Slashdot Mirror
Preview of The GeForce 256
Tor Arne writes " Gamecenter is doing a preview of the Nvidia GeForce 256. Not as amazing as it initially sounded, it's still pretty amazing looking."
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
If you can get better performance at a lower clock speed, that is much better than running it fast. Less heat, less stress on the silicon and there's a better possibility that the chip CAN go faster. I'd love it if I could get the performance of an athlon 600 out of a chip that runs at 100, even though it doesn't sound as sexy.
-----
Do you even know anything about perl? -- AC Replying to Tom Christiansen post.
Some games can take more advantage of T&L than other games - Unreal stresses fill rate while Q3 Arena will benefit more from T&L.
m l
:-)
Voodooextreme has asked a lot of game developers, what they think of this hole issue about T&L vs fill rate - you can find the article here: http://www.voodooextreme.com/articles/fillvstl.ht
I like the first comment from Tim Sweeney - Epic Games
A lot of the benchmarks that has been published doesn't take advantage of the T&L, and therefore the benchmarks doesn't look really great, just great. But the fill rate of the GeForce isn't *that* much better than the TNT2 Ultra or the Voodoo3 3500.
What you can't se from the benchmarks either, is the picture quality - with games that uses T&L, you might not get a frame rate that is much higher than others, but you'll get a much nicer picture.
You'll need to get games that stresses the T&L chip to see the difference, and there are not many games which does that today (are there any at all?!)
Download the tree demo from NVIDIA's website and run it on you 3D-accelerators - it crawls!
I tried it on my Celeron450/128MB RAM/Voodoo3 2000 - it was a slideshow!
Besides, Geforce is the only next generation card which is available in the next couple of weeks - S3 Savage2000 will be available before christmas, but that's a long time in the graphics bussiness. It is even worse with 3dfx's Voodoo4/Napalm - it will maybe not be available before february!
If NVIDIA continue to deliver a new product every 6 month, then the will have their next generation card ready a few month after Voodoo4 arrives.
Rumors about NVIDIA's next card/chip/GPU will certainly be all around the net at that time, which may hurt 3dfx's sales, if they don't deliver something quite extraordinary...
> Hardware T&L greatly increases the amount of onboard memory needed.
.. I dunno, between 64 and 128 bytes (you need your texture indices too, remember)), you're using between one and two GB/s of bus bandwidth if all you're doing is reading each polygon once.
No facts to back this claim up. How exactly does hardware T&L increase the amount of onboard framebuffer required? With AGP, there really is no need for local video memory at all, except to use for the actual visual screen, and maybe as a texture cache. Sure the geometry system will need somewhere to cache scenes, but to fill up 128MB with just _geometry_ information you'll need something as complicated as that huge landscape scene in the Matrix.
Certainly hardware T&L does not increase the size of the framebuffer needed. However, AGP is really not as fast as the RAM they're putting in these systems - hell, all bus issues aside, system RAM is only 100 MHz, while most video cards local memory is way faster, or on a wider bus or both.
When doing the geometry, you don't want to tie up your bus to read and write and read each vertex as you translate and light each frame. Sure, it is possible to do it with AGP, but is it efficient? Let's see, they say it can push 15 million polys a second? Say a poly takes up
If this is 60 fps, each of those frames is 16-32 MB. 128 MB will be more than most applications will need. But using an extra 32 for geometry information is not unwarranted, in their pushing-the-card-to-its-limits case.
Disclaimer: I'm not as smart as I think.
Trees can't go dancing
So do them a big favor
Pretend dancing stinks!
The deal with 3dfx is that they mean to be able to provide over _100_ fps, or 60fps with antialiasing. This is different from motion blur- for one, the antialiasing will work with all old games as it's nothing but a scaled-up screen bilinearly resampled down. Antialiasing does look good (it's widely used in raytracing) and this will indeed cause existing games to look better, as well as the future games. ;)
GeForce will not be able to do this as it is grossly fill-rate impeded compared to its competitors. GeForce is all geometry and no fill rate- the next 3dfx thing is all fill rate and no geometry- the Savage one is somewhere in the middle.
The only way you'll get the antialiasing and motion blur ('cinematic' effects, kind of like how 3dfx rendering seems dirtier, more contrasty, more photographic as opposed to 'rendered') is with the 3dfx stuff as none of its competitors are willing to put that much effort into fill rate. The only way you'll get 20 times the geometry (rounded curves, 3d trees in games etc.) is if you get the GeForce and also wait to have developers write games for it, many of which could be Win-only *grumble*. My money's on 3dfx actually- I'm biased because I always think 3dfx screenshots look more 'photographic' (grain? contrast? some factor of their 22-bit internal calculations to 16-bit display?) but there's another factor- competitiveness.
If you read folks like Thresh talking about what they use, it turns out that they crank everything down to look as ugly as possible and run as fast as possible. I've done this on q3test and got a solid 60fps in medium action using only a 300Mhz G3 upgrade card and a Voodoo2. It looks awful, especially when you really pull out all the stops and make things look absolutely horrible- but it's sure competitive! You can track enemies and gib them much better, even if you're not all that hot at Quake.
How does this relate to the GeForce? It's the fill rate. Even on a normal AGP bus the thing can't be fed enough geometry to max it out- but the actual filling of the screen is unusually slow, and this expands rapidly with larger resolutions.
The result is this- somebody trying to max out, say, q3test but run at 1600x1200 in lowest image quality will be able to see accurate (but nearly solid color!) enemies in the distance and be able to make out subtle movements. This also applies to the antialiasing- that will help as well, even at normal resolutions. The result is that the person running on something with insanely high fill rate and using that combined with very low graphics quality, will get more visual information than the other players will, and will be getting it at a frame rate that is competitive (to a Thresh, there's a difference between 100 and 150 fps- while in a crowded fight, with 'sync' turned off).
By contrast, users of a geometry enhanced card will not get a competitive advantage from their form of graphical superiority. It is strictly visual eye candy and will not significantly add a competitive advantage...
For that reason I'd say, DON'T write off 3dfx just yet. Their choice for technological advancement is tailor made for getting a competitive advantage, and when you start maxing out the respective techie wonderfulness, the competitive advantage of 3dfx's approach will not be subtle. Likely result- 3dfx users may not be looking at comparably pretty visuals, but can console themselves by gibbing everybody in sight
--
?
I think this card will perform as advertised. Unfortunately, software must be written specifically to take advantage of the hardware and there is no way to test this at the moment. Right now, without taking advantage of the T&L engine it is the fastest "conventional" video board. That says quite a bit.
What will be interesting in the future, will be the test results of the Voodoo4 (or whatever it will be called). This card should be able to do more with a lower framerate due to it's support of motion blur. The corrollary that 3Dfx is making with the V4 is between film which runs at 24fpx and todays video games which must run at 60fps. I think that most people would agree that film looks very, very good. If what 3Dfx says about their board is true, then FPS will no longer be a suitable judge for a boards performance. Assuming that a game is written to take advantage of motion blur. The downside... games depend on reaction to a controller and must be able to display these small changes. If the game is only updating at 24fps, then you may feel as if you don't have precise control over the game (but it will damn well look good!!).
It will be interesting to see how the 6 to 9 months of graphics card pan out. One thing is for certain though, by the time the PSX2 ships in North America, the PC should be well beyond it visually.*
*Of course the world is ending on Y2k, so these are hypothetical hardware progression estimates.
It may be too early to tell performance issues until the product actually hits the streets. I recall Nvidia downplaying many of these "previews" because many sites are using outdated drivers. Direct X 7 is susposed to take advantage of T&L and Open GL already does so for years. So out of the box it'll be a great card for 3d applications like maya, lightwave, etc. And all the open gl game games out on the market [except for GL Quake1]. Apparently this started with a preview of a creative GeForce256 card from an asian review site which pegged its performance not that much faster then a TNT2... I guess it depends on what tests.. I find alot of the benchmarks pretty useless. If it runs quake3 good it should run most games good. As for applications.. heck if I get any speed performance increase in Maya over my Viper 770.. i'm sold =).
More importantly, a fixation on clock speed is quite silly, as it is merely one of the factors in how fast a system is. I'm rather more impressed if they produce a faster product that doesn't have as fast a clockspeed.
Furthermore, keeping the clock speed down has other merits such as that it likely reduces the need for cooling, as well as diminishing the likelihood of the chips being stressed into generating EMI.
(Entertaining rumor has it that 900MHz systems that are likely coming in the next year may interfere with the 900MHz band used by recent digital cordless phones...)
If you're not part of the solution, you're part of the precipitate.
Does any one know if the Linux drivers for this thing (which have been promised) will be day and date with the cards themselves?
And when will ID start supporting Q3A on the nVidia cards? I HATE rebooting into Windows!
I want one of these SO BAD, but I'm not going to give up XF86 for it!
-- IANAEG - I am not an elder god.
However, games will need to be specially written to take advantage of this geometry acceleration.
This is only true if you were unfortunate enough to write your game using Direct3D. OpenGL games will be able to take advantage of geometry acceleration without even recompiling. You reap what you sow when you use a Microsoft API.
Whether or not hardware T&L is of any benefit to current or future games is yet to be seen though. Games lately have been getting more and more fillrate bound and less geometry bound, as game creators take advantage of higher resolutions and larger textures.
The GeForce, on the other hand, supports up to
128MB of local graphics memory. Hardware T&L greatly increases the amount of onboard memory needed. The first boards aimed at consumers should come out at 32MB, with 64 MB and 128MB cards to follow later on.
No facts to back this claim up. How exactly does hardware T&L increase the amount of onboard framebuffer required? With AGP, there really is no need for local video memory at all, except to use for the actual visual screen, and maybe as a texture cache. Sure the geometry system will need somewhere to cache scenes, but to fill up 128MB with just _geometry_ information you'll need something as complicated as that huge landscape scene in the Matrix.
Texture compression allows the use of much more detailed textures without overburdening graphics memory or bus bandwidth.
My jury's still out on texture compression. For games that are poorly written (i.e. that load and release textures on the fly, each frame) compression can help, but for games that use a more intellegent caching scheme for texturing, there really isnt much of a point.
Like the TNT2, GeForce supports the AGP 4X standard.
Definitely "A Good Thing".
The GeForce also introduces a new feature, cube environment mapping, that allows for more realistic, real-time reflections in games.
Similar to the Matrox G400's env mapped bump mapping but not quite the same.
Other things to note: 4 texel pipes (fills at four times the clock rate). Watch for all the other chip makers to do this too, limit of 8 lights in hardware (what happens when a scene requires more than eight? They don't say.. hmmm......)
Basically nVidia is gambling with hardware geometry. The gamble is, that future host cpu's (Pentium-4's or whatever) will not be able to beat them in doing transformation and lighting, and that if they don't, gamers are going to really even benefit from T&L. We'll see if that pans out. Unless they have a very sophisticated ALU on that chip, it will doubtlessly only speed up certain types of scenes. (We've all seen the "tree" demo).
You can play Q3A on nVidia. Check out nvidia's linux FAQ. It's got links to the drivers, and instructions for Q2/Q3. Yes, they all say it can't be done, etc etc, but believe me, I run Q3test on a TNT2 all the time, it works fine. It's just not officially supported. Have fun! :-)
----
We all take pink lemonade for granted.
There is no K5 cabal.
I am not the real rusty.
I would like to see some day a cheap 3D card supporting windowed 3D acceleration with full OpenGL support. Nothing else...
from the myth-ii-will-look-so-nice dept.
Unfortunatly, it won't, as Myth II only supports hardware acceleration on 3dfx cards via the glide port; OpenGL support is not even planned. To quote briareos, a Loki developer on loki.games.myth2:
You're really asking "will we take the time to write an OpenGL
rendering module for Myth2"? The answer is: if someone finds the
time. That's all I can really say.
The idea of Hardware T&L and texture compression are nice improvements but as has been repeated many times for many hardware additions (MMX, 3dNow, the new IBM crypo chip) software (read apps) has to be written to take advantage of these new features. I would much rather see a CPU with enough FPU power and cache to handle the software T&L as it exists now in games. If film runs at ~24FPS what makes it look as good as a game running at 40-60fps? In my opinion it comes from the frame rate drops and stutters that occure when rendering scenes or perspective changes. When in Half-Life and I walk out of a hall to an open area my TNT drops from about 20-25 fps to 12-17 fps. Throw in a couple of light sources in any rendering and your slowing the whole thing down more.
Why does this matter you ask, well, your eyes can definately see the frame rate jumping up and down, even if you see no major difference in the smoothness of the animation. I would rather have any type of T&L as well as buss and fill rates that allowed me to push a steady 25-35 FPS in what ever game/app was rendering in the resolution i wanted. If it pushed 70-80 and dropped to 30 on a tough scene it would not matter much as I would cap my frame rate at about 35-40. Then my fps stays a STEADY 25-40 and doesn't drop to an un acceptale rate, also the app code doesnt get bottenecked when the frames push super high. I dont want 300 fps at 1280x1024, i just want ROCK SOLID frames between 25-40 when rendering ANY scene. Of course Having hardware T&L and for pipelines is nice for being able to do more detailed goemetry and faster as long as the software offloads its T&L to the hardware. However I think that we are getting pretty damn close to the maximum detail level thats needed in games. We can use some more but not a whole lot. The most important thing is that we can dot it at a steady rate.
Flame Away!!
www.mp3.com/Undocumented
You look at it's scrores compared to the TNT2 and see only an 8fps increase in certain tests, remember those 8 extra frames are another 6 million texels. You also have to take into account that you're not utilizingf all aspects of the chipset, the T&L on hardware isn't being utilized and neither is it's control of the scene so you can't really compare apples to oranges here. It's actually like the P3 and Apple's G4, at the same clock speed as the old chips running the same software they are only marginally faster but when you actually use their features to their fullest you have a much faster result.
As for being a transition technology thats exactly what I think it is, soon you'll see S3 and 3Dfx do something similar if not better, then nVidia will come out with a more powerful GeForce and so on and so forth. One area I really think this kind of technology will do alot of good is in the console market. If you look at the N64 and Dreamcast they both have a super fast CPU (relatively) and then a powerful graphics chip for the actual rendering, theres not alot of technological different between the two besides word size and the number of transitors. On the the other hand if you used a technology like GeForce in a console you'd have a much more versatile machine that would be cheaper to manufacture. Your CPU handles the game code and does the physics calculations using a standardized chip that can preform just about any task you assign it and then your graphics card uses a part of it's chip for scene control and another part for lighting and textures and then a final part for the actual rendering of a scene. Each job is done on a specialized processor on the chip which means it can be done faster and more efficiently than can be done on a general purpose chip. This means consoles can more easily run complex code and physics because the processor isn't as tied up with the graphics processing not to mention run application style programs with heavy graphical content without a dip in performance. This would give future consoles more leverage when it comes down to a choice between a full fledged PC or a console that has much of the funtionality but less hassle.
I'm a loner Dottie, a Rebel.