shitty drivers that are causing so much with Vista in the first place. I'm genuinely surprised that they haven't been sued by Microsoft at this point, seeing how much of the Vista-hate is their fault.

      --
      How are sites slashdotted when nobody reads TFAs?
    8. Re:Vista cuts performance... by bigstrat2003 · · Score: 1

      Exactly! People refuse to get this through their heads for some reason. Microsoft has a lot of pull in some circles, but that doesn't include Hollywood. When it comes to content delivery, the studios don't need Microsoft, Microsoft needs them. It's no different in this case than it is for smaller companies.

      --
      "16MB (fuck off, MiB fascists)" - The Mighty Buzzard
    9. Re:Vista cuts performance... by XnavxeMiyyep · · Score: 0, Offtopic

      Offtopic, but

      You could change One day I'm going to get to One day I'll get and you'd have room to spell out last in your sig.

      --
      I put the 't' in electrical engineering.
    10. Re:Vista cuts performance... by zehnra · · Score: 1

      Maximum PC does mention the details of their testing, but as it is primarily a print magazine first and a website second, these things aren't included in every article. I believe they publish an article on how they're benchmarking hardware about once a year (or whenever they revamp the procedure). There's only so many pages in an issue, whereas web space is much cheaper and therefore easier to expand.

      I don't see Maximum PC so much as a "serious review site" as i do a "serious review magazine" with a web presence. Similarly, if Anandtech or Tomshardware put out a magazine, I'd look at that as a secondary venture and view their sites first.

    11. Re:Vista cuts performance... by steveaustin1971 · · Score: 0, Troll

      Why would anyone concerned about performance EVER install Vista to begin with? for directX games XP works MUCH faster, and has fewer issues... as far as I'm concerned Vista is just another layer of bloat.

    12. Re:Vista cuts performance... by cduffy · · Score: 1

      Microsoft owns the desktop. Content creation and delivery folks want the desktop. What does their (lack of) position in the content market matter?

      Do you really think anyone could sell video content that wouldn't play on Vista?

    13. Re:Vista cuts performance... by drsmithy · · Score: 3, Insightful

      Microsoft owns the desktop. Content creation and delivery folks want the desktop. What does their (lack of) position in the content market matter?

      The content folk are, at best, highly suspicious of "the desktop". With good reason.

      Do you really think anyone could sell video content that wouldn't play on Vista?

      Of course they could. Most people consume their content from standalone commodity appliances like DVD players and iPods. This hasn't changed in the last few decades (substitute "VHS", "Cassette", "LP", etc as necessary) and there's little reason to think it will in the future.

    14. Re:Vista cuts performance... by cduffy · · Score: 1

      The content folk are, at best, highly suspicious of "the desktop". With good reason.
      Old content, sure. New content embraces it. That said, old content isn't that wary -- they keep selling DVDs with bonus data-track content, after all.

      Most people consume their content from standalone commodity appliances like DVD players and iPods.
      Most people also tend to wait it out before adopting new technologies; being an early winner or an early loser can make a difference between being Blu-Ray or HD-DVD.

      All that said, my only relevant industry experience is new content.
    15. Re:Vista cuts performance... by pdusen · · Score: 1

      You can always tell someone who has never tried to game on Vista by comments like this.

    16. Re:Vista cuts performance... by Firehed · · Score: 1

      No, I have. Game performance was perceptibly the same as in XP when nVidia's drivers weren't crashing my system (I really don't care what the numbers said - I could play at the same settings with the same smoothness; if it was 2fps lower, I couldn't tell and therefore I don't care). The only consistent performance issue I've seen in Vista (on a half-decent computer, rather than the tremendously half-assed gear you'll get from HP and such) was related to network file transfers; my understanding is that this was resolved in SP1.

      --
      How are sites slashdotted when nobody reads TFAs?
  9. GX2 Cheaper and Faster by Colonel+Korn · · Score: 4, Informative

    In most reviews, the 9800GX2 is faster, and it's also $200 cheaper. As a multi-GPU card it has some problems with scaling, and micro-stutter makes it very jumpy like all existing SLI setups.

    I'm not well versed in the cause of micro-stutter, but the results are that frames aren't spaced evenly from each other. In a 30 fps situation, a single card will give you a frame at 0 ms, 33 ms, 67 ms, 100 ms, etc. Add a new SLI card and let's say you have 100% scaling, which is overly optimistic. Frames now render at 0 ms, 8 ms, 33 ms, 41 ms, 67 ms, 75 ms, 100ms, and 108ms. You get twice the frames per second, but they're not evenly spaced. In this case, which uses realistic numbers, you're getting 60 fps might say that the output looks about the same as 40 fps, since the delay between every other frame is 25 ms.

    It would probably look a bit better than 40 fps, since between each 25 ms delay you get an 8 ms delay, but beyond the reduced effective fps there are other complications as well. For instance, the jitter is very distracting to some people. Also, most LCD monitors, even those rated at 2-5 ms response times, will have issues showing the 33 ms frame completely free of ghosting from the 8 ms frame before the 41 ms frame shows up.

    Most people only look at fps, though, which makes the 9800 GX2 a very attractive choice. Because I'm aware of micro-stutter, I won't buy a multi-GPU card or SLI setup unless it's more than 50% faster than a single-GPU card, and that's still ignoring price. That said, I'm sort of surprised to find myself now looking mostly to AMD's 4870 release next week instead of going to Newegg for a GTX280, since the 280 results, while not bad, weren't quite what I was hoping for in a $650 card.

    --
    "I zero-index my hamsters" - Willtor (147206)
    1. Re:GX2 Cheaper and Faster by The+Living+Fractal · · Score: 1

      You don't have SLI... I do, and micro stutter is barely noticeable at worst. And I can play at resolutions and anti-aliasing that no sigle card could've made playable.

      Short answer: it's fine. if you have the money, and want to play at extreme resolutions, get SLI.

      It almost seems like micro stutter is some kind of viral ATI anti-marketing bs.

      --
      I do not respond to cowards. Especially anonymous ones.
    2. Re:GX2 Cheaper and Faster by Colonel+Korn · · Score: 4, Interesting

      You don't have SLI... I do, and micro stutter is barely noticeable at worst. And I can play at resolutions and anti-aliasing that no sigle card could've made playable. I've had three SLI setups (an ancient 3dfx X2 and two nVidia pairs). I liked my first SLI rig but I felt not to satisfied with the feel of the last two when compared to a single card, and now that I've learned about this issue I know why. Lots of people say that microstutter is barely noticeable, but lots of people also insist that a $300 HDMI cable gives "crisper" video over a 6 foot connection than a $10 HDMI cable. The micro-stutter effect that you can barely notice is the inconsistency in frame delay, which I mentioned ("For instance, the jitter is very distracting to some people."), but beyond that, there's the problem I described with the bulk of my comment. It's not just a question of whether you can tell that frames are coming in in clumps. It's a question of whether you can tell the difference between 60 fps delays, which is what you paid for, and 40 fps delays, which is what you get. SLI definitely improves performance and for those of us who don't mind the jitter (I never did, actually), it is an upgrade over a single card, but even with 100% scaling of fps, the benefit is more like a 33% increase in effective fps.

      It almost seems like micro stutter is some kind of viral ATI anti-marketing bs. Definitely not, and definitely not BS, but speaking of ATI, rumor has it that the 4870x2 may adapt the delay on the second frame based on the framerate, eliminating this problem. If it's true, then it will be the best dual-GPU card relative to its own generation of single card ever, by a very large margin. But of course, the rumor may just be some kind of viral ATI marketing bs. ;-) I hope it's true.
      --
      "I zero-index my hamsters" - Willtor (147206)
    3. Re:GX2 Cheaper and Faster by wc_paladin · · Score: 1

      this is interesting. does AMD's crossfire do this as well?

    4. Re:GX2 Cheaper and Faster by Anonymous Coward · · Score: 0

      Yes, but if you use vsync, microstutter goes away. Completely. You should never run SLI without vsync -- it just looks weird.

    5. Re:GX2 Cheaper and Faster by PIBM · · Score: 1

      I've had an sli rig before, and I noticed what you'd call microstutters on a good old CRT but it wasn't as you pointed. Also, depending on which SLI mode you were using, the frames rendering where split up between the two cores ( dynamically, so one could end up doing 1/3 the other 2/3 if there was less things to render in ) at which point your explanation fail to explain the stutterings (maybe it was just bad drivers, that was 6600GTs at the beginning).

      Also, for your explanation, on any good game running on sli high-tech cards, your CPU usage should be pretty topped off, meaning that it will be spending most of it's time preparing the data for the next frame, thus preventing the kind of results you spoke of from happening.

      Anyway, I switched to a 7800GTX as soon as they got out for not much more than I sold both my 6600s and I've been happy ever since (did the same thing to go to the 8800GT, but this time I won't make it to the gtx280 I believe :( )

    6. Re:GX2 Cheaper and Faster by AbRASiON · · Score: 1

      Wow, I'd never heard of this, thanks for bringing it to our attention.

    7. Re:GX2 Cheaper and Faster by wildstoo · · Score: 1

      Sorry AC, but my experience of this issue varies.

      I have a 9800GX2. If I turn vsync on in World of Warcraft I get a BSOD within about 2 minutes. With vsync off it's absolutely fine. Yes, this is with the latest drivers, though the same thing happens with previous drivers. Even if I could get vsync to work, Triple Buffering with SLI results in massive flickering of textures with alpha channels. Not good. This is all under WinXP, I haven't tried it with Vista.

      And yes, I notice microstutter with vsync off. Microstutter isn't the same as the regular tearing that you might associate with no vsync, although the end result - a distracting, flickery pseudo-jerkyness - is the same.

      Other games don't exhibit this behaviour; Source Engine games, for example, are fine with or without vsync. Certain other games have their own quirks with SLI, it's a mixed bag.

      Ultimately, SLI is fantastic when it works well. Unfortunately that's not always the case, even with recent drivers and modern games that *should* support it properly.

    8. Re:GX2 Cheaper and Faster by Anonymous Coward · · Score: 0

      I've had three SLI setups (an ancient 3dfx X2 and two nVidia pairs). I liked my first SLI rig but I felt not to satisfied with the feel of the last two when compared to a single card, and now that I've learned about this issue I know why. Not sure what you saw then, because 3Dfx SLI did not suffer from micro-stutter. 3Dfx had true SLI: both GPUs worked on the same frame at the same time, using scan lines as the unit of work distribution (one card generated even scan lines, the other card odd). All the geometry went through one side (the geometry units on the other card would just sit idle) so basically the system functioned like a single Voodoo GPU with twice the normal fillrate.

      As you say, all modern 'SLI' implementations seem to be AFR (Alternate Frame Rendering) in practice, not SLI (Scan Line Interleaving).
  10. Thank goodness! by DRAGONWEEZEL · · Score: 1

    I was just about to go buy a new video card! Now I'll hold out!

    --
    How much is your data worth? Back it up now.
  11. Noise leveb by eebra82 · · Score: 4, Informative
    Looks like NVIDIA went back to the vacuum cleaner solution. Blatantly taken from Tom's Hardware:

    During Windows startup, the GT200 fan was quiet (running at 516 rpm, or 30% of its maximum rate). Then, once a game was started, it suddenly turned into a washing machine, reaching a noise level that was frankly unbearable - especially the GTX 280.
    Frankly, reviews indicate that this card is too f*cking noisy and extremely expensive ($650).
    1. Re:Noise leveb by clampolo · · Score: 2, Informative

      and extremely expensive ($650)

      Not at all surprising. Did you see the size of that chip die? You can fit 6 Penryn on it!! I used to work for a semiconductor company and the larger the chip the more expensive it gets. This is because the larger the die is the less likely it is to be defect free when it comes out of the fab.

    2. Re:Noise leveb by p0tat03 · · Score: 1

      You would think for that price they'd give you a nicer fan. This reminds me of my old Toshiba laptop - fans were as loud as small jet engines. I was very pleasantly surprised by my new laptop (MacBook Pro), which is whisper quiet, and even when going full tilt you can only hear some wind from the fans, and not the fan motor itself. Awesome. Makes me think that nVidia cheaped out on the fan...

  12. Re:9800GX2 != 2x9800GTX by pdusen · · Score: 1, Informative

    and while they support DX10, who needs that when games under the wonderful OS Vista run twice as slow than they do on XP? That would be a good point, if it were true.

    Only you can prevent FUD!
  13. can't... resist... by tripmine · · Score: 0, Troll

    But does it run Crysis on full? I mean, can it REALLY?

  14. More details and analysis, here by Anonymous Coward · · Score: 0

    More benchmarks, power numbers and GPGPU testing here - http://www.hothardware.com/Articles/NVIDIA_GeForce_GTX_280_and_GTX_260_Unleashed

  15. I call bull on those conclusions. by JohnnyBigodes · · Score: 2, Insightful

    and in some cases beat â" two 9800 GTX cards running in SLI, a fact that explains why Nvidia expects the 9800 GX2 to fade from the scene rather quickly.

    Bullshit. The 9800GX2 is consistently quite a bit faster (TechReport's very detailed review here), and it costs around $450, while the GTX 280 costs $650 (with the younger brother the 260 at $400), with the only drawbacks being more power drawn and higher noise. Even then, I think it's a no-brainer.

    Don't get me wrong, these are impressive single-GPU cards, but their price points are TOTALLY wrong. ATI's 4870 and 4850 cards are coming up at $450 and $200 respectively, and I think they'll eat these for lunch, at least in the value angle.

    1. Re:I call bull on those conclusions. by Xelios · · Score: 1

      The 4870 will be $350, not $450. And at that price Nvidia is going to have a hard time convincing me to buy a GTX 280, even if it does turn out to be marginally faster.

      Lets see how the reviews of the 4800 series pan out.

      --
      Murphey's fighting Occam, and we're in the stands.
    2. Re:I call bull on those conclusions. by CastrTroy · · Score: 1

      ATI's 4870 and 4850 cards are coming up at $450 and $200 respectively, and I think they'll eat these for lunch, at least in the value angle.
      People buying $400 video cards aren't looking for value. Around $200, I could see the price being a factor. However, once you've decided to spend $400 on a video card, price isn't even something you are considering.
      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    3. Re:I call bull on those conclusions. by JohnnyBigodes · · Score: 1

      Not everyone who makes a relatively large investment is "mindless", so to speak. That's what I call an expensive graphics card, an investment. And even the big ones must pay off somehow, and these new cards don't.

    4. Re:I call bull on those conclusions. by CastrTroy · · Score: 1

      I find it really hard to follow the logic that an object that will be worth 50% of it's current value in a year (and in each consecutive year) to be an investment. It would be hard to argue that a new production model car to be an investment. If you kept it in the garage until 25 years later it might be worth more than the original, by 3 or 4 times. However, if you just took the original money you spent on the car and invested it for 25 years, you would end up way ahead. With a car, at least you could argue that eventually it will be a collectors item. I don't see a graphics card ever being a collectors item.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    5. Re:I call bull on those conclusions. by JohnnyBigodes · · Score: 1

      I didn't say it was meant to be a collector's item, that would be quite ridiculous :) What I mean is that for me, a proper investment is to stay on top of what I intend to have in regards of resolution/image quality (1680 @ 4xAA) for a reasonable timeframe without upgrades. My 8800GTX has served me extremely well in this regard, as I can now still sell it for almost half its original value and pay roughly half or more of a new card, give or take a few.

    6. Re:I call bull on those conclusions. by TheLink · · Score: 1

      Well if he can play his desired game NOW in its full "maximum quality" glory, and he typically spends USD400 a night on entertainment, then it could actually help him save money.

      Basically he spends USD400, plays computer games for a few nights, and actually ends up with more money than he would otherwise (I actually know someone who did save some money in a similar way). In contrast if it were USD4000 for a vid card, the calculation could be different - he could get bored of the various games and go back to spending more.

      The other way it could be an investment is if he's a professional gamer. Higher fps is better for many games.

      --
    7. Re:I call bull on those conclusions. by Lord+Ender · · Score: 1

      If you play video games for, say, two hours per day, compare the cost of a $400 graphics card every two years to the cost of seeing a two hour movie every day (or pick your own form of entertainment).

      A $400 video card really is a smart business decision when you look at entertainment-hours per dollar.

      --
      A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
  16. So, practically who bought 9800 by unity100 · · Score: 2, Interesting

    are royally screwed ? it was a 'new' card and all.

    well done nvidia. very microsoft of you.

    1. Re:So, practically who bought 9800 by Jellybob · · Score: 4, Insightful

      AMD and NVidia are always going to release new cards. It's just the way of the industry.

      If you buy a graphics card in the hope that it's going to be the top of the line card for longer then a few months then you're very much mistaken.

      Buy a card that will do what you need it to, and then just stick with that until it stops being powerful enough for you. Anyone hoping their computer will be "future proof" is heading towards disappointment very fast.

    2. Re:So, practically who bought 9800 by Richard_at_work · · Score: 1

      So tell me, which idiots believe the world is going to stand still just because they paid out money for something?

    3. Re:So, practically who bought 9800 by Barny · · Score: 1

      You must be new to the whole "computer part" thing...

      --
      ...
      /me sighs
    4. Re:So, practically who bought 9800 by CastrTroy · · Score: 1

      Does the 9800 stop working, or in some way slow down, because NVidia came out with something new? Did they say they were going to cut off driver support for the 9800? You got what you paid for, and you still have what you paid for. New stuff will always come along that's cheaper and more powerful than what already exists.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    5. Re:So, practically who bought 9800 by rjstanford · · Score: 1

      Everyone who bought an iPhone at launch, apparently. Hell, I did - and didn't feel ripped off when the price dropped later, but apparently I was the only one...

      --
      You're special forces then? That's great! I just love your olympics!
    6. Re:So, practically who bought 9800 by Lord+Ender · · Score: 1

      The price of a 9800GTX has gone from $350 to $220 in a very short time. I just bought a 9800GTX for $220 from NewEgg a few hours ago. They are dropping the price to make room for these new cards, and you would be wise to take advantage of it. The new card may be 50% faster, but it is much more than 50% more expensive at today's prices.

      --
      A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
    7. Re:So, practically who bought 9800 by LarsG · · Score: 1

      You were one of those that bought an iPhone before September '07, right?

      --
      If J.K.R wrote Windows: Puteulanus fenestra mortalis!
    8. Re:So, practically who bought 9800 by unity100 · · Score: 1

      pushing a 'new' card just before you are going to release a next gen card is just bad trade practice and nothing else.

    9. Re:So, practically who bought 9800 by unity100 · · Score: 1

      world neednt stand still, but at least tarry for 4-5 months before phenomenally slashing the value of whatever something that has been sold. at least, more than 1.5 months ffs.

    10. Re:So, practically who bought 9800 by unity100 · · Score: 1

      i can accept 6 months. but i cant stomach 1.5 months. 9800 cards were just new.

    11. Re:So, practically who bought 9800 by Xugumad · · Score: 1

      What puzzles me is the people who build these insane watercooled dual-SLI rigs. Sure, it's fast today, but give it 9 months and standard kit will do that, and 18 months down the line it'll be out of date.

  17. no surprise by unity100 · · Score: 2, Informative

    8000gts were much louder than their 3870 counterparts too.

    i dont get why people fall for that - push a chip to limits, put a noisy fan on it, and sell it as high performance card.

    at least with ati 3870 you can decide whether you gonna overclock the card and endure the noise or not.

  18. ATI's Response? by mandark1967 · · Score: 2, Interesting

    Anyone know when the new ATI card will be released?

    Based on the information I've seen on it, it will be pretty comparable in terms of performance, but at a far cheaper price.

    I'm hoping that the new ATI card performs within 10% - 15% or so of the GTX280 because I'm getting a bit tired of the issues I have with my current nVidia 8800GTS cards. (SLI)

    I cannot set the fanspeed in a manner that will "stay" after a reboot.

    My game of choice actually has some moderate-sever issues with nVidia cards and crashes at least a couple times a week due to some issue with nvcpl which I have bugged for 10 different versions of drivers and they never fix

    My last ATI Card was their 9700Pro. I switched to nVidia because, while their drivers are closed source, the installation in Linux is far easier and their performance was pretty top-notch. Now I'm considering switching back to ATI if they can deliver a decent competitor.

    --
    Sig Follows: "Suppose you were an idiot. And suppose you were a member of Congress. But I repeat myself." -- Mark Twain
    1. Re:ATI's Response? by UncleFluffy · · Score: 1

      Looks like it's being released right now. Best wait a week or two until the reviews are out and you can compare the two before wasting your valuable beer tokens.

      Inquirer camping outside the NDA session.
      --

      What would Lemmy do?

  19. Tame Crysis at 1900x1200? by L4t3r4lu5 · · Score: 2, Insightful

    I run Crysis, all maxed out, on an 8800gtx, and only get lower than 30fps in the end battle.

    If I want more speed, i'll get another 8800. That card is phenomenal, and about to get a lot cheaper.

    --
    Finally had enough. Come see us over at https://soylentnews.org/
    1. Re:Tame Crysis at 1900x1200? by ady1 · · Score: 1

      Exactly what I was thinking. I am able to run crysis on 1900x1200 with 8800GT sli with >40fps (all setting very high).

      I understand using crysis as a benchmark but pretending that there wasn't any setup capable of running crysis on 1900x1200 is exaggerating

    2. Re:Tame Crysis at 1900x1200? by Doddman · · Score: 1

      Hell, I run it on a mediocre rig (AMD 2.5ghz @ 2.9ghz, 4gb ddr2-800, 9600gt) at medium with high texture quality (which looks amazing, compared to most shooters) and I get >40fps at 1680 except during the final battle or other high-explosive events (bombing the destroyer being one).

      --
      If creativity is the field, copyright is the fence.
    3. Re:Tame Crysis at 1900x1200? by CodyRazor · · Score: 1

      Also, accordign to this article crysis runs at 22fps, and I suspect thats without much AA. How is this taming it?

      --
      So Skulldilocks threw acid on the schoolchildrens' faces, cause somebody from the bible told her to do it!
  20. DirectX 10 is the reason by DrYak · · Score: 4, Informative

    I could see XP running say 20% better with both cards, but why does Vista penalize the new card so much? Crysis is a DirectX 10 game.
    When run under Vista, it features tons of additional effects. Those are the reasons why the speed improvement in Crysis aren't that much impressive under Vista.

    PS: And for the record, Radeon HD3870X2 uses the exact same GDDR3, not GDDR4 as TFA's review says. ATI choose to go for GDDR3 to cut the costs of the dual GPU setup. (Only a few non standard boards by 3rd party manufacturer use GDDR4 and a PCI-express 2.0 bridge).
    --
    "Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
    1. Re:DirectX 10 is the reason by rgviza · · Score: 1

      Crysis is a DirectX 10 game. When run under Vista, it features tons of additional effects. Those are the reasons why the speed improvement in Crysis aren't that much impressive under Vista. ----------------- /QFT There's a lot of additional rendering happening with a DX10 vs. DX9 even using the same game, if you have the DX10 features enabled in the game. A more interesting comparison could have been done using a game which supports DX9 or DX10 on vista, and running it in DX9 mode to compare to XP. -Viz

      --
      Don't kid yourself. It's the size of the regexp AND how you use it that counts.
    2. Re:DirectX 10 is the reason by Anonymous Coward · · Score: 4, Informative

      When run under Vista, it features tons of additional effects. Stupidly enough, a 20% loss in framerate is all you get by running the game in DX10 mode. All of the 'DX10' effects can be enabled in DX9 (on XP as well) with simple modifications to the game's configuration files. When you do it, it looks exactly the same as DX10 without the deleterious effect on framerate.

      I suppose if the reviewers were stupid, they may not have run the game in DX9 mode on both XP and Vista, which would account for the difference even if the graphical options were set to the same levels. But it doesn't make the game look better just by running it in DX10 mode, the only difference is that it's slower.

      For those who aren't familiar with this: in XP/DX9 mode the highest-level graphical options are grayed out. This is an entirely artificial limitation; the configuration changes I mentioned simply replace graphical options with a higher ones (they're just integers in a plaintext file), so for example 'High' (DX9) becomes 'Very High' (the 'DX10' effects) in practise.
    3. Re:DirectX 10 is the reason by Artem+S.+Tashkinov · · Score: 0

      When run under Vista, it features tons of additional effects. I doubt that. M$ marketers are masters of brainwashing.
    4. Re:DirectX 10 is the reason by Anonymous Coward · · Score: 0

      Crysis is a DirectX 10 game.
      When run under Vista, it features tons of additional effects. Those are the reasons why the speed improvement in Crysis aren't that much impressive under Vista. Actually the restriction is wholly artificial. If you look at the plaintext files in CVarGroups you can clearly see the DirectX10 values under 'Very High' differ only slightly from the the DirectX9 'High' options.
      When one modifies the config files by replacing another setting with the 'Very High' configuration variables the DirectX10 exclusive features magically appear. The water is more realistic, the light looks different in the trees and during sunrise/sunset, tire tracks are amazing, etc.

      On my system (2180, 2Gb, 8800GT, XP SP2) I edited the config files to turn only the DX10 options on- leaving the other settings such as motion blur and depth of field on their High equivalents. The performance difference was a maximum of 9fps. Usually it was under 5. On true Very High settings the difference increased considerably but that's understandable due to the actual differences in rendering options.
  21. Great news - not that I want to buy the thing... by Enleth · · Score: 1

    But every time Nvidia releases "THE new, big thing" the prices of the previos and, especially, the second-previous generation cards drop by a significant amount, making them worth the buck for an occasional gamer who doesn't want to spend a fortune to play games and is happy with his games running on the low to medium details settings.

    --
    This is Slashdot. Common sense is futile. You will be modded down.
  22. Seriously by Anonymous Coward · · Score: 1, Funny
    Did you really read the "MAXIMUM PC" article
    did you read

    The 8800 GTX has 24 ROPs and the 9800 GTX has 16, but if the resulting pixels need to be blended as they're written to the frame buffer, those two GPUs require two clock cycles to complete the operation. The 9800 GTX, therefore, is capable of blending only eight pixels per clock cycle. The GTX 280 not only has 32 ROPs but is also capable of blending pixels at full speed--so its 32 ROPs can blend 32 pixels per clock cycle. The GTX 260, which is also capable of full-speed blending, is outfitted with 28 ROPs. without you eyelids drooping and your head nodding onto your chest
    You did ?
    You may enjoy this or this then.
  23. Noise... by Guanine · · Score: 2, Funny

    Yes, but will you be able to hear your games over the roar of the fans on this thing?

    1. Re:Noise... by dingleberrie · · Score: 1

      The word on the street is that the flight simulators are very immersive.

  24. RTFA by ruiner13 · · Score: 2, Informative
    If you'd RTFA, you'd note this part:

    Power Considerations

    Nvidia has made great strides in reducing its GPUs' power consumption, and the GeForce 200 series promises to be no exception. In addition to supporting Hybrid Power (a feature that can shut down a relatively power-thirsty add-in GPU when a more economical integrated GPU can handle the workload instead), these new chips will have performance modes optimized for times when Vista is idle or the host PC is running a 2D application, when the user is watching a movie on Blu-ray or DVD, and when full 3D performance is called for. Nvidia promises the GeForce device driver will switch between these modes based on GPU utilization in a fashion that's entirely transparent to the user. So, yes, they hear you, and are making improvements in this area.
    --

    today is spelling optional day.

  25. The scene has changed. by wild_quinine · · Score: 5, Interesting
    I used to be near the front of the queue for each new line of graphics cards. I used to wait just long enough for the first price drops, and then stump up. Cost me a couple of hundred quid a year, after the sale of whatever my old card was, to stay top of the line. Compared to owning and running a car, which I don't, owning and running a super-rig was positively cheap (in the UK). Some might call it a waste of money, and I have sympathy for that argument, but it was my hobby, and it was worth it, to me.

    This year I put my disposable income towards getting in on all three next generation consoles, and the PC will languish for a long time yet.

    I don't think I've changed, I think the market has changed.

    They're getting bigger and hotter, and no longer feel like cutting edge kit. They feel like an attempt to squeeze more life out of old technology.

    DirectX 10 as a selling point is a joke, with the accompanying baggage that is Vista all it does is slow games down, and none of them look any better for it yet. In any case, there are only five or six of them. You can pick up an 8800GT 512 for less than 150 dollars these days, and it's a powerhouse, unless you're gaming in full 1080p. There is no motivation to put one of those power hungry bricks in my rig. Nothing gets any prettier these days, and FPS is well taken care of at 1680x1050 or below.

    Game over, graphics cards.

    I wonder what will happen if everyone figures this out? Imagine a world in which the next gen of consoles is no longer subsidised, or driven, by PC enthusiasts...

    1. Re:The scene has changed. by ZERO1ZERO · · Score: 1

      150 dollars, 75 GBP, or 150 GBP?

    2. Re:The scene has changed. by TheNetAvenger · · Score: 1

      DirectX 10 as a selling point is a joke, with the accompanying baggage that is Vista all it does is slow games down, and none of them look any better for it yet

      1) There hasn't been a real DX10 released to date. All DX10 games are hybrid DX9 games with DX10 features like larger textures strapped on. When you see a title that is DX10 ONLY, then the performance DX10 offers will be noticeable, until then, all we are getting is eye candy.

      2) If you are going to argue DX10 is bad, then explain to everyone how larger texture sizes for more detail is a 'bad thing'... (I know this is just one example, but semi-important.)

      3) With NVidia pushing Physics, after Microsoft begging them to adopt to an open Physics model almost 4 years ago, it is down right ironic watching NVidia today. (This was part of the pissing match of the XBox to XBox 360 timeframe, and why MS designed their own GPU with early DX10 concepts, like Physics available on the GPU.)

      4) With games implementing 'fragmented' physics implementations, it will hurt ATI and NVidia more than help them, and NVidia is the donkey pushing the cart here. DirectX10 specficially provides a common framework for doing physics and supplying the interface to games so that people don't have to look for CUDA-NVidia, Aegis, etc...

      DX10 can do GPU physics and opens this world to developers because of the Vista WDDM, that multi-tasking the processes to the GPU - 3D is no longer cooperative multi-tasking like XP, and surprisingly you would think tech geeks at SlashDot even would, go, oh that is cool, pre-emptive GPU at the OS level. (No OpenGL or DirectX yielding needed.) Vista also provides SMP GPU processing. (Hopefully SLI and Crossfire will die a slow horrible death once XP is dead.)

      NVidia is still making their GPU technology based on pre-Vista world technology, because they don't want to break with the XP market. Once the XP break is designed into the GPU, and Vista is required for the multi-core, blah, blah features, we will see some really nice GPU technologies that will spin heads.

      As for your comments about Vista being slower than XP in gaming, you are a year behind.

      Here read a few pages:
      http://www.extremetech.com/article2/0,2845,2302499,00.asp

      About Sept of 07, the drivers caught up and passed XP on a lot of games (SLI is the exception, as they can be equal or XP has an edge). This is why when you look at reviews at Tom's and other sites they are all using Vista for baseline profiling GPUS/Video cards.

      The only 'bad' thing about a lot of the review sites, they don't respect the whitepaper on how to test a 'new' Vista install, and are running benchmarks while Vista is doing background I/O or other really stupid stuff.

    3. Re:The scene has changed. by wild_quinine · · Score: 1

      150 dollars, 75 GBP, or 150 GBP? I'm a bit off at the moment. The cheapest 512 I can find is just under 100 pounds, not 75. (ebuyer quick code 145605) 256 can be had for 60ish, although this one sold out this week: http://www.saverstore.com/productinfo/Product.aspx?product_id=20020220&rstrat=1032
    4. Re:The scene has changed. by wild_quinine · · Score: 1

      When you see a title that is DX10 ONLY, then the performance DX10 offers will be noticeable, until then, all we are getting is eye candy. Maybe your eyes are better than mine, but I don't think we're even getting that.

      If you are going to argue DX10 is bad I'm not. Bundling it into Vista is bad, for a slew of reasons, and the shit they've pulled with several 'Vista only' titles, and the Crysis ini-locking is enough to errode customer confidence in there being any kind of need to have done that.

      I see absolutely nothing to recommend Vista over XP, at this time or in the near future. Sure, there are things that are better in Vista than XP, but none of them, or even all of them, combine to reason enough to switch.

      As for your comments about Vista being slower than XP in gaming, you are a year behind. Here read a few pages: http://www.extremetech.com/article2/0,2845,2302499,00.asp

      Hardly a year out of date. The figures you post are one month old, and involve Vista SP1 final, vs SP3 of XP. I admit I am impressed by the evening out that Vista has managed to achieve, in those tests.

    5. Re:The scene has changed. by TheNetAvenger · · Score: 2, Informative

      Maybe your eyes are better than mine, but I don't think we're even getting that.

      Texture size and number of objects in a scene on Crysis would be the best examples, there is a difference. Games are moving to levels (especially for HD or 1920x1200&up players) that the texture limitations of DX9.0c can't bring the detail needed, and this is just one 'tiny' aspect of DX10.

      http://www.tomsgames.com/us/2007/09/18/dx10_part3/page3.html

      Bundling it into Vista is bad, for a slew of reasons, and the shit they've pulled with several 'Vista only' titles,

      DX10 has specific reasons why it only runs on Vista. Go ask the people hacking the libraries for XP. They will run, but it expects the OS to be handling aspects of the GPU that XP isn't doing.

      http://arstechnica.com/journals/microsoft.ars/2007/2/14/7060

      DX10 is designed around Vista because it expects GPU RAM Virtualization to be available from the OS. (Only Vista can do this) DX10 expects even 'in-game' threads/processes to be prioritized and handled by the OS, and only Vista can do this because it has a pre-emptive scheduler for the GPU, XP don't (in fact no other OS has one). To put these things in XP would be to make a full WDDM for XP, and that is not quite so easy.

      The DX10 stuff like this is a tie over from the XBox360 development team, and DX10 is what MS and Robbie learned to take gaming forward on the PC.

      As for the 'Vista Only' titles, there were reasons for them at the time. For example Halo 2, as its online play is Games for Windows Live, and at the time used Vista's communication framework, and Live for Windows (the Gaming connection) was a Vista only technology. So the Halo 2 development went forward with these considerations, and other internal optimizations in the game just exepcted the Vista WDDM to be there, etc. Microsoft went back and wrote Live Games for Windows for XP from the ground up. (Hence some of the new networking features in XP SP3, just to support it.)

      So it may have seemed nefarious, but was not a con, just a platform specific feature and optimization design, pure and simple... Sadly MS was counting on NVidia and ATI to have their WDDM drivers at XP levels at release of VIsta, and this didn't happen. When MS jumped in with NVidia and ATI and 'helped' their driver development the fruit of this was seen around June 07, as Vista was catching to XP in gaming performance, and by Sept 07, had equaled it.

      I see absolutely nothing to recommend Vista over XP, at this time or in the near future.
      This is where Microsoft's marketing sucks. They should do like Apple and list every tiny feature.(Remember the 300 list about Leopard?) If Microsoft did a list like this for Vista, it would be around 10,000 items in their list.

      If I had time this morning, I could take your circumstances and make a very credible case for Vista. I also understand where you are coming from as Vista is a plumbing and architecture shift, they burned their time to build more features based on these changes with the iniitial dump of Longhorn. Windows 7 is basically going to be a showcase of what is already in Vista, since it doesn't have any major architecture changes planned.

      Hardly a year out of date. The figures you post are one month old, and involve Vista SP1 final, vs SP3 of XP. I admit I am impressed by the evening out that Vista has managed to achieve, in those tests.

      Ok, year was a bit of tongue in cheek.

      A lot of people didn't realize that NVidia and ATI had to write the Vista WDDM drivers from scratch, as it is a dramatic different model than XPDM. From letting Vista do scheduling to RAM virtualization and handing over more to the OS from core driver level to even Aero Composer.

      And even though I think NVidia and ATI could have done better at launch, as they didn't provide drivers to beta testers

  26. Does it sound like a jet engine? by alen · · Score: 2, Funny

    how loud is it and does it need the hoover dam to power it up?

    the way things are going you will need 2 power supplies in a PC. one for the video card and one for everything else

  27. Not fitting the narrative!!! by Anonymous Coward · · Score: 1, Funny

    The GTX 280 delivered real-world benchmark numbers nearly 50 percent faster than a single GeForce 9800 GTX running on Windows XP, and it was 23-percent faster than that card running on Vista.


    Please: We here at Slashdot are trying to catapult the propaganda that Vista is a failure. Your bringing in "facts" and "benchmarks" isn't helping.

    Will nobody think of the FUD?
    1. Re:Not fitting the narrative!!! by clampolo · · Score: 1

      It's not saying anything about Vista being good or bad. It is saying that the GTX280 on Vista will run 23% faster than a GeForce 9800 GTX running on Vista.

    2. Re:Not fitting the narrative!!! by Vectronic · · Score: 1

      Yes, but... if it runs 50% Faster on XP, and 23% Faster on Vista, thats saying something... im not sure if its saying that Vista, or DirectX10 is worse, or that the card, or its drivers/instructions are "worse" for Vista, or even that the game(s)/3d software they were using was more targeted at DX9 and Shader 2/3x instead of DX10 and Shader 4, or something.

      You'd think they would be aiming for something hat runs faster on Vista than XP, if they are trying to step forwards, because although XP still has the share, I wouldnt want it to stay there for much longer.

    3. Re:Not fitting the narrative!!! by thePowerOfGrayskull · · Score: 1

      Maybe the baseline is already faster on XP, thus the same increase yields a lower percent?

  28. Well, there goes my upgrade plan by kiehlster · · Score: 4, Insightful

    No wonder people say Console killed the PC game star -- "Alright, got my hardware list done. Time to order. Oh, look what just came out, guess I'll wait for prices to drop. Alright, they dropped! No wait, a new processor is out, think I'll wait. Sweet, think I can order now. No, nevermind, Crysis just came out, I'll have to wait until I can afford the current bleeding edge. Awesome, I can afford it now! No, a new GPU just came out that runs the game better. Oh, SATA 600 is coming out. Ah, forget this, I'm buying an Xbox."

    1. Re:Well, there goes my upgrade plan by aliens · · Score: 1

      And this has been the same story for how many years now?

      --
      -- taking over the world, we are.
    2. Re:Well, there goes my upgrade plan by Rhys · · Score: 1

      Those same people will sadly discover the consoles aren't much better. The PS3/Rock Band setup downstairs needs far-too-common patching (particularly the PS3) and that patching is usually horribly slow. Not to mention little nuggets like I don't think we can back up our Rock Band "saves", so if the hard drive dies, so does our save file.

      --
      Slashdot Patriotism: We Support our Dupes!
    3. Re:Well, there goes my upgrade plan by MindlessAutomata · · Score: 1

      Not to mention the peripherals themselves cost some money, too, and if a computer breaks, you usually just need to replace a component, and not send the whole thing back.

    4. Re:Well, there goes my upgrade plan by aarmenaa · · Score: 1

      The problem is that people have this perception that you absolutely need these super high end setups just to play games with. The reality is that most of these reviews are done at giant resolutions like 2560 * 1600 on 30" LCDs or 1920 * 1200 on 24" LCDs, which is unlikely to be a common setup. Valve's hardware survey says 3 out of 4 people still use non-widescreen monitors, with the most common sizes being 16", 17", or 19" monitors. Meaning most people probably don't game any higher than 1280 * 1024 or 1600*1200. For those 1 out of 4 users with widescreen monitors (probably LCDs), only about 35% of them run a monitor 24" or larger, so people with these huge monitors are really not quite the top 10% of respondents to the survey.

      This is about what I would expect: that top 10% or so is who these cards are for. Everyone else can get something that's cheaper and runs cooler, and be perfectly happy. These hardware enthusiast websites are for that top 10% - basing your purchasing decision on how well games run on a 30" LCD is kinda pointless when you still use 19" CRT.

      For a more realistic take on hardware, I offer my gaming system: a 1680 * 1050 22" LCD (which is large, but not near as huge as a 30"), an 8800GT video card, and an E6300 processor. None of it's what you would call high end - 22" monitors are about $200 on sale (and Dell ships them withs ome desktops now as well!), the video card is less than $150 right now, and that processor is so old that it's not sold anymore (but comparable things can be had for around $130). I've been perfectly happy with this system, and it runs games like a champ. The video card is new, but the old 7900GS that I was running was perfectly fine if you weren't interested in Crysis. Interestingly enough, I paid more for the 7900GS (I belive close to $180) than I did for the 8800GT, though both seem to have occupied similar market segments, which tells me that getting good gaming hardware is only getting cheaper, even if the high end cards get more outrageously expensive with every iteration.

      --
      "I do a grep for shit, bollocks, and tits before checking in code. I'm professional..." -RECURSIVE_META_JOKE, reddit.com
    5. Re:Well, there goes my upgrade plan by Kelbear · · Score: 1

      My geek friend spent 3 years timing the purchase of his "new PC". He is aware that new tech constantly comes out, so he just kept waiting for new tech releases to coincide in a short time span so that he could maximize the value between iterations. His computer became obsolete within 3 weeks.

      To be fair, he's an extreme case. He's also waiting for an Xbox360...waiting for the RRODs to be solved in the 3rd or 4th production generation which will address graphics cooling(unlike the last one which changed the heatsink on the the processor).

    6. Re:Well, there goes my upgrade plan by dunezone · · Score: 1

      Its so true. I just built my first computer in six years. Building it was ten times easier, buying my parts was ten times harder. I had to do so much research on parts. Keep the price reasonable but still maintain decent power. I was so hesitant to order because I knew that if I just wait another month I can get something even better then what I'm getting now, but it already been six months so I just bit my lip and ordered.

      My worst experience was researching video cards. This GTX200 series just popped out of nowhere, I don't know how I missed it. The 9800GX2 is a nice card but I don't think dual-core video cards are driver ready or cooling ready as these things heat up fast as sin and the drivers crash alot(something not mentioned in reviews). Idle temps are 50-60c, unless you speed up the fan, at 100% you can get mid 40's. And I'm not trolling either, I own this card.

      Another thing that really pissed me off was reviews. I didn't want to go SLI, I think its bullshit/legacy technology thats too expensive. But every review, every top videocard list, just mentioned SLI configurations and never reviewed just single cards.

      I mean consoles are still expensive when you start adding in all the extras to it. Mainly the extra controllers. But at least their designed to last 4-5 years. My computer will need a video card change in about two years to stay competitive. My Geforce 3 lasted about 1.5years before I had to switch to a 9700Pro.

    7. Re:Well, there goes my upgrade plan by DirePickle · · Score: 1

      "Not the fastest" != "obsolete."

    8. Re:Well, there goes my upgrade plan by Anonymous Coward · · Score: 0

      The heat-related RROD problem has been solved. Period. And they have a three year warranty on them now just in case.

      If he doesn't want to believe it, he probably didn't want an Xbox anyway. Which is fine, and one really doesn't have to make excuses for it.

    9. Re:Well, there goes my upgrade plan by WuphonsReach · · Score: 1

      For a more realistic take on hardware, I offer my gaming system: a 1680 * 1050 22" LCD (which is large, but not near as huge as a 30"), an 8800GT video card, and an E6300 processor.

      That's nearly identical to my gaming setup, except I went with (2) 8800 GT 512MB cards in SLI mode and I'm running an AMD Athlon64 X2 5600+. Those 8800GT 512MB cards are probably the best bang for the buck at the moment (mine are 6 months old). It's a good kit, definitely medium-upper range, but you can build one for about $1000-$1500.

      (I need a faster CPU... Oblivion is CPU-limited. *sigh* But I'm not sure I want to put a 125W beast into this box.)

      The 22" 1680x1050 displays are a great size. Not overly large (a bit wider then a standard keyboard), with good resolution and size without paying a bundle. Unlike the 19" 1680x1050 displays, the pixels are still large enough on the 22" that you won't be squinting.

      I'm sure I'll move up to the 24-25" 1900x1200 displays at some point...

      --
      Wolde you bothe eate your cake, and have your cake?
    10. Re:Well, there goes my upgrade plan by mjwx · · Score: 1

      FFS, PC gaming is far from dead, its been declared dead more times than Lazarus but PC gaming is still here. PC games are still selling, more console games are ported to PC than PC games to consoles, PC exclusives are still being made and latest generation consoles are beginning to resemble PC gaming from 4 years ago.

      PC's are more expensive but the Xbox 360 and PS3 are trying to close the gap, in addition to that component prices are dropping like a brick (a processor that would have cost me A$600 two years ago now costs me A$150, RAM that would have cost A$200 in 2005 now costs A$60) A State of the art gaming PC in 2004 would have cost A$3500, now days costs under A$2000 including a top end graphics card (A$700) not including a monitor as consoles don't come with a TV, in order to get HD gaming out of a console I need to drop A$1400 on a 42" TV (720p, try A$1700 for a generic brand 40" 1080i), a PC for HD gaming requires a A$300 (20", A$400 for 22" or A$550 for 24" which is full HD (1920) prices are for a Samsung monitor, not a cheap Chimei or Viewsonic) so in order to get the same effect from a console you need to pay almost the same amount on a Console and TV setup as you do on a PC/monitor setup.

      Now then there are the peripherals, I haven't had a controller last the entire lifetime of the console since my SNES, my PS2(A$70 at the time) and Xbox controllers (A$50 after the console was superseded) both broke on me even though I never used them that much, with a PC, peripherals are a lot cheaper and durable. I've been using the same MS Natural Keyboard for 6 years (A$40), I've got a MS basic optical mouse(A$20) for 5 years (peripherals are about the only thing MS has done right) although I bought a Logitech G5 laser mouse last year for gaming.

      In addition to this, at the end of its gaming life (2 Yr's for a dedicated gamer like me or 3-5 years for the more casual gamer) it can be put to use for other things (given to wife/GF/child so they stop using your gaming machine for emails, web server, media centre, cheap gift for extended family), after a console is superseded by the next generation of console (3-5 years) it at best has the job of a A$50 DVD player and I have a PS2 and Xbox (still, they are sitting in their boxes in the top of a cupboard) and am yet to see a PS2 or Xbox be able to do the job of DVD player as well as a cheap A$50 dedicated DVD player.

      BTW, Crysis on an AMD 6000+, 2 GB of RAM, Geforce 8800GTS, will run at 1024x1200 at 20-30 FPS just fine with the exception of particle effects which slows it down to between 9 and 15 FPS (24 FPS is the limit of the human eye to distinguish frames, most people cant see past 16 FPS and pretty much anything past 12 FPS is barely perceived as slow or jerky). Supreme Commander Forged Alliance (expansion pack), with the larger maps (81x81 KMs) starts to tax my system at around the 2 hour mark if I need to re-install Windows (flawless performance on a fresh install) at 1024x1280. Less resource intensive games like Galactic Civilizations II, Bioshock, COD4, S.T.A.L.K.E.R will run at 1600x1200 without a problem.

      For the record I have a gaming PC and a Wii, they are for two separate purposes, the Wii is great for casual games and gaming with a bunch of friends. IMO, this is the way consoles are going, theres more money in it by targeting the casual audience who doesn't want to think about their games. When I want to spend hours playing a good strategy or FPS game or want to play online against 20 or 30 other people I use my PC, the PC will regain the hardcore gaming crown as Microsoft and Sony have both learned from Nintendo's success at making a more entertaining and affordable console.

      --
      Calling someone a "hater" only means you can not rationally rebut their argument.
    11. Re:Well, there goes my upgrade plan by mgblst · · Score: 1

      Just as you don't need to buy every game and accessory for your console, PC games don't need to buy every new technology for their PC.

      The ability to upgrade your PC is actually a huge advantage, especially when you consider that 5 years after your console was released, you will be runnning on very old hardware.

    12. Re:Well, there goes my upgrade plan by Anonymous Coward · · Score: 0

      The heat-related RROD problem has been solved. Period.
      Which clearly explains why I was dropping off my 10 month old 360 at UPS this morning. Destination: Microsoft's Xbox repair center.
  29. Research before you spread FUD. by ericvids · · Score: 3, Informative
    From the Tech Report:

    TR: Does the removal of this "render pass during post-effect" in the DX10.1 have an impact on image quality in the game?

    Beauchemin: With DirectX 10.1, we are able to re-use an existing buffer to render the post-effects instead of having to render it again with different attributes. However, with the implementation of the retail version, we found a problem that caused the post-effects to fail to render properly.

    TR: What specific factors led to DX10.1 support's removal in patch 1?

    Beauchemin: Our DX10.1 implementation was not properly done and we didn't want the users with Vista SP1 and DX10.1-enabled cards to have a bad gaming experience.

    --
    Pet peeve: Profane people propagating perfunctory pedantry.
    1. Re:Research before you spread FUD. by Anonymous Coward · · Score: 0

      If by "bad gaming experience" you mean running the game significantly faster under DirectX 10.1 on ATI cards - I'll live with the supposed bad gaming experience.

      See Anandtech - NVIDIA's Dirty Dealing with DX10.1 and How GT200 Doesn't Support it

      and TechReport's GTX 280 review showing a Radeon HD 3870 X2 with top score for the original DirectX 10.1 release of Assassin's Creed at 4X AA - 2560x1600.

    2. Re:Research before you spread FUD. by ericvids · · Score: 1

      No, by "bad gaming experience" I mean running the game with inaccurate visual effects. I, for one, would prefer running the non-10.1 version just to see ALL of the effects properly, even if it costs me a few FPS.

      Your own TechReport link only states that the original DirectX 10.1 implementation is faster on the Radeon than it is on the Nvidia by 3fps on average, but you are forgetting that the game is not rendering the exact same thing across both cards in this way. (Oh, and note that the median low score of the Nvidia card is actually faster. That matters a lot because that means the card is still faster during parts of the game with incredible slowdown, regardless of whether DX10.1 was available on the Radeon or not.

      Ubisoft itself has already stated that it was a bug (and a very legitimate one), and it was THEIR bug to fix. You might complain "oh it's a very minor bug, there was no reason to remove DX10.1 support altogether", but if you're like me (I'm a graphics programmer myself), you'd understand the need to make all of the visual effects accurate, and apparently the only USE of the DX10.1 support in Assassin's Creed was for this one (buggy) part of the game, so Ubisoft was right about disabling 10.1 support on their game. Those people who are claiming "Nvidia's just dirty dealing so Assassin's Creed would run slower on Radeon's" are just blatantly uninformed, or probably just biased to begin with. (Oh, and might I point out that your "dirty dealing" article does not mention anything about Ubisoft's alleged deal with Nvidia either; you're just grasping at straws.)

      --
      Pet peeve: Profane people propagating perfunctory pedantry.
  30. Re:Great news - not that I want to buy the thing.. by cptnapalm · · Score: 1

    not to mention that the games that are still played a few years after release will look absolutely fantastic.

  31. The Voice of Reason by Einstein_101 · · Score: 1

    First of all, I would like to remind everyone that benchmarks are subjective. I know this may come as a shock to you, but just because someone types with perfect grammer, has a pretty banner and a nice website layout that doesn't mean they aren't immune to bias or hyperbole. Even further, sometimes their intentions are honest, but their results just aren't typical

    A perfect example of this was the ongoing debate about the 9600GT vs 8800GT vs HD 3870. Go through about 5-6 different reviews and you'll realize that there clearly isn't a winner in the fight, yet some websites won't hesitate to call the HD 3870 inferior, or try to pidgeon hole the Radeon card as being a "better value" - overlooking the fact that it outperformed the NVidia card on some games, while using less power and running cooler. The fact of the matter is all 3 of the cards perform better that the other 2 in at least 1-2 situations, but some people are just plain reluctant claim an ATi card is better than it's NVidia counterpart (due to it having almost the same performace with lower power consumption).

    It's all subjective people. Benchmarks have never been an exact science anyways.

    (ps: You all aren't kids anymore. Google this stuff for yourself and save me some time. I'm at work too you know :oP)

    1. Re:The Voice of Reason by DragonTHC · · Score: 1

      of course they're subjective.

      But there are a great many of us who can translate benchmark scores into real world performance.

      all it takes is a comparison of those scores to your own scores to know what type of performance to expect.

      I was considering upgrading to a 9800GX2, but given these scores, I think I will wait. My 8800GTS 640MB is performing fairly well.

      --
      They're using their grammar skills there.
    2. Re:The Voice of Reason by Einstein_101 · · Score: 1

      Then I wasn't talking to you.

      Granted, there are a great many of you who can translate benchmarks into real world performance, but you and I both know that your kind is in the minority. Like 10% of all gamers minority. The rest will either live and die by reviews and benchmarks (with no independant thought of their own)), or acknowledge that there is more to gameplay and performace than benchmarks - and just leave things alone at the acknowledgement.

      I fall into that second category, and even I am in the minority.

    3. Re:The Voice of Reason by MindlessAutomata · · Score: 1

      Benchmarks are objective. A card will perform one way on a game given everything being equal on the set up (and ignoring possible variations between cards and other hardware) regardless of what or who runs the test.

      The end conclusions, however, are what is subjective--that X is a better buy, or so forth, are what is subjective, although statements like "Card A performed significantly faster in all tests" are objectively true statements.

  32. What about Aero graphics? by Joce640k · · Score: 1

    Aero graphics must surely be bad for the environment - it prevents most of the GPU from powering down.

    --
    No sig today...
    1. Re:What about Aero graphics? by Petersson · · Score: 1

      Aero graphics must surely be bad for the environment - it prevents most of the GPU from powering down.

      Windows 9x was also bad for environment - it didn't send IDLE instruction while doing nothing, so the CPU power consumption was still at max.

      This guy http://estu.nit.ac.jp/~e982457/other/cpuidle/idle.htm has done some measurements proving how silly OS design increase power draw. 15 Watts difference doesn't look much, but just multiply it by number of Win 9x installation.

      Windows NT were already using the IDLE command, so what was reason for not implementing it to Win9x, while some freely availabe utilites (Waterfall, Rain, CPUidle) did the job? Shame on you, Microsoft, again.

      --
      I'm not insane. My mother had me tested.
  33. Don't worry.... by Totenglocke · · Score: 2, Funny

    They still have 10 more years to develop video cards before Duke Nukem Forever comes out!

    --
    "The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants." ~Thomas Jefferson
  34. Crysis? by steveaustin1971 · · Score: 1, Flamebait

    From the look of the benckmarks, neither the SLI 9800's nor the new card tame crysis at all. You can't make a poorly coded game fluid with more power I guess.

  35. And it uses more power than your A/C by mario_grgic · · Score: 1

    and it's probably noisier too :D.

    --
    As the island of our knowledge grows, so does the shore of our ignorance.
  36. Re:Performance crown my butt by Applekid · · Score: 1

    in case you are wondering, whether i am a fanboi or not - i dislike being fanboi of anything. but when i see some party pulling bullshit, i side with the other. I don't know much about 10.1, but if what you say is true, your beef is probably with Ubisoft for actually doing it, not NVidia for requesting it.
    --
    More Twoson than Cupertino
  37. I have to agree by default+luser · · Score: 1

    I've been holding out this generation, holding onto my 7900GT. I like the GT because it delivers solid performance for only 60w, which is half the power ATI's x1900 series was offering at the time. I've also been able to stall because games like Team Fortress 2 and Quakewars ET still look great on my current card.

    Only now with the release of the 8800GT and 9600GT is the power consumption/performance ratio getting reasonable (and yes, the ATI 3870 has similar power consumption to the 8800GT, but cannot match it in performance). I'm actually enticed by the 55nm 9800GT (due out in July), which should cut the power consumption of the 8800GT to the same as my 7900GT :)

    The only other card I'm considering is the 4850. Only time will tell if it delivers better performance than the 8800GT without breaking my power budget.

    --

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

  38. When Vista is faster there's 'something wrong'? by Joce640k · · Score: 1

    Mod parent "informative". Thanks for the info.

    --
    No sig today...
  39. I doubt it's noisier than MY A/C by Joce640k · · Score: 1

    Kid today don't know they're born.

    I sat for a year next to a Silicon Graphics twin-tower GTX. Now THAT was a noisy machine. Any GeForce is whisper-quiet compared to that.

    --
    No sig today...
  40. Impressive. But impractical. by fuzzyfuzzyfungus · · Score: 3, Insightful

    The GTX280 part looks quite powerful; but its die size is really extreme. Anandtech claims that maximum best case yield for the 280 is 105 chips per 300mm wafer. TFA notes that Intel can put 6 dual core Penryns in the same space These guys quote just under $3,400 for a single 300mm wafer. So, assuming absolutely optimal yield, the GTX280's core costs ~ $300 to manufacture, not counting R&D, packaging, distribution, etc. A gigabyte of RAM suitable for a high end graphics card (read, not 10 dollars worth of DDR2) adds some more, and the board, passives, and assorted other logic do as well.

    Obviously, the above numbers are wild speculation; but the punchline is that these parts can't possibly be cheap to manufacture. I suspect that NVIDIA will see some nice sales to lunatic early adopters, and they'll probably have a compute only version of this card for high end computing; but there is no way that it could hit mass distribution price points. Even at $650, I'm not sure that NVIDIA's margins are all that exciting on this particular part.

    1. Re:Impressive. But impractical. by PIBM · · Score: 1

      3400 / 105 = ~300 ?

      Which system is this approximation based upon ?

    2. Re:Impressive. But impractical. by ThreeGigs · · Score: 1

      but there is no way that it could hit mass distribution price points

      I'm not so sure of that. Considering it's not unusual for a supercomputer these days to have 5000 or more processing units (Roadrunner has 6000+ Opterons and 12,000+ Cells), I can see a large demand for the 280 part in high performance computing applications. I'm guessing nVidia sees things the same way, which is why we're seeing so many non-graphics specific improvements and tweaks to the GPU. I'm thinking nVidia is counting on leveraging CUDA based supercomputing apps to sell a lot more of these than they would if it were just gamers buying them. The double-precision floating point capability alone signifies they're targeting a non-gaming market. With 8-core (4 way SLI dual core) units having made the front page of /. recently, and things like protein folding apps being ported to GPUs, I'm thinking using GPUs for massively parallel computing is becoming much more mainstream, as folks realize just what kind of bang for the buck they can get from GPUs. That, I think, translates into enough demand to push the 280 into mass-manufacture and distribution price points. I just hope the demand from well funded organizations buying these things for their supercomputers doesn't keep the demand so high that we're all bidding on eBay for the occasional part that isn't slated for a supercomputer!

    3. Re:Impressive. But impractical. by mako1138 · · Score: 1

      The Inquirer (yeah, I know, hold your nose) has more believable numbers: 40% yield, $5000 per wafer, 40 good dice. So $125/die. Add NRE, testing, etc.

      When there's so many stream processors on a chip, it's going to be big, no doubt. That's why a 55nm version is coming. Cell also underwent a die shrink shortly after it came out.

      I've been itching to upgrade, but I guess I'll be waiting some more.

  41. It's official by marco.antonio.costa · · Score: 0, Flamebait

    Vista is a slow, crawly, disgusting vomited animal. And people are buying that slow crap from microsoft as it's the second coming of christ.

    --
    Send your spendthrift head of state this
  42. We're gonna need CUDA benchmarks by schwaang · · Score: 2, Insightful

    More and more these commodity graphics cards are being used for non-graphical high speed computing by taking advantage of the insane parallelism of the GPUs.

    Someone please develop CUDA benchmarks to be included in future reviews.

    We need several apps: one with a kernel that is trivial enough to be constantly starved for memory, one that is the opposite (compute heavy, memory light), integer vs. FP, and something that specifically benefits from the new double-precision floating point that only the newer stuff has.

    Get back to me soon, mmmmK?

    1. Re:We're gonna need CUDA benchmarks by LotsOfPhil · · Score: 1

      The CUDA SDK includes dozens (maybe more than 100) of sample programs. Use these.

      --
      This post climbed Mt. Washington.
  43. Re:Power vs Intel. DirectX problem? Use OpenGL!! by Anonymous Coward · · Score: 0

    That is DirectX fault, not hardware vendor's!

    And why that doesn't happen with OpenGL? If you take the latest bleeding edge graphics card and play the first version of Serious SAM on it (which requires a very old OpenGL version), it will run perfectly.

  44. Real-time idle for video cards? by antdude · · Score: 1

    I would like to see video cards slow down clock cycles like CPUs (e.g., AMD Cool'n'Quiet). I don't always play games, 3D stuff, etc. Most of the time, it is surfing the Web, e-mails, newsgroups, watch videos and Flash (might need video card's accerelation for this so speed it up), etc.

    --
    Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
  45. At 255 watts, it better be faster by Anonymous Coward · · Score: 0

    That's a space heater by any definition.

  46. Re:Power vs Intel. DirectX problem? Use OpenGL!! by p0tat03 · · Score: 1, Flamebait

    Blah blah, more trolling. Try developing games or visualization apps sometime. You will come to appreciate how ridiculously superior (albeit closed) DirectX is compared to OpenGL. Open source apologists tend to defend OpenGL's shittiness despite all evidence to the contrary, but from an objective, developer-ease point of view, there's simply no comparison.

    There are also plenty of cards that ship with crappy OpenGL drivers, increasingly moreso thanks to the decline in popularity of OpenGL. DirectX is fairly well defined, and is one of those few products where I honestly believe that MS won the market share fair and square.

  47. Nvidia platform by Latinhypercube · · Score: 1

    Nvidia should write an os for their chips, and bang them into a custom nvidia board, an Nvidia only platform. It'd be the most massively parallel computing capable system out there I'm sure. Beats the trouble most exiting cpu based systems seem to be having moving into multi-threading etc.

  48. So. by Anonymous Coward · · Score: 0

    I figure by the time my kids are looking at porn the the graphics will be life like and there will be no need to have real sex.

  49. Noise levels? by DGolden · · Score: 1

    Given the amount of power these things apparently need, are they going to be whirring monstrosities too?

    I like having a quiet machine (though my current quiet PSU probably wouldn't be enough for more than one of these things, yeesh...).

    Can you / do you need to fit third-party quiet fans to these enormous-and-apparently-enclosed-form-factor cards?

    --
    Choice of masters is not freedom.
  50. Bullshit, 30fps XP, 8fps Vista. by Anonymous Coward · · Score: 0

    From the fine article:

    our engineering sample delivered the game at more than 30 frames per second running DirectX 9. Games still run slower on Vista, however; Crysis, for example, shed about eight frames per second running DirectX 10,

    Posting AC to avoid you and your shill accounts. Vista sucks and so do you M$ astroturfers.

  51. 9800GX2 and GTX 280 by Coax123 · · Score: 1

    My current pc: Vista 64 bit premium evga 790i ultra q9450 @ 3.3 GHz 4 GB DDR3 @ 1800MHz evga 9800 GX2 2 power supplies: PSU 650W (powering SATA drives and MOBO) Graphics PSU 800W (fits into cd bay and can power up to 4 cards or two 9800GX2's) I'm not sure what everyone is bitching about but if you don't own or have personally tested a 9800GX2 or a gtx 280 then, like others have posted, do not trust some web sites review. Yes, some are good, but lots are biased. That's not what this post is about though.... My 9800GX2 is great; I've yet to come across a game that cannot handle all high settings. I run Age of Conan at all high with 16xq anti alias with 50-90+ fps. I don't base my graphics card performance on reading other editors/companies benchmark tests. I base it on actually owning it and applying hands on gaming and testing it with the particular games and programs that I use. (AoC, CoD, etc etc, & programs like SolidWorks) I AM UPGRADING to the gtx 280. Yes it's an upgrade. It's a better/newer card. However, I cannot fully vouch for it until I receive it. When I get it (aprox 1 week) I will let you all know how it compares to my current set up. I'm confident that it will perform better, and I KNOW that it will perform better in dual SLI than a 9800GX2 will in dual SLI (2 cards - i.e. 'quad sli'). Also, the 280 will allow me to run true triple sli if needed, but is probably unnecessary atm. As for people complaining about fan noise- 9800GX2 is very quiet, I was surprised since it's keeping to GPU's cool. As an engineer and computer builder (hobby) I find it humorous as to how many people waste their time nagging about things they don't and probably will never own. Go buy one and try it out for yourselves, it's truly and amazing card. Ps. The 280 will be tested by me when I get it in a week. I do not care to give it a great review or a bad review. I am simply going to be comparing it to the 9800GX2. Out of curiosity I would then like to test it against the best ATI card that will be out in about 2-3 weeks. Any suggestions on that please let me know. -thanks