Slashdot Mirror
3dfx Unveils Info Regarding Voodoo 4 & 5
A reader wrote to us about the latest press release from 3dfx regarding the Voodoo 4 and 5.
The V4 and V5 will apparently be released in March of 2000. The V4 will be
single processor, but the V5 will have both a commercial and professional
version, respectively supporting up to 4 and up to 32 VSA-100 processors, and up to 128 and 2GB of RAM
each. The release for the V4 and V5 is rolled in with the VSA-100 talks - definitely worth checking out.
3dfx sux, first post (yet again) -- I suck worse.
I'm posting for this one, hope it'll be replicated on the other :)
Sweet moses! 2GB of RAM! Is that right!? hehehe
For the consumer market, products based on the VSA-100 deliver from 333 megatexels/megapixels per second up to 1.47 gigatexels/gigapixels per second fill rates using 16-128 MB of video memory and one to four processors per board.
This has got to be the greatest bit of kit I've seen in a long time!
Powerful? Without question!
Linux support? I sincerly hope so!
first
we'll see when this board's released if it truly has these specs. plus, don't forget about nvidia, they have released new hardware so quickly this past year, i wouldn't put it past them to have a voodoo4/5 killer out by that time (or before). the geforce shows promise, but we'll have to wait until software is coded for these chips in order to see it perform in all of its glory.
-
This Post has been brought to you by the letter "E".
Now look what you have done! You scared the other one away.
I can't figure out why everyone is so happy about 3dfx putting out another Voodoo chip. They're pushing a proprietary interface (Glide), where a perfectly good standard exists instead (OpenGL). They're using market pressure to get game manufacturers to adopt their standard, and lawsuits against developers who try to write Glide wrappers so that Glide-only games can be played on other video cards.
Doesn't this sound a bit like another company that everyone is up in arms about?
- Drew
- In Capitalist America, law violates YOU!
- Linux support?
- This is just another trick by Microsoft
- Wow, I'd like to see a Beowulf cluster of these..
Personally, I think that this is great... let's hope iD is keeping up and giving us RealGuts(tm) in Quake 4.Remember how the cheapo motherboards used to be able to allocate some of the system RAM for video RAM? It would be pretty funny if these cards could do the opposite.
--
I've heard of SMP computers, but SMP video cards?
Looks like we'll have to all compile our kernels for SMP machines now.
--
Lab test show that use of micro$oft causes deadly cancer in lab animals.
But I do like a sense of humor. It's nice to see 3dfx sticking to the Glide standard rather than some proprietary OpenGL nonsense that doesn't port anywhere. Every tried to run a TNT in anything except Windows? The 3D works like shit. Sure, Glide might be slower in Linux/FreeBSD than Windows, but it is a simple, fast, and more efficient protocol than the crap that nvidia is turning out anyday. But don't fret nvidia users, you own a very nice 32MB card that can do wonders on a 2D desktop...look at X go!
3Dfx unveils good number for his new 3d architecture...but It lacks of geometric acceleration. Do you still want 200 fps at 1600x1200 32 bit with 3 big polys at each frame? :) No thanks, I prefer lower fill rate and higher polys counter.
This thing is incredible. My jaw dropped to the floor and didn't stop till it went down a few floors. This is the first time 3dfx will put something out that is good not only for gaming. They are using a completly new architecture this time. It's about time I was wondering when the kings of 3d would retake their crown.
It has been statistically shown that helmets increase the risk of head injury.
I can't wait to see how this card works. It has to be better that Nvidia's unsupported T&L, Matrox's unsupported Enviroment Bump Mapping and the now ancient Riva TNT2 ultra. I doubt Nvidia will have anything out by then that can touch this.
Tom's Hardware noted that the new parallel-processing video card from ATI, the Rage MAXX, has to wait 2 frames to accept a user input, as opposed to 1 for most single-chip solutions. Even though SLI has every chip working on the same frame, does it still suffer from the same delayed-input problem?
It wouldn't have been a problem in the days of the Voodoo2 SLI setup, as any player with one could get frame rates typically twice as fast or faster than pretty much every 3D accelerator out there, so the 2-frame lag would be the same or less time as the single-frame delay. However, with the ungodly frame rates offered by a single GeForce 256 with Double Data Rate RAM, if there were a two-frame delay for someone with a Voodoo5 5500, in a LAN game of Quake III the Voodoo5 user would be toast.
"Even genius needs a competent technique."--Robert Fripp
Still no geometry acceleration?? Bah!
Windows 95, 98, NT4.0 and Windows 2000 drivers Allows you to run the Voodoo5 6000 AGP with all popular operating systems.
Okay, not only am I defending linux on this one, I am also wondering where the MAC drivers are. If 3dfx wanted to have some incredible benchmarks they should write a MAC driver and throw it into a G4. They say the Voodoo 5's aren't only for gamers, why not port the drivers to the most popular graphics design platform?
I fail to understand why this stuff excites people. I've always thought that the market for add-on 3D graphics cards was going to develop a lot like the market for add-on sound cards did, and so far I'm seeing nothing that indicates otherwise.
What I mean is -- consider for a moment how the market for add-on sound cards developed. Up to 1992, sound on the x86 PC was basically nonexistant, unless you owned a flaky almost-compatible like the Tandy 1000. Then the multimedia tidal wave hit and suddenly there was consumer demand for hardware sound support -- and a market sprang up to fill the demand.
Once the demand for sound cards sprang up, the market developed through 3 distinct stages in the next 5 or so years:
So this is where we are today in sound cards -- while a few enthusiasts care about buying the latest Sound Blaster Live! or whatever, the vast majority of users are happy with the 16-bit audio that's hardwired into their motherboards. It's Good Enough!
And that's what's going to happen in the 3D card marketplace, IMHO, fairly soon. We've already passed through stage 1 (I remember agonizing over whether to buy a Voodoo1 or a Rendition Verite card) and stage 2 (with 3Dfx milking their brand name for all it's worth through the Voodoo3). But now Good Enough 3D hardware is starting to come integrated on motherboards, and 3Dfx's Voodoo-only APIs have been almost entirely forsaken in favor of Direct3D, which is integrated into the OS. I've run 3D games on cheapo PCs using this integrated hardware, and while the performance isn't great, it's Good Enough -- while the add-on card companies fight over which card can provide 80 fps in Q3Test, or other "features" which would be lost on the average consumer anyway. So watch for it -- in a year I'd be amazed if there's still a market for whizbang add-on cards. Most people will be just fine with the Voodoo2-level hardware they'll get free with their PC.
-- Jason A. Lefkowitz
Read my blog.
Upon seeing the specs for that baby, part of me just screams I want it, but the other, more rational part of me wonders what the point really is.
I mean, great: gigatexels per second. As much RAM as I currently have on my mainboard. Meaning what? I can now play Quake3 at 4,000*3,000 resolution? Yay. Yes, I know about anti-aliasing, but this is overkill for even that if not running very righ resolutions (1024*768 and above).
Read my lips, 90% of all speed problems with games on current hardware is the geometry setup bogging down the processor. Unless you play at above mentioned resolutions, or happen to have dual athlon 700s and are playing at 100 fps already (and if I am right in assuming that this does not have a Geometry chip like the GeForce) this card will be exactly 0% faster for you.
In my opinion Nvidia have taken a much wiser approach to the whole 3d acceleration concentrating on the weekest pointinstead of just pouring in endless amount of pixel fillrate that the processor can't render anyways unless you are stairing at a blank wall.
-
We cannot reason ourselves out of our basic irrationality. All we can do is learn the art of being irrational in a reasonable way.
Look at the Voodoo 1 :
- 1 pixelFX
- 1 texelFX
Now Voodoo 2 :
- 1 pixelFX (but running a bit faster)
- 2 texelFX (work on the same pixel, apply different textures, that's multitexturing)
Voodoo 2 was just a 'SMP' version of Voodoo 1.
ATI rage fury maxx use the same technique to boost frame rates (two Rage128 chips working in parallel).
The page with detailed info concerning these boards is www.3dfx.com/prod/voodoo/newvoodoo.html
The really interesting thing is that *once again* 3dfx promised us more than it will deliver. On the low end (Voodoo4 4500) these babies are getting smoked by the GeForce 256, which will be a half a year older! The GeForce can do 480 Megapixels per second, about 1.3 times as fast as a Voodoo 4 (which clocks in at 367 Megapixels per second).
If the past is any indication it at least a few more months for the Voodoo 5 to be released (ignore what 3dfx says), by this time Nvidia will probably already have a better card.
In summary, the Voodoo 4 is slower and less feature rich than the GeForce 256, plus is won't be out for 4 more months. It could take longer for the Voodoo 5 which will probably be an anachronism before it is released.
Come on 3dfx! This is *not* the technology that will keep us ahead of the PSX2!!!
I've been waiting to my old 90 Mpixel voodoo 2(which still is pretty decent on most games really)...
;-)
If Linux drivers come out, I'll probably go for the Voodoo 5 6000 quad beasty.. That should hold out for a while..
Anyone find anything on the memory technology yet?
This card would have to have some massive memory bandwidth to keep up with those fillrates..
I know even the sdram GeForce is memory bandwidth limited at much lower fillrate.
It appeares that each graphics chip has its own dedicated 32 megs from the specs(with up to 64 being addressable by each chip), so that is one memory trick I'm sure they are using.. Any other details?
(anyone notice that 3Dfx, long saying "32 bits doesn't make much difference" cuz they didn't have the technology, is now pushing it
Blessed are the pessimists, for they have made backups.
Unfortunately a 24 bit z buffer will not cut it for any sort of advanced application.
Old SGI Indigoes (1991!) had 24 bit z buffer and one could frequently see through polygons.
JJ
Why should they bother writing MAC drivers when they don't even have an AGP slot to plug the card in ??
Does anyone out there have any idea what comes after geometry acceleration, if anything?
a typical Slashdotter...I will pay $5000 for a video card, but $40 for Word Perfect is a travesty, because I can't see the source code.
So you don't have delayed input.
BUT performance wise, the ATI approach is better.
You're the one who's wrong. glide is a proprietary API that 3dfx cooked up to lock people into their proprietary standard. While OpenGL is relatively new on windoze, it's the open standard the Big Boys use in the Unix world, for serious stuff.
>But now Good Enough 3D hardware is starting to come
:)
For most people "Good Enough" hardware is here right now. A huge majority of gamers should be happy with the current generation of 3d cards, the improvements between generations are getting smaller and smaller. Also, the limiting factor in multiplayer is not the 3d card but your modem.
Maybe its just me but pretty pictures are impressive for the first hour. After that I want game play. Its the software which is more important to me.
I still have to go home and see how Q3Demo plays on my machine at home
The surprise isn't how often we make bad choices; the surprise is how seldom they defeat us.
How do you know more geometry will be better? I'm not sure myself, but IMO 3dfx has been king, more or less, up to this point because they choose the best features to include. That is the heart of engineering. Deciding what to include and what to sacrifice, which trade-offs are the best trade-offs. Other manufacturers have included more and "better" features, yet when you look at price and speed and quality, I feel that 3dfx has been the best so far. Although the TNT2s are very good products also, and I'm sure the GeForce will be too.
What makes geometry better than fillrate?
I wonder if this heralds an era in which a video accelerator has several processors dedicated to individual tasks(i.e. one for bump mapping, one for texture mapping, etc.) We could be on the virge of very fast graphics indeed.
makes sense... but they could port the PCI versions...
Check your facts.
To spur "enterprise Linux," Big Bang, the distributed two-phase commit.
If you have seen ray traced graphics and compared them to the current crop of 3D graphics accelerators then you will realise there is a huge difference in quality, and it will take many years for you to have that level of graphics in realtime, maybe then it will end up as an integrated part, I would guess we are 5 to 10 years away from that.
Any sufficiently advanced man is indistinguishable from God
Yes read the FAQ, it says it will have beos and linux support! lets see if pmesa will support it.
That's what it might look like at face value, but I have found rarely that engineering makes these decisions:
>because they choose the best features to include
This is usually done by the marketing department, and often to the chagrin of the engineers, who would like to think they know better...
Seth
$5 / month hosted VPS on linux = awesome!
- Quantum's dual Voodoo1.
- 3dfx Voodoo2 (SLI)
- ATI Rage Fury Pro 128 Max Turbo Fast Thing, or whatever the hell it is called.
- Bitboys Glaze3D (if/when it comes out).
Everyone implements it bit differenty, but the idea of SMP Video is not new. CAD-people, who use cards that cost more than a decent car, have had this stuff forever.That made me laugh! Actually, you're an idiot for not knowing how/when to use the contraction of "you" and "are". Duh. Go back to grammar school.
I can't see why people are up-in-arms about this paticular release, it's just the continuation of the graphics-card cycle. nVidia, S3, and the rest will continue to release new chipsets along with 3dfx. 3dfx is far from king of the hill: namely, they release products at such a slow rate that for much of the time it's better to get a TNT2 Ultra, for instance, instead of a Voodoo3.
Sometimes 3dfx will be on top, sometimes nVidia will be reigning champion (the NV15 is in the works, i hear...)
The Savage2000 and ATI Maxx are almost out...it really doesn't matter. These cards are way more powerful than anyone really needs, or will need, for a while. Until software comes out that really needs a couple hundred megatexels a second, I don't really care who's on top of the hill (and by that time, there will be even more powerful cards coming out)...
Then again, I could be wrong.
And not only that, but there are Voodoo 3 drivers available in Beta from 3dfx for those cards to run on Macs. I'd imagine they will continue evolving that code base to support the 4 and 5 cards on the Mac platform as well.
Seth
$5 / month hosted VPS on linux = awesome!
OK. From the neat little Shockwave Flash thinger on the 3dfx website, it says "So powerful, it's kind of ridiculous." Actually, I find it highly ridiculous.
Come on. Do we -really- need to be able to frag at 3200x2400x32bpp at 60 frames per second? Well, ok, I guess piping this to a 57" big screen TV would be nice.
But seriously, isnt this approaching overkill? I find Quake II to be quite fun on a Voodoo3 3000 AGP. Granted, my shitty 14" monitor is the limit, and why I'm only running at 8x6. But alas.
"To err is human, to forgive is simply not my policy." --root
now do we have the hardware to support Virtual Reality?
Macintouch is reporting that the PCI version is already supported by the company's existing Macintosh drivers. You can read the FAQ yourself. No doubt, if there is enough interest, drivers for AGP Macintoshes will be forthcoming.
Sig goes here
T&L is a gamble. High fillrate is a gamble. Bump mapping is a gamble. Any new feature a chip manufacturer puts on thier chip is a gamble.
No one knows what game companies are going to try next.
There is no way of telling whether hardware transformation and lighting is going to make any difference at all in future games. Sure, nvidia is going to tell you that future games will depend on it! There is no way of knowing that 5 gazillion texels/sec is going to really make much of a difference to future games, although 3DFX doubless wants you to think that. No one knows whether game companies are going to stuff their games with bump-mapped polygons, no matter how much Matrox tells you it's the truth.
Point is, each of these hardware developers are hedging their bets, that game companies will favor their technology.
As for us consumers, I would take a "wait and see" approach. I'd never go out and buy the latest and greatest until I see what games run well on them and what games do not run well on them. Specs from pre-released hardware are meaningless, and even released hardware that runs a FEW games spectacularly is nothing to base a purchase on.
Look for the architecture that stands the test of time, and has support for the platform and games you play.
In order to push that many pixels/texels fast writes will need to be enabled since the current memory bus is just too slow.
Also, is the current PC-133 Mhz standard fast enough (for workstations not gamers) to send 2GB worth of textures?
I don't know. That's why I'm asking you guys...
--Al
Winmodems do suck. Do you know what a winmodem does? It makes the CPU do the work that a $16 chip could do, for a 10% performance hit. Stupid, stupid, stupid. Also, winmodems destabilise your system, since they have to have priveleged memory and CPU time.
BTW, USB mouse and keyboard support is in linux kernel 2.3 as strandard now, and can be patched into linux kernel 2.2 too.
USB mice are superior to standard serial mice. (higher sampling rate). Winmodems are inferior to real modems. Your argument was, I hope, based merely on incomplete data, rather than any intent to disseminate misinformation...
I saw an interesting preview over on http://www.sharkyextreme.com/ this past week. The ATI Rage Fury Maxx (2 processors, each rendering a different frame) smoked the GeForce 256 in just about every test.
Funny, that ATI card doesn't have hardware T&L that everyone's claiming is going to end world hunger and lots of other stuff. So, why's the ATI card faster? It's got way better fillrate (aka polygon painting) than the GeForce. So, as benchmarked, the GeForce may have a spiffy T&L engine, but it's crippled by its slower fillrate on current games.
For now, I'd say that 3dfx is right: fillrate is king, T&L isn't worth spending the $ or transistors on until the fillrate problem is 'solved' (hah)
All of these Nvidia GeForce/3dfx Voodoo 4 and 5 boards are technically amazing but this level of consumer 3D hardware is in desperate need of a new killer app. I have a Voodoo 1 and I'll likely be more than satisfied with the performance of Quake Arena on that thing. I simply refuse to drop hundreds of dollars more to play games at higher resolutions/framerates.
Why, when we all have a global network right in front of us that is ablaze with information and commerce, is no one strapping a hardware-accelerated 3D engine onto the net? When am I going to be able to navigate the web in 3D? When can I use my 3dfx or Nvidia board to do real work, or to shop, or to explore real information and news? Why is the web still 2D? Wake the fuck up. Screw VRML - I'm not even asking for any sort of server tech - just give me a fly-through 3D-abstraction of the HTML/XML content that is already there. If people know the engines are out there they will start to build for them.
Bottom line - I won't give 3dfx or anyone else more of my money to play another FPS. I will part with more money for 3D hardware when I can use a Voodoo 5 6000 to give me an Ono-Sendai Cyberspace VII-like window into the net.
Night
Billy "Wicked" Wilson of Voodoo Extreme asked a bunch of high-profile devs exactly this question.
The majority response, was that if they had to choose, they'd pick a card with accelerated geometry processing and a mediocre fill rate over a card with an insane fill rate and no geometry acceleration.
What does that tell you about the direction the game developers want to go? They want to build games with higher-polygon engines/content. My guess is that's what we're gonna see.
No one's that dense. This guy was yanking your chain.
First some quotes (from the press release):
Re: the Voodoo4
The boards, which render two fully featured pixels per clock, will deliver between 333 and 367 megatexels/megapixels per second
Re: the Voodoo5 5000/5500
The board, which renders four fully featured pixels per clock, will deliver between 667 and 733 megatexels/megapixels per second fill rate
Re: the Voodoo 5 6000
The Voodoo5 6000 AGP, which renders eight fully featured pixels per clock, will deliver between 1.33 and 1.47 gigatexels/gigapixels per second fill rate
Now, if you'll notice they state how many "fully featured pixels per clock" each card delivers. Also, notice that the V4 does 2, the V5-5500 4, and the V5-6000 8. Along with that, as I guess one would expect, the V5-6000 has double the fillrate of the 5500 which has double the fillrate of the V4.
So? What's my point? Well, with the Voodoo2 -- which could render two pixels per clock -- the full fill rate was acheived only if the app was rendering two pixels per clock. (ie. multitexturing) If the app wasn't multitextured, the effective fillrate was actually only half the "marketing" fillrate. I think this was also the case with the Voodoo3, although I'm not positive.
I'm not saying that this is definitely the case with these cards, but:
Correct if I'm wrong, but I think these cards are still based on the same architecture as the V1, V2, and V3.
3DFX is somewhat notorious for advertising the higher "marketing" fillrate as opposed to the true fillrate.
The fact that they qualify the fillrate of each card by stating the number of render pixels per second kind of worries me.
If this is the case, apps that don't take full advantage of the high end cards (ie. have less than 8 pass multitexturing) may leave you with nothing more then a glorified and expensive Voodoo4.
A lot of people don't realize how important PCI is for dual-head displays. And I really like 3dfx for maintaining PCI products around? Why? This way we can have dual-head displays with two graphics cards of the same quality, one on AGP and the other on PCI, and they deliver a great ammount of performance. Here is 3dfx, your dual-head display king of the next millennium. Nobody matches these babies. And that is why people shouldn't criticize PCI technology, until we can have two AGP slots (I don't see that happening in the mainstream any time soon). So stop bashing on PCI, it will take 3dfx where nobody else can go! :-)
yeah
good lord, now this news...is there any reason to doubt how much of a 3d powerhouse, Linux/INTEL really is??
heck, a properly configured xeon machine with a TNT card smokes an SGI Onyx2 in most performance and 3d benchmarks. And SGI calls itself an innovator?? Please, theve done nothing for 3d/animation/graphics compared to what Linux has done for the world.
Look for Redhat boxes to Invade the movie scene soon...
About 8 months ago, I saw two guys kissing in the park. That was the gayest thing I had ever seen until I saw the junk you're blowing out your hole.
Alright, here we go folks, one by one...
Starting off, OpenGL is slow. Period. While the licensing issues of Glide are a bit iffy, it sure as hell beats the pants off anything else significantly. This isn't something too open to debate, as benchmarks that we don't see from Mindcraft show us over and over and over...
3D works well with OpenGL in linux? Well that's news to my developer ears. I could have sworn the first thing you see in the video section of the UT demo readme is that anything attempting OpenGL is dog-ass slow. In fact, I seem to find that in _any_ game except Q3 (kudos to Carmack).
Glide is faster in Linux? Okay, lemme think...no. It's cut down to about 3/4 of the Win9x speed. Sometimes worse. Ever tried, bub?
I'm sure the 2D looks very nice, and I'm sure my 4MB generic Cirrus PCI card does a 1024x768x32bpp desktop just as nicely. Smart guy. No, just kidding. The poster that you're flaming might be a little out of touch, but come on man.
Good. I saved $16. Why do I really care if it uses 10% of the CPU? My PC is idle most of the time anyway. Linux is single-user on my machine anyway.
-- Freedom means letting other people do things you don't like.
Hmm...It's fair to say that most people don't use 100% of the cpu very often. Why is saving 16 dollars (manufacturer's cost, consumer's cost difference will be greater) at the expense of giving up 10% of your cpu so bad? It's not my choice personally. But for the general market, there's no reason not to save a few bucks and let a $50, 3 or 4 hundred mhz cpu do the work.
Genuine ray tracing, probably backwards at first, later forwards. Lots of parallelism, one or more chip/functional unit/whatever per screen pixel. Enough RAM to texture map the planet.
I think the real reason that Linux has no drivers for Winmodems is not that they suck, but that the manufacturers have chosen not to make programming information available.
I haven't ever used a Winmodem, so I don't know how hard they suck, if at all.
Molly.
Performance issues aside, Winmodems suck because they won't work without Windows, and their manufacturers won't give out enough specs to allow them to work with other operating systems. Is this a MS plot? Who knows. It seems like a better "low-cost" alternative would be to build the modem processor into the motherboard, like the sound chips and video processors on many bulk systems.
How you managed to avoid having your post not moderated to flame-bait is beyond me.
Matrox offers an AGP card with two outputs. It's called the G400. Unlike the V4 and V5, it's already released and available.
Anyway, I'm somewhat off topic. But, I needed to correct this. The issue with PCI is bandwidth and texture swapping. If the card truly can have up to 2GB of RAM, (Yes, there are many simulation visualizations and mappings that can use this.), you'll need more than the 533 MB/s provided by the PCI bus. Even full AGP 4x (w/ RDRAM) has 1.06GB/s. An approximately 2s delay is damn noticeable.
If they can put multiple graphics processors on one card, why can't they put multiple output ports on the same card?
--Al
Good point, the bottom line is 3dfx and nVidia and all the other companies are building these chips and cards to make money. 3dfx has gotten some bad press because they didn't include 32 bit color on their Voodoo3 line. To me, that doesn't seem indicative of a company that is throwing on features mindlessly to market their product. On the other hand, one could argue that 3dfx didn't include 32 bit color for other reasons. 1. 3dfx sucks, nVidia rules 2. they just missed the boat on this one I don't know which is the case, probably no one but employees of 3dfx truly know. I do like 3dfx, but my BS meter goes up a little when they rant about no one needing 32 bit color, 60 fps is more important. Come on, any company is going to say what they have is better.
Of course, if no-one has high poly-count cards, then these games will be longer coming, because the market won't be there.
I truly hope that nVidia can gather enough support for the GeForce's geometry engine so that this battle is fought and won on which solution is technically better, not which company bullies the other better.
When the V3 was introduced last year you all complained that it lacked 32bit color. Well I am still waiting for a game with 32bit color that I want to buy. (Don't say Q3 or UT, they haven't shipped) So lets clarify: The V4 is for people that what 32bit color and do not want to spend a lot of cash. It is NOT a Geforce killer. And wow it comes out just when the 32bit color games are coming out, bet it kicks tnt2 hands down. The V5 will be the fastest card on the market until the T&L titles start to come out. Which by that time the V6,V7 and Geforce2 will come out. Look at how well your TNT 1 does in 32bit color and you will see how well the Geforce will do when the real T&L titles come out. The Geforce is irrelevant. nVidia will replace it before it gets a chance to show how good it really is.
the G4's have a 133mhz AGP 2x slot
In case you did not read futher, the 32 chip version with 2GB of RAM will cost around $40k. Also, why in the world would you want one, unless you were rendering high quality video in faster than realtime.... The thing could render Toy Story (for the lack of a better example) in about 10-20 minutes. Also, if you do the math, it could run resolutions 4092x3072 in 30FPS, 32bit, or higher, but what monitor could support it. At least the thing does 32bit, finally. Im opting for the V5-6000 (AGP, 4 chip, 64MB), for myself, and I think it costs over $300.
Why would you want to BUY TWO video cards to run dual monitor support when you can buy a Matrox G400 dual head and get dual monitor support from a single graphics card???
--- Errr......No I don't need more oral sex thank you, Windows goes down on me all the time.
Geometry and lighting setup is extremely important. For my stuff, it is a very significant bottleneck. It is the perfect type of operation to offload to hardware. Not only is 3dfx missing out on that, they have been dicks about the source code. When they go down, they'll have noone to belame but themselves.
Glide is going the way of most proprietary API's... the way of the scrapheap. Even Brian Hook, the author of Glide and former id programmer sees it as irrelevant.
Daryll Strauss, the developer responsible for most of the 3DFX driver code so far, wrote this in the 3dfx.glide.linux newsgroup; I'll preempt the first question: Yes, we'll be supporting them under Linux. - |Daryll
"All the action in the consumer marketplace is happening at the low end around cheap PCs, not at the high end around 1GHz Athlons."
I understand what you are trying to say (I think), but the explosion in growth in the low-end is because the prices are falling. This makes PC's more accessible to people who aren't willing to spend $$$ on a computer. In other words, different consumers are buying the cheap PC's, not the older PC consumers who have been buying every couple years. And definitely not the hardcore gamer.
I also believe that this enthusiast 3d market has a chance to expand and grow over time into more main stream. Sure, it's a toy now, but being able to render near photo-realistic worlds in real time in response to user inputs...that has potential. As the internet and technology grow and find new ways to solve problems, I think we may see some very neat applications for 3d technology.
One final note, I just helped my girlfriend order a new Dell computer. It's not bottom of the barrel cheap, but it was only $1100. It included an Aureal 3d chipset sound card. Sound isn't a commodity. Common, "good enough" sound is a commodity
But one thing to note, Nvidia doesn't actually make cards unlike S3 and 3DFX. They just sell the chips to OEMs, and the OEMs set the final clock speeds, memory used, etc.
Grumph! What happend to the HTML formatting I did?
It showed up in the preview, but not in the final post!
NOTE: Each sentence should be it's own paragraph. The first one should be in bold italic.
Finally getting rid of the 256x256 texture resolution limit is a Good Thing as well. Even Quake2 uses textures larger than that, and on the Voodoo[1-3] chips it just looked blurry and crappy because of it.
That said, I wonder what sort of marketing spin 3dfx's wonderful PR people will put on this decision, when for the longest time they were constantly saying how worthless 32bpp rendering and large textures and the like were. I also wonder if these chips will have true accumulator buffers (the press release didn't say anything about this) or their bastardized, crippled "T-buffer" crap. I also wish they'd drop extending Glide (for a number of reasons) and only have Glide 3 for backwards compatability, especially since Glide can be relatively trivially implemented in terms of OpenGL and adding on more features to Glide to try to make it catch up will just cause more cumbersomeness and an even greater rift between their Windows and Linux support. (I feel even sorrier for Darryl Strauss if he's got to do even more for-free work on extending Glide for a relatively thankless company.)
On the whole, though, 3dfx has a chance to actually redeem themselves with this new card. I hope they don't blow it; I'm all for giving them another chance. I just hope that they decide to actually have a good product instead of good marketing. For the longest time they seem to have just been resting on their laurels from having been the first usable (and not even decent) 3D card on the market. Maybe now that can finally change.
---
"'Is not a quine' is not a quine" is a quine.
"'Is not a quine' is not a quine" is a quine.
Quine "quine?
Ibag (I'll Buy A Geforce...with linux support, I do believe)
sigh...
yeah... for real... if anything it should have been moderated interesting
I think that the newer game consoles have this already, but the next great feat of accelerating games on a computer is a chip designed specifically to do the physics for you. Then we'd obviously need a standard world-description API, and then there'd be one more thing to fight over.
I also think that a very cool thing would be to have a Dreamcast or PlayStation *card* that you insert into your computer. It could use the CD/DVD drive to read its roms, and then you can play some really cool-looking game while recompiling something, with a minimal drop in speed for each of them.
-S
In the words of the immortal Buzz Lightyear:
"You poor, sad little man".
Maybe we will have working Linux Glide drivers for Q2-3 for the Voodoo3 by the time the Voodoo4 and 5 come out. I bought my Voodoo3 because 3DFX had a good record of Linux support. I'm growing a bit tired of waiting for working Glide drivers!
According to FAQ #16 on 3dfx's FAQ page Linux will be supported along with BeOS... On the other hand, I really don't care because I will not buy one of these cards anyhow. Who integrates a DVD decoder on their card and doesn't put a TV out on the board because "very, very few end-users want to play PC games on a TV." Ok, fine. Gamers don't want to play the games on TV but what about those of us who would like to watch DVD's on our TV? Come on 3dfx, use your brain...
of course the numbers are outstanding. But did you see the price of the flagship V5. $600!!! I'm really turned off by the PC gaming market. The turn-over rate for video cards is like 6 months and the price keeps going up. I bought a playstation for $250 its lasted me for years. When the Playstation 2 comes out, I will buy one and be set for gaming for the next 3-4years. Plus everytime I buy a playstation game, I know it will work. No do I have enough ram, a good enough video card, enough harddrive space. PLEASE! I program and surf the web on my computer. I game on my playstation. 'NUFF SAID!
>Maybe its just me but pretty pictures are impressive for the first hour. After that I want game play. Its the software which is more important to me. I agree, that's why my favorite first person shooter is Starsiege Tribes, unfortunately win95/98/NT only. For those of you who like quake style games but find them a little boring this is a great game. The engine isn't nearly as good as quakes. But the gameplay is great. Typical games are capture the flag. Great weapons balance. Team play with vehicles and objectives that must be held and captured.
Oops... That should have been 1.6GB/s for AGP 4x.
Standard AGP 2x (what most of you have) is 800MB/s.
PC-133 RAM provides about 1.06GB if I remember correctly.
Sorry about that.
--Al
Is is Office suites? Operating systems? The internet? The answer to all of these is no. What drives the 3d video card industry are the games. As long as game developers continue making games more and more complex, there will continue to be improvements in video acceleration. The reason I bought a Rendition v2x00 in the first place was to play Quake. You are foolish to think any gamer will be content with "good enough" when Quake 4 (?) or whatever comes out. The average consumer may be content with what comes in the box they buy at CompUSA...the person interested in games will never be content with that.
-------
"Every artist is a cannibal, every poet is a thief."
But they don't list Linux. Hmm, why should I spend $600-$700 for a piece of hardware from a company that doesn't consider my operating system of choice "popular"? Unless they change tune, Voodoo3 3500 will be the last card I purchase from them. Here's your chance to get my money, 3Dfx...
>Good. I saved $16. Why do I really care if it uses 10% of the CPU?
It depends which 10% performance hit you are getting.
Surfing: There is no problem with a 10% hit.
Playing Online games: Big problems. Its worth paying the $16. Your PC is not idle when playing games, even when you are just standing still in a multiplayer game.
The surprise isn't how often we make bad choices; the surprise is how seldom they defeat us.
Actually this is pretty insightful discussion about the evolution of graphics cards -- someone oughta compile an editorial with some of these comments. Beats the hell out of any off-kilter bizarro view of the industry by Dvorak or Berst.
That aside, I gotta agree with a couple posters above who indicate that new features on graphics cards are a gamble -- T&L, Bump Mapping, etc.
Myself, I pretty go by the Quake rule for hardware: whenever a new Quake is out, it's time to upgrade the mobo/CPU/RAM/Graphics/sound card configuration to whatever works best (i.e. way, way above the minimum requirements).
I've done this three times now -- just finished my third major upgrade last week -- and it seems an okay rule-of-thumb -- especially with regard to graphics cards.
At a threshold of 3 I saw a lot of "so what" posts. Here's my response.
Naturally, I'm all for a "let's just see when it comes out" attitude, but to answer the "oh great, even more texels" argument, remember that a massive fill rate means your geometry engine can have more overdraw without hurting performance. (Overdraw is where you "draw" multiple pixels into the frame buffer at the same place, and the one with the lowest "z" or distance from the observer is the one that actually shows up). If you had an infinite fill rate, you could draw the entire world as fast as your geometry setup would give you vertexes. Up until now engine designers have had to use tricks like BSP trees (ala DOOM) and Portals (Decent) to get overdraw as close to 0 as possible. With a high enough fill rate, you can get sloppy with your hidden surface removal and focus on other things. Of course, this is an over-simplication, but the point remains that more texels/s is not a bad thing.
Also, CPUs are still getting faster and cheaper. It will not be unusual to see dual-processor machines in homes next year. With Athalon using Digital's bus technology, quad processor machines could become Christmas pressents in 2000.
To answer the "what do I need 2G of textures for" question, think computed textures and textures with more information than use colors. If a texture has depth (bump mapping) or material information (alpha channel, refraction), it adds up. quake 3 uses 32-bit textures: 8 bits each for red, green, blue and alpha (transparency). Now let's immagine what we could do with another 32 bits: 8 bits of depth, 8 bits of reflection (I forget what this is called), and 16 bits for whatever effects would look good if they varied over the face of a polygon. Also, animated textures will quickly use up texture memory.
Yes, what we have now is pretty cool. Yes, 3DFX is being unfriendly to open standards. Yes, other cards may be a better bet. No, this is not the end-all-be-all of real-time scene rendering. Personally, I can't wait to get my hands on almost any of the cards that's going to be coming out next year.
Disclaimer:
I do not work for or even know anyone who works for 3DFX. I know two people who work for Creative Labs, and they hate 3DFX.
If 3dfx was primarily motivated by their marketing department, they would have used 32 bit color on their voodoo3. Instead, they used 24 because you don't really need 32 bit. Granted that 32 bit produces slightly better image quality; given a choice between 2 cards at $125 that will do 85 fps (Q1 demo1) on a k6-3 450 at 800x600, one being a 24 bit card and one being a 32 bit card, I would choose the 32. But even today, almost a year after I bought my Voodoo3 3000, you have to spend $200+ to get similar peformance at 32 bit. And to be honest, I can't tell the difference between 24 bit color on my machine and 32 bit color on my sister's new p3 550 gateway box (on my .22 19" monitor). My eyes are pretty bad, but still...
If 3dfx was primarily motivated by their marketing department, they would have also made 32 and 64 mb versions of the card as well, even though the increase in performance would have been negligible- just to sell more cards at a higher price. They didn't. Unlike some video card companies I might mention, they did the math and determined what the maximum memory required would be, then built the card. By building a little card and cranking up the real technology, they gave me a card that doubled the framerate I could get otherwise. Without tricking me. Without telling me that the V3 would make the net go faster.
A definition of an engineer is "Person who uses technology to solve problems". This is as true in computers as it is in any other field. Auto engineers trade horsepower and cabin size for fuel economy; aero engineers doing cfd neglect certain terms in order to get 'viscous' solutions in weeks instead of years. I happen to have a lot of respect for the engineers at 3dfx because they were honest about the capabilities of their product; anyone can put all the bells and whistles on and charge you a fortune. Engineering occurs when you start adding constraints, forcing engineers to make design choices, and then have the integrity to state the limitations instead of hiding them.
Thanks 3DFX.
(Now if they would just fix the alt-tab bug in gl quake...)
... and there is no doubt, that one day he will be
where the eye of his telescope has already been
*sigh*
Open Source. Closed Minds. We are Slashdot.
I see a few things here:
.18um processes, no single-chip solution is going to go faster than ~150MHz (safely). using the SLI-type design, 3dfx can ship faster solutions than the competition, using a cheaper process (.25um), with better reliability. Then when they have a cheaper .18 process, up go the clock rates and out comes a new board! Remember, for the V5 6k, clock rate increases are magnified by a factor of four!
1. It looks like this is an extension of the original Voodoo architecture, but with 32bpp rendering, T-buffer, LOTS more RAM, better Z-buffering, etc.
2. I'm not really sure that geometry accel is a terribly important issue. Modern CPUs (P3/450) can push 100 FPS in Q3 right now. The limiting factor to date has been fillrate. If they can pull off what they say they can, you'll be getting 100 FPS in Q3 at _all_ resolutions. Also, increases in FP capacity in x86 or whatever CPUs will continue to allow more complex models, engines, etc.
2a. Additionally, if you're using a geometry accel, you're EXACTLY LIMITED to the capacity of the chip. No faster CPU is going to give you better performance. What happens when your 10 MFLOP processor isn't enough for Q6?
3. Metal process: It looks like that until they move to
I think they've picked the right solution for the times, kept themselves from pissing off Intel (which Nvidia is sure to do), and delivered on their promise of but-kicking features and performance!
...which is real-time 3D raytracing (or something similar) at 60 frames per second, 32bits per pixel, at 1600x1200 resolution. Although personally, I would go for 1024x768 resolution.
Your analogy with sound cards is a good one except that sound cards have pretty much achieved their goal: essential 20 or more digital channels, 44.1kHz stereo or four speaker out. With 3D video cards, I personally think we are still five or more years off for the goal.
Currently real-time rendering in 3D games and applications is done with relatively low polygon counts, shadows are done using a hack (if they are done at all), reflections are also done using a hack. Shapes are confined to rectangular solids, or now, bezel curves. Ideally, the the gaming developers and players would like real-time rendering at high resolutions using extremely high polygon counts (infinite would be ideal), and realistic shadows, reflections and shading. We are a long way from this goal. Look at the very best rendered still images on PC, and then compare them to a game like realtime Quake3 (pretty much state-of-the-art for PC's). There's a huge difference.
Sound card technology is essentially commoditized because sound cards have no more technical hurdles to complete. 3D video cards are 3-5 years away from this state, and big leaps forward in technology (such as the Voodoo5) are exciting because they take us one step closer to that goal.
fine. I use what works, open/closed source, free or fee based. I don't treat it like some sort of religion.
"GeForce eat your heart out. 2gb, 32 processors"
"This is the kinda stuff dreams are made of come Q3"
Of course this is all true.
Question: how many gamers, hardcore or otherwise, can justify FORTY THOUSAND dollars on a video card? Yes. 40,000. That's the topend model's anticipated asking price. 2gb of ram don't come cheap, which should have been the first clue. The second being how much cards are these days with one or two of these 'chips' which I admittedly don't know too much about.
Open Source. Closed Minds. We are Slashdot.
actually, 3dfx cards can only do 16 bit color, not 24 bit.
___
If you think big enough, you'll never have to do it.
Don't think 3200x2400x32x60: think 1024x768x32x60...
...with five specular highlights from five dynamic lightsources, flickering of torchlight on faces and more sharply from metal surfaces, and every barrel or crate or object slightly different from having each one overlay about three slightly different 'dent' or 'dirt' layers. ;) if you did, you'd know that this is _precisely_ what ILM did to make the battledroids photorealistic- they were all identical models, but you had the texturemap for the robot, and then five different overlay textures putting different patterns of dirt and wear onto the droids- which were applied in combinations, of course.
I take it you don't read Cinefex
'Cinematic' means impressive- means multitexturing that would _choke_ a GeForce (or indeed a Voodoo3, but that's a given). It means the modellers will still be caring about polys, but the _skinners_ can go HOG WILD. Surfacing is not merely choosing a really big texturemap- talk to rendering people- overlaying translucencies and transparent textures is when you start getting really startlingly impressive effects. This throws the door _wide_ open for really amazing stuff. Polys aren't everything (it should be OK on polys anyhow, but polys aren't everything).
I have a g4 450 with an AGP slot and have had a v3 3000 AGP card sitting in it for weeks now without a problem. The "PCI" Mac Drivers also work for AGP ports. Something about AGP being a proper subset of PCI, but don't ask me. I just know it works. ::grins::
For a powerful design like Voodoo 4 and 5, having 32 processors working for the same frame is an extremely wasteful design. They should have tried AFR (not the ATI's flawed AFR, but the one that rasterizes the screen as soon as a screenful of data are received).
*g* ;)
Seriously. It's coming out in PCI, I want that because I'm hanging onto my nice old powermac for a while. I know _exactly_ what to do with all that texture bandwidth- multitexturing babeee! *g* forget polys. You'll end up with really boringly textured well sculpted shapes- geometry is NOT the weak link. I've appreciated the 3dfx strong points even through the drawbacks of 16/22 bit color- I've seen the transparencies and shading and tonal values looking better at 16 bit than the competition at 32 (not always, but in a number of cases, and always due to the 32 bit card drawing washed out tonal values). Now that 3dfx is ready to do the card with antialiasing that works with all my existing games, and with so much texture memory and fill rate that you could use it for fscking _filmmaking_ without it breathing hard *hehe*, well, I'm there. Build it, I'll buy it. My voodoo2 needs replacing, and I've never been more pleased that I didn't start planning to try and get a GeForce or something.
The output of this card _will_ look better than GeForce, by an order of magnitude. That's a prediction. That's also assuming a lot of multitexturing, but hey- if it's good enough for ILM, it's good enough for _you_
There isn't really a difference between 24 bit colour and 32 bit colour. They both have 8 bits each of Red, Green, and Blue. In a 32-bit mode, the extra 8 bits are your alpha channel. If you have equal 24-bit and 32-bit acceleration, you get the same performance for less RAM. You might even get slightly higher performance because you blit one less byte per pixel to your frame buffer in 24-bit mode.
Who cares about video cards I am telling you that macos x in it's internal builds has support for Altivec, and I have heard of framerates over 120fps, with the standard video, imagine something highend. Figure it out, altivec+OpenGL+Quartz+Purchase of Raycer+Pixar and Jobs+Apple's history in the graphics business+NextSteps superior 3D=the best graphics systems out there.
Microsoft's Direct3D has been tracking the latest developments in cards: D3D 7.0 will have direct support for lightmaps and stencil buffers, for example.
Where is OpenGL headed? Is anyone furthering its development? Is it going to track the features it needs to stay competitive as a games API?
Otherwise we're stuck with Glide and D3D. Talk about a Hobson's choice.
I've finally had it: until slashdot gets article moderation, I am not coming back.
T & L is all nice and good but all I want is a little hardware T & A
11 was a racehorse
12 was 12
1111 Race
12112
2 points to make here....
if T & L are not that important in a product, then
why are other manufacurers now adding it to their
consumer/gaming cards (yes I know that it has been
in professional cards for some time)?
Also,
3dfx has now gone OEM. They are thinking WAY TOO
corporate for my tastes, great products notwithstanding (still pro Nvidia here). The
numbers may be just a bit off, but you'll get the point:
the same year that the voodoo2 sold 5 million units, ATI sold 25 million of their crappy little
cards through OEMs, and yes, ATI is under the impression that their cards can play games well.
Starting with the TNT, 3dfx has been thinking of new ways to get your money with only small
increments of innovation/features in their products.
Is there anyone left anymore that cares just a bit
more for the respect of the community/public than
how much profit the quarter brought in?
3dfx is a great company, as is Nvidia, but people,
you must know that you vote for something in the market everytime you spend your dollar.
Thanks for your time.
...used to be a library...now it's just a mind-cemetary
As a note, the Rev 2 G4's (supposedly due sometime in April/May) will have 4x AGP... I'm still building a fantasy system based on the nVidia simply because Halo by Bungie is supposed to be a supporter - but we'll see... 2 GB of VRAM certainly sounds impressive !
1. No T&L. Sure, it may only be useful for games after 3-6 months, but this is exactly when the Voodoo 4&5 will launch, and Quake 3 already plays faster with NVIDIA and S3 products even now.
2. AGP support is still crippled. AGP execute mode is non-existent, just like Voodoo3. Although the local memory is supposed to solve this problem, this solution is not cheap enough to enter low cost market. On the other hand, AGP4x support does not include fast write to compete with NVIDIA.
3. While others have put motion compensation (and ATI also includes iDCT transform) into the chip, the only video acceleration VSA-100 has is planar to packed-pixel conversion, which only offloads a little CPU resources.
4. Multiprocessing uses the inefficient SLI. Although it has pretty balanced load, AFR offers similar results with greater efficiency.
5. Single clock is used to control memory and core, which leaves a little control for overclockers.
To sum it up, with exceptions of texture compressions, T-buffer support and highly parallel SLI setup (required for using T-buffer), VSA-100 is just what Voodoo3 should have been for many months ago. And even that, it fell a little short over TNT2, and certainly too late to compete with NVIDIA and even S3 for single-chip setup.
As for killer bandwidth goes, it works great with 1GHz Athlon, but for those who have less cutting edge CPUs (ie most of us), T&L is a much better gaming choice. And for those who don't play games, GeForce 256 works greater and definitely cheaper than $10k graphics workstations too.
There's some new driver on 3dfx's website that is a .src file. Are they opening Glide up????
Shuddup you sycophantic "Geeforce is gonna change the wurld" toadie
Hello. It's called lasting value. Five years down the road, you can plug in the Voodoo 5 and get your Pentium VI to pump out 100 times the performance than if you plugged in your Geeforce accelerator. Yeah, it's a stretch, but I haven't thrown away any old Voodoo 3d cards lately. It's called scalability.
To play the new games with geometry acceleration that don't exist yet. By the time they do exist, the main processor will be able to pump out more polygons. Geometry acceleration is like MMX. It ain't worth shit unless you actually code for it. T&L? How about T&A? I can implement software lighting the way I want to... not using those stupid dinky 8 hardware lights.
#include
/*
All the better to frag you with maaaaytee. Harharharharhar.
*/
Crud I should have used preview.
Forgot about slashdot html syntax.
#include ""
/*
All the better to frag you with maaaaytee. Harharharharhar.
*/
Wow... I am having serious problems today. The preview actually showed the include line, but when I posted it wasn't showing up. Sorry for all this interruption... I was just trying to be funny, but the attempt(s) failed horribly.
#include linux/mean_pirate_voice.h
/*
All the better to frag you with maaaaytee. Harharharharhar.
*/
Yea.. I totally agree T&L is a waste - Its only the 1st generation of this technology and 3dfx are wise to let Nvidia and GeFarce simulate the market whilst developing a powerful T&L engine.
24bpp: RVBR VBRV BRVB RVBR VBRV BRVB
Think word alignment.
The geometry calculations would be handled IN SOFTWARE but the 3D API in use (i.e. OpenGL or Direct3D). The game uses the 3D API, and that can make use of whatever acceleration is availiable in hardware, since for the game there is no difference at the API level.
We need a new moderation category.
Something like, "idiot", or "misinformative",
or "dead wrong".
Macs do now have an AGP port.
Who is the worthless moderator who marked this informative?!?
---CONFLICT!!---
Ever hear of a Console System call 3DO? It was pretty bad ass in the Console market for a little while, but then it died. I think it was creative that made a 3DO card that could do just that (im not sure creative made it, I think they did) Im not saying that either will die (Dreamcast maby?), But why spend money on a gaming computer ($1000) and an addon card ($100) just to play inferior games? Sure it would be nice to have all thouse extra games, but they really cant commpare to most PC games just because they cant suport many things, like Mutli-player, Updates, New Maps, etc... So why get hung up on the inferior stuff, if you have the PC take advatage of it.
I think hardware limits depend not on market share but on what software developers want to get done. Today's hardware may be capable of 8-speaker 3D sound and photorealistic 3D graphics at 120 FPS, but if developers have no need for it - perhaps they're more into interactive smell or force feedback - then the hardware guys just aren't going to put it in. Technology only exists to assist the rest of society, after all.