Slashdot Mirror
GeForce 3 Demoed - Running DOOM 3
green pizza writes "Yesterday at Macworld Tokyo, Steve Jobs and John Carmack demoed the new nVidia GeForce 3 (73 GIGAFLOPS of power)... on a G4... running DOOM 3. Please excuse me while I pick my jaw up off the floor. You can get more details
from MacNN." [michael adds: VoodooExtreme has screenshots of Doom 3 running on the new GeForce card; Shugashack has more. Looks like Doom 3 will be another game where the color palette is "shades of black". Sigh.]
Yes, DOOM3 is dark, but probably not as dark as what you are seeing in the screenshots-- look at those shots on a Mac or SGI instead of a PC and you'll be surprised how bright they are.
magic
a realplayer version of the video is available at CNet, so the rest of us can watch it...
Walter H. Trent "Muad'Dib"
Padishah Emperor of the Known Universe, IMHO
--Fesh
--Fesh
Kill -9 'em all, let root@localhost sort 'em out.
God forbid people using technology to have fun.
Well I guess this will be a nice workout of the new FreeBSD server. For the curious the screens.x script is PHP that works off a bunch of f_checks based on file and directory names to generate those screenshot pages. No database hits involved in serving up those DOOM images.
Oh yeah, and DOOM looks amazing. Too bad NVidia cant make up their mind about releasing more info on the GeForce3. They have changed their mind 3 times in the past 24hours.
-Steve Gibson
-Steve Gibson
Shacknews.com
Well, it certainly speeds up the renderer if you just assign half the screen pixels to black ;-)
--
--
"I find your lack of faith disturbing." -- Darth Vader
You stupid jackass monkies. NVIDIA is NOT in with Microsoft! All they're doing is making a chip for the XBox. Think of all the things they've done *against* MS. They basically told them to shove DirectX and make a totally changed version for 8.0. They are comitted to supporting MacOS X, which takes away business from MS (especially 3D artists would rather use MacOS, but can't because the current state of HW-accel). They are helping SGI by making Linux drivers for their cards, which again takes business away from Windows2K (SGI used to ship NT workstations, now the low-end is Linux /w NVIDIA). They were the first good consumer graphics card company to come up with a solid OpenGL ICD, which again took business away from MS and made it possible for so many OpenGL titles to be out there. Quit trumping up phony charges.
A deep unwavering belief is a sure sign you're missing something...
I mean, is this supposed to be meaningful in any way?: http://216.105.168.97/cgi-bin/image-o-matic.cgi?d
So close, but yet so far!...hehe...But this video clip is kinda cool.
2) Anyone know any concrete scheduling info. on Doom3?
3) So, how much do people think that Apple paid nVidia for the whole "out on Apple first" deal?
Yes, DOOM3 is dark, but probably not as dark as what you are seeing in the screenshots-- look at those shots on a Mac or SGI instead of a PC and you'll be surprised how bright they are.
Cranking the Gamma on this monitor improved the movie no-end, so I expect the game will be the same. Think of the Quake3TeamArena demo level - with the gamma slider at the lowest setting you could barely see anything.
I'm amazed at the number of people who seem to think that DOOM3 should be some multicoloured bright-light party. Thats what Nintendo games are for. Doom's legacy demands dark rooms, illuminated by flickering flourescent lights, monsters which appear out of the shadows or drop on the player from the ceiling.
And seriously, looking at the video footage of Doom 3, this is going to be a game to give you nightmares. The characters are going to be closer to realistic images than ever before - those Maya-produced animations are pushing several thousand polys when up close (I assume that the meshes will have Level-Of-Detail) - playing this is going to be like starring in a good (or maybe even a bad :-) ) horror movie. I fully expect to see some Army Of Darkness mods based on this engine :-)
Don't expect Doom3 to be a game for kids. This one will earn a 'Mature' rating almost straight off the bat.
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
Doom3 just looks amazing. Even these early tech demo scenes make it clear that they have reached the photorealistic level. Say whatever you want about NVidia, but Carmack made that card fly...
-magic
You know, there's another term for that; color calibration, monitor calibration, gamma correction. All Apple machines automatically do that, it's built into the software. The monitors are also shipped calibrated. The video cards all support it as well. On the PC world it wasn't really a feature until a few years ago, and a lot of people have no idea that it exists. There are reasons why Apple still sells big in the publishing industry, because they want this feature; what they see in a brochure is identical to what they scan is identical to what they print is identical to what they email to their partners in the UK is identical to what is printed at the bookshop in Singapore is identical to what they see in their heads.
Geek dating!
GPL Deconstructed
Microsoft is the only supplier of Windows, and Creative is the only supplier of Sound Blaster-compatible sound cards. nVidia, on the other hand, is by no means the only supplier of OpenGL and Direct3D accelerators.
Really, what does nVidia have that no one else is theoretically capable of matching?
I don't see any kind of proprietary API that only runs on nVidia hardware. I only see them supporting OpenGL and Direct3D - two APIs that anyone else can use just as well. The only way they've brought themselves into such a powerful position is simply good products and good business. Ever since the TNT they've been aggressively pushing new features like 32-bit colour and hardware geometry acceleration. It's not they're preventing other companies from adding new features, so tell me, what egregious tactics have they used to shut out competitors?
Like Citizen Kane. What is that? CRAP!
Or The Maltese Falcon. That's crap too.
Or the original Gameboy. Shades of black on the screen. Ergo crap.
Or even that asshole M. C. Escher. Lots of shades of black there. And it was craptacular.
sigh
Curmudgeon Gamer: Not happy
Still, that card's feature-set is certainly a force to be reckoned with. I'll stick with the X-Box for now.
Since it doesn't appear that there will be a separate story talking about the other new announcements at MacWorld Expo, here are some of the highlights most interesting to Slashdot readers:
- The Power Macs were subtly altered. The former build-to-order dual 533 MHz model is now a standard configuration from the Apple Store (which means retailers will start carrying it as well), and there's an option to purchase a 733 MHz model with a CD burner in place of the DVD-R/CD-RW combo unit (saving $400 in the process).
- Those wild new iMacs have at least upped the specs slightly for the graphics chipset; still Rage 128-based, but at least there's more memory (16 MB) on board. Plus, the 500 and 600 MHz models are the new G3 chip with the full-speed onboard 256K cache.
- The Cube now offers the GeForce2 MX card as a build-to-order option (standard w/ CD burner on the high-end model). Guess ATi's still on Apple's shit list to some extent.
- Fellow Mac users should try running Software Update and see if they get CarbonLib 1.2.5.
- The $49.95 5-pack of DVD-R disks is finally available from the Apple Store, but the estimated ship time is 45 days.
Now, if I could only get a Flower Power G4 Cube..... mmmmmmm....
The Mac forums are blazing with commentary on the new iMac colors. Personally, I kind of like them, and hope they do a good job of stimulating interest in a highly overlooked demographic; women computer users. Anything that brings computing power to a wider audience can't be a bad thing. Besides, the effect is supposed to be slightly 3Dish, with the pattern all the way through the case instead of merely stuck onto the surface. I can't wait to see one in person.
--
I use Macs for work, Linux for education, and Windows for cardplaying.
Or at least that's what The Register says. :-)
Well, today's the 22nd.
So, that's 2/22/2001
Three two's in the day (without the year)
2+0+0+1=3
2222001, or 222*3 = 666!
We're all doomed!!!!
Or something.
Raptor
Raptor
"Procrastination is great. It gives me a lot more time to do things that I'm never going to do."
Niven a pacifist? No way. A pacifist would never write stories with themes like, "but sometimes you just have to fight anyway." That is exactly the theme that is presented in the short story "The Warriors," where pacifist golden-age-of-no-violence humans first meet the Kzinti and STILL kick their furry asses.
I know, I know, I shouldn't feed the troll. But I hate to see someone slander Niven that way. Well, go ahead and slander his last few years of work, but the classic Known Space stuff is still some of the best SF ever, IMHO.
"Optimized for Direct3D" does not mean that it isn't also "Optimized for OpenGL". And if NVidia is so closely tied to Microsoft, then why are they releasing the GeForce 3 for the Mac first?
------
To generalize your question, what software is used that sound cards could accelerate?
How about digital USB speaker output? Right now that sucks up CPU resources that a good sound card should be able to handle.
How about MP3 encoding/decoding? Right now it's a trivial 2% of my system, but if I up the bitrate, the number of channels, and the 'effects', I can start soaking up CPU. Why not have a soundcard accelerate it the same way video cards accelerate 3d graphics?
How about voice recognition software? Hardware accelerate that!
3d sound: Anything that uses a 3d library should be able to use 3d sound. Imagine Quake3. If the soundcard could access the level data, the walls, the enemy placement, the weapon type, etc, it could actually do occlusions, echoes, reverbs, damping, amplification, cancelation, etc.
Geek dating!
GPL Deconstructed
How can you, as a BeOS fan, stick up for nVidia after they've given BeOS the ding-a-ling dinner* for so long?
They provide binary-only support for Linux, no support for BeOS, and they've entered into a deal with Mac to artificially release their GeForce 3 on Mac first. (I say "artificially," since you know damn well they have more people in the PC world that want the technology.)
nVidia blows goats. I have proof.
-thomas
* Ding-a-ling dinner. Translation: Telling someone to "blow you." AKA "kiss my ass."
"And like that
Well, I am really supposed to keep my mouth shut, but I will tell you one thing. Don't tell anyone else, ok? ...Not only does the new Doom have barrels *and* crates, but it has radioactive and explosive barrels *and* crates!
You can tell by my blatent overuse of bold and italics that these two technologies combined obviously represent a tremedous leap forward in how we play computer games !
I wish the blinking tag worked so I could explain this better :-(
Liscencing fees for the BC technology will start with the sale of your soul and go up from there.
Unfortunately, the giant projector screens are not color calibrated the same way at all.
The colors did get rather washed out on the big screen.
John Carmack
If only there were a way we could see images generated directly from the source...
Okay, so this is totally off topic, but I just ordered a Titanium PB, I'm going to get Mac OS X the minute I hear the PB has shipped, and I have a hobbiest's interest in gaming.
Prelude aside, I haven' been able to find much in terms of cross platform programming on Linux, Mac OSX, and Windows 2000, three of the least popular gaming platforms out there behind PSX, Gameboy, and Windows 98. Though I suppose what with the DX 8/7 support in W2k, Windows 2000 isn't a problem.
I *really* want to learn and use Objective C, the Cocoa libraries, and OpenGL. I know that's not a problem with the Mac, given that Apple has made them all first class citizens of Mac OSX; is there any chance of being cross platform?
Or do you just code straight C for the game (and thus target every platform on the planet I guess), with platform specific code for the input and display handling?
I really wish someone had a book published, using iD as the case study, on cross platform development. IDEs, compilers, best practices, optimization techniques, workarounds, etc.
Of course, just saying all this out loud has given me a solution ^^; Code in C, abstract out the platform specific display, device, and input handling routines into a separate library, and use the 'best' software for each platform, whether that be Metrowerks, GCC, Visual Studio, etc.
Actually, I guess you could use Metrowerks for all the platforms, couldn't you? Is that what you do?
-AS
-AS
*Pikachu*
Yes, that explains why NVidia's GL drivers have always been faster than their D3D drivers.
NVidia implements their OpenGL drivers in hardware too. This is unlike other handware companies (*cough* 3dfx *cough*) whos GL drivers are simply wrappers around some other API (like GLIDE).
You will also notice that NVidia's GL drivers always support the newest features of their cards long before D3D does. For example, NVidia's GL drivers have had the "NV_vertex_program" extension (for programmable vertex shaders) since long before D3D8 was released. Similarily, when the GeForce was released, the GL drivers instantly supported T&L, while D3D users had to wait for Microsoft to released D3D7 for that support.
Might I remind you that NVidia employs several people like Mark J. Kilgard, the author of OpenGL Programming for the X Window System (the definative work on the subject), and the GL standard windowing library, GLUT? (If you have ever done any programming with OpenGL at all, you probably used GLUT to do it.) MJK is one of the biggest names in OpenGL on the planet, and I suspect he is part of the reason that NVidia has the best OpenGL implementation in existence.
Please make sure you know what you are talking about before you talk. Thank you.
------
We did a ton of testing the last two weeks while we were putting the demo together.
The 733 G4 was not as fast as my 1 ghz PIII in any of the trouble areas.
Apple is doing a lot of good work, but the CPU's just aren't as fast as the x86 ones.
AltiVec can compensate in some cases, because it is way, way easier to program for than SSE, but it takes a very simple batched, computation intensive task for it to pay off in any noticable way. Amdahls law and all that.
We did a couple functions with AltiVec, but they didn't make much difference.
Video encoding and large image processing are two areas that it can pay off, because you may be spending 90%+ of your time in one page of code.
Even then, it takes a special balance to let a G4 come out ahead, because it has less memory bandwidthd than a high end x86 system.
John Carmack
The AltiVec unit requires a couple of things that you dont have to deal with in normal code. The data has to be nicely laid out all in one place, and you have to be willing/able to deal with it more than one element at a time. Most programs are not designed that way (or if they are it is only by chance) so usually it is hard at first to find ways to retrofit Altivec into existing code. Certain trends in programming (e.g. OOP) make it even less likely that you will see these kinds of data structures. So to make a long story short, to get pervasive use of AltiVec, you have to design around it. I cant speak about the Quake engine, but from John's comments that sounds a little bit like part of the problem.
This is not to say that AltiVec is useless for such apps. AltiVec more often than not just so happens to be very good at accelerating that 10% of the code that consumes 90% of the CPU time. So, a little bit of work often goes a long way.
What you learn working with it is that memory bandwidth is almost always the problem. The result is that certain common old-gen programming techniques for code optimization that rely on memory access to save CPU time (longer code, lookup tables, etc.) are about the last thing you would want to do with AltiVec. You only unroll loops as much as is necessary to get proper scheduling and stop there, for example. Almost everything, including constants are best calculated on the fly rather than loaded from memory. As John points out, you are better off with one big complicated function rather than a lot of simple ones. Accessing memory is so expensive that once you have the data you had better do a lot with it. You typically have about 30-40 cycles of time per 32 byte block of memory that is "spent for you" every time you load memory. If you dont do something worthwhile during that time, it is your loss.
AltiVec is also good for sound code. It is a great way to reduce the overhead associated with single sound channels so that you can have lots playing concurrently without having time spent at interrupt running amok.
Ian Ollmann iano@cco.caltech.edu
Umm, MS got to be a monopoly by doing tricks like shafting OEMs for using alternate OSs. They got to be a monopoly by playing the API-of-the-day game with IBM. They got to be a monopoly by doing everything they could to get Netscape off of Windows. They got to be a monopoly by tieing vendors to the closed DirectX API.
NVIDIA cannot be a monopoly. Monopoly is more than just market share. NVIDIA, even if it gets 100% market share, cannot be a monopoly because it is based on open standards. They don't have anything like Glide to depend on. The minute somebody comes out with a faster card, they lose the top position. Simple as that.
A deep unwavering belief is a sure sign you're missing something...