backtop


Print 42 comment(s) - last by RobberBaron.. on Jul 3 at 8:54 AM

ATI's physics goes on the back burner

During ATI's 2006 third quarter financial results conference call this morning, ATI CEO Dave Orton revealed to analysts that despite demonstrating physics processing earlier this year, volume availability will not be for another "9 to 12 months." Orton indicated that the technology was still undergoing testing and preparation. Interestingly,  ATI indicated earlier that its physics technology was already on the market and needed only driver updates.

DailyTech has attempted to test some physics-enabled demos on a pair of Radeon X1900 XTX boards but drivers were not ready. Interestingly, ATI demonstrated physics on its RD600 chipset at Computex earlier this year but a year from now, ATI's RD600 will more than likely be phased out. In fact, ATI said during the call that it was already developing and testing RD700, its next generation core logic.

After ATI's announcement about its own physics solution during Computex, AGEIA responded by indicating that ATI's method was crude and did not really address the core needs of physics processing. AGEIA also indicated that its own widely accepted technology was already available on store shelves. AGEIA's availability argument was somewhat called upon by ATI customers, citing that ATI too had boards already in the market. Orton's announcement today unfortunately reaffirms AGEIA's point. Many ATI customers will have to wait a while longer.

Details on ATI's R600 core -- the replacement for its X1900 family -- is expected to be available late this year or in early 2007. It is unclear whether or not ATI's physics API will continue to support the X1900 family a year out from now.

Update 06/30/2006
:  We earlier reported that "retail availability" was not to be available for 9-12 months.  Instead, Orton's exact meaning was for "volume availability," and as such DailyTech has changed the article text. Will Willis, Senior PR manager for ATI sent us the following comment about ATI's physics solution:
Dave’s reply was that we won’t see physics in volume (i.e. hundreds of thousands of cards being used for physics) and therefore a material revenue stream from physics for 9-12 months (i.e. another 3-4 quarters) as that’s when there will be some substantial volume of ATI graphics cards being used for physics purposes.

ATI GPU’s that support physics acceleration are already in retail (i.e. the X1900 and X1600 series of cards for example).  All that’s required is a driver update that enables physics processing, and more importantly, game content which we expect around the holidays.  We’re very likely to see revenue from physics before 9-12 months, but as Dave said it won’t be in volume, it will likely be early adopters.
Willis claims content for ATI physics will be available this holiday season already.  However, there is no word yet on the driver other than it is "several weeks out."


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Slow And Steady Wins The Race
By Ulfhednar on 6/29/2006 10:35:32 AM , Rating: 0
After the joke that was Ageia PhysX, it's good to see that ATi are not going to rush on this. I hope Nvidia decide to take it steady too, as physics really is not a desperately needed feature just yet and I'd rather it be done right when it is.




RE: Slow And Steady Wins The Race
By segagenesis on 6/29/2006 10:46:02 AM , Rating: 3
PhysX is suffering the same fate that killed off the Kyro series of chips (remember when everyone salivated over these "ATI/nVidia killers"?) which is lack of good support. With the first game showcasing PhysX being at best ho-hum performance benefits you wont be getting people to line up in stores over this. In a similar vein I remember how the Kyro II would be twice as fast as a Geforce 2 MX (its competition) in Quake 3 but in just about every other game there were massive bugs or it ran incredibly slowly.

Perhaps its not thier fault because they are not developing the games but at the same time its not helping thier situation the games available to use it dont really use it well.


RE: Slow And Steady Wins The Race
By Slaimus on 6/30/2006 10:37:16 AM , Rating: 2
A comparison with S3 Savage would be just as good. S3 had more technologically advanced chips, but they were hampered with poor drivers and slow memory.


RE: Slow And Steady Wins The Race
By Wolfpup on 7/1/2006 11:07:07 AM , Rating: 2
You've GOT to be kidding!?! S3 and Kyro having more advanced chips? Not even close! S3 was the first with texture compression (I think), but other than that they made no contributions at all to graphics).

Those guys were trying to fight Nvidia and ATi's LOW end cards. And what's the point in that? Even someone going for a low end card is better off buying from a brand with proven drivers.

But...yeah...chicken and egg problem for that physics card. We don't even have proof that it's better than a CPU or GPU doing nothing that physics, but even if it is, it won't be used to change GAMEPLAY until almost everyone has one, and that won't happen until it's used to enhance gameplay. Don't know how they thought they could get around that issue.


RE: Slow And Steady Wins The Race
By Xeeros on 6/29/2006 10:47:05 AM , Rating: 2
Oh well this is kinda disappointing though as I was hoping it would lit a fire under AGEIA's arse to make a competitive/performance part with options in the driver set to IDK say not compromise system performance just for a few extra moving crates.

Either way though ATI's solution does have some fans simply because the prospect of using older model cards when you upgrade to do physic work is tempting.


RE: Slow And Steady Wins The Race
By MacGuffin on 6/29/2006 11:12:18 AM , Rating: 5
Pure Speculation: ATi is probably going to support Microsoft's DirectPhysics. Ageia is banking on their own PhysX engine. nVidia is supporting HavokFX. Havok 4.0 is apparently a winner with great use of dual/multi-core CPUs. ATi and DirectPhysics might be a winner.

Personally, I have this feeling that DirectPhysics will become the defacto standard and ATi is probably waiting to see how that works out. I just hope this doesn't inspire nVidia to deliver a rushed product to 'capitalize on the situation by launching a product that will serve no purpose'.

Then again, this whole PPU concept makes me mad. Isn't this what multi-core CPUs were gonna be used for (from a gaming perspective)? Are 10 times more flying crates/bouncing sparks really worth a dedicated processing unit? I seriously hope this whole concept fades away and developers decide to support multi-core CPUs.


RE: Slow And Steady Wins The Race
By Trisped on 6/29/2006 11:48:00 AM , Rating: 2
Microsoft has never actually said they are creating a physics API. For all we know they might have been covering their bases with that job posting, or they could still be looking for someone to fill it. Even if they are working on it, it will be at least 3 months after the leak that Microsoft confirms it, and at least a year before cards that support it start to ship. ATI is not going to put its cards on hold for the Microsoft API.

As I have said before, PPU and GPUs are a large number of cores on one die. For example, x1900s are 64 cores in 4 core groups (for the 1:3 pipeline ratios). Most multi core processors are dual core (2 cores), with at least one reserved for the main game code, and the other partially used by system resources to manage data transfers etc (unless they game is properly threaded, in which case all cores are going to be pushing the game engine). In most cases you will not have sufficient CPU power left to quickly process a large number of dependent calculations necessary for complete physics.


RE: Slow And Steady Wins The Race
By bersl2 on 6/29/2006 1:42:18 PM , Rating: 2
SIMD instructions on CPUs only go so far. If we had a general calculation library that abstracted the mass calculations needed from any specific chip... but of course, we can't, because we don't have the documentation necessary to create such a thing.


RE: Slow And Steady Wins The Race
By saratoga on 6/29/2006 2:05:00 PM , Rating: 2
Conroe and the K8L should roughly double the vector throughput per clock of the K8, single threaded. Add to this quad core and you have another 4x. That works out to an 8 fold increase over the current single core, A64/P4 generation, assuming no clock speed increases. Couple with large caches, lightening fast access to system memory and single digit latency for interprocess communication, I think this will more then give dedicated PPUs a run for their money.

The biggest issue though is going to be the bus. I don't know how fast Ageia's future PCI-E cards will be, but thats going to seriously limit the range of problems they can be applied to because at a certain point its just faster to let a local core do it then wait hundreds to thousands of clock cycles for a PCI-E device to do it.


RE: Slow And Steady Wins The Race
By Hydrofirex on 6/29/2006 6:42:04 PM , Rating: 1
Reading the reviews of the Ageia PPU I start to think that quadcore processors, more cache, and better system performance overall might actually manage to make a PPU worth it! As it is now they're horribly bottlenecked by current systems. I'm sure the coding will get better, but if you think about it all these dependent and related calculations - they must start to bump up overhead.

Just my opinion, but PPU's will come in (Come on, they make way more sense then spending ~$1000 for 2 VGA's!) - The real question is if they'll be intigrated into some other component or will systems get so fast processors can just handle it. I would wager intigration in the mid-run. I'm sure everyone's happy to let Ageia be the 'front-guard' here.


RE: Slow And Steady Wins The Race
By saratoga on 6/30/2006 8:28:43 PM , Rating: 2
quote:
Reading the reviews of the Ageia PPU I start to think that quadcore processors, more cache, and better system performance overall might actually manage to make a PPU worth it! As it is now they're horribly bottlenecked by current systems.


I don't think you understand what a PPU does. It offloads functions from the system. The system doesn't bottleneck the PPU since you're not interested in PPU performance, but rather are interested in overall system performance. What actually happens is the faster the system gets (particularly in vector performance), the less interesting a PPU gets simply because you don't get such a large advatage from offloading things. As more and more cores become available, the utility of coprocessors in general will drop.



RE: Slow And Steady Wins The Race
By AndreasM on 6/29/2006 4:57:02 PM , Rating: 2
quote:
Pure Speculation: ATi is probably going to support Microsoft's DirectPhysics. Ageia is banking on their own PhysX engine. nVidia is supporting HavokFX. Havok 4.0 is apparently a winner with great use of dual/multi-core CPUs. ATi and DirectPhysics might be a winner.

Personally, I have this feeling that DirectPhysics will become the defacto standard and ATi is probably waiting to see how that works out. I just hope this doesn't inspire nVidia to deliver a rushed product to 'capitalize on the situation by launching a product that will serve no purpose'.

Then again, this whole PPU concept makes me mad. Isn't this what multi-core CPUs were gonna be used for (from a gaming perspective)? Are 10 times more flying crates/bouncing sparks really worth a dedicated processing unit? I seriously hope this whole concept fades away and developers decide to support multi-core CPUs.


Actually Ageia has stated in interviews that they will support a unified physics API if someone makes one, this is the same way 3dfx operated, by having their own proprietary API (Glide) and then supporting other APIs (OpenGL, Direct3D). It's really just a matter of Ageia making drivers for DirectPhysics, and they'd be pretty stupid not to. This would be a win-win situation for everyone, as customers would be able to choose between Ati/Nvidia's (Nvidia would also be stupid not to support DirectPhysics) GPU physics, Ageia's PPU or multicore. Then we could also finally see which solution is the fastest.


RE: Slow And Steady Wins The Race
By meyerds on 6/29/2006 7:02:08 PM , Rating: 2
Correct me if I'm wrong, but wasn't there an article either here or on Tom's Hardware/Anandtech that stated that both ATI and nVidia were working closely with Havok for their physics implementations? So AGEIA is the only one with the truly proprietary system (made in-house). Microsoft will undoubtedly work with Havok (or who knows, maybe aquire them ;-) for their DirectX Physics implementation (which was announced a few days ago, I believe).

Personally, I question the need for dedicated PPUs (Physics Processing Units). Looking back in the industry, what has come out as a significantly successful technology tends to be good for more than just games. Most physics processing in the business/science world are done by CPUs/supercomputers made for the job. I do not anticipate this technology becoming useful/succesful in any other application than gaming. Unlike graphics accelleration/OpenGL (speaking of which, where is OpenPhysics in all of this?) which was used VERY heavily in workstation graphics processing, etc. before and after games began to take advantage of the system, PPUs started, and perhaps are going to be used exclusively in the game industry. Not a formula for success, in my opinion.

In short, I agree with your comment. Perhaps a better solution is to create a standard physics engine (DirectX Physics/OpenPhysics) that takes advantage of Multi-core CPU hardware for physics accelleration in software - similar to what Havok has already started for software-based physics. I don't think a few extra sparks and randomly-bouncing crate explosions are worth $150+.


RE: Slow And Steady Wins The Race
By PrinceGaz on 6/30/2006 9:10:11 AM , Rating: 2
Yeah, I was wondering where OpenPhysics is too. Though it would probably be called OpenPL, after all we've already got OpenGL (graphics) and OpenAL (audio).


RE: Slow And Steady Wins The Race
By meyerds on 7/1/2006 12:04:25 AM , Rating: 2
You're right, of course... it is/will be called OpenPL (Open Physics Library) if that standard becomes a reality. I was simply making reference to it's Open Source nature w/o acronyms for the sake of clarity. Actually, I recall running accross an "OpenPL" project not too long ago (shortly after AGEIA was released). If I remember correctly it was an MIT project. Doing a quick Google search just now didn't reveal anything, but who knows?


RE: Slow And Steady Wins The Race
By crystal clear on 6/30/2006 7:43:56 AM , Rating: 2
The Forecast for AGEIA
Kristopher Kubicki (Blog) - June 5, 2006 6:43 PM

Pls read this article-good reading.


RE: Slow And Steady Wins The Race
By daddypop on 6/30/2006 10:23:21 AM , Rating: 2
As far i can see, the majority of you guys are complaining about Physics in general. Why complain over EVERYTHING, if you dont agree with physics, here's an answer, dont but the f---ing card, dont worry what NVIDIA and ATi are doing, who cares you arent interested right, so quit all your b----ing and dont buy anything.

To use a dual core CPU to process physics is great, most games dont even run any better with dual core CPU's anyway. who needs 2 3.0 CPUs to run Half-Life 2 or F.E.A.R.??? Nobody, if you think you do then you are an idiot.


RE: Slow And Steady Wins The Race
By Xavian on 6/30/2006 5:59:24 PM , Rating: 2
true but lets not forget that dual core has other applications, i could encode and burn something to dvd with one core and play a game with another (i also have three seperate HDD's for different uses and im sure 2GB of RAM would be enough for both purposes).

Not to mention you could have one core take care of the crap running behind the game (windows and such) and have one core completely and utterly to the game. With the X2's and their seperate caches (linked by the very fast HTT links) you can easily have both cores doing completely different tasks with very little impact on either task.

Physics cards right now are useless. They add eye-candy and no more (they even slow down the games while producing that eye-candy), plus considering they can't be used for online competitive play (since people without the card will see something completely different to ones that do). In the end utilizing graphics cards as virtual physics cards is going to be the winner, they use the fastest possible interconnect (PCI-Express x16) and have much lower latency then PCI. Graphics cards themselves are pretty well built for processing graphics, all that is required is new drivers to take advatange of the GPU and a game written with the new Havok API with ATi and Nvidia support (future tentative title :P).

However i must end this post because i have begun to wander (someways back) and i might end up on some completely different subject.


RE: Slow And Steady Wins The Race
By saratoga on 6/30/2006 8:34:54 PM , Rating: 2
quote:
true but lets not forget that dual core has other applications, i could encode and burn something to dvd with one core and play a game with another (i also have three seperate HDD's for different uses and im sure 2GB of RAM would be enough for both purposes).


You could already do this provided you had one CPU that was fast enough. Adding 2 cores doesn't change the equation at all. You could have 2 cores devoted to the game, and then let an encoder share them with the game, or you could have the game not be multithreaded. Either way you can still do both at once, so its not like most people are going to care.

quote:
Not to mention you could have one core take care of the crap running behind the game (windows and such) and have one core completely and utterly to the game.


Theres no performance advantage to doing this though (since you can't really dedicate any core entirely to one thing on a PMT OS like Windows, Linux or MacOS), so I don't see why anyone would. If there was a performance advantage, the X2 crush the equivilent 64 or single core FX. But they're essentially identical.

I agree with your remarks about GPUs. If you look at the stuff MS, Ati and Nvidia are talking about for the next 5 years, GPUs start to look a lot like PPUs for many applications. Not quite as good a solution, but probably 80% of the way there, and dual and quad core are likely to pick up the remaining slack.


By MacGuffin on 7/2/2006 11:16:37 AM , Rating: 2
Dude, nobody is here is complaining about the advancement of technology. Everybody WANTS great new technology to come out AND offer a consistent and long-term solution for all parties involved - consumers and developers as well.

Look at high-end audio: DVD-A/SACD both offered remarkably better technology for both consumers (increased fidelity and extras) and labels (better DRM). All they had to do was up the production of more albums and CD would have been on its way out today. But nobody liked the idea of two formats: CD still reigns supreme and the RIAA is still crying hoarse. Those who spent big bucks on upgrading their sound-systems now feel burned over the lack of new discs to play on their setups.

BluRay/HD-DVD is going the same route, some say.

Physics in 3D gaming might die and be a has-been if there is no uniform way to implement it. What a lot of people agree with, is a $300 expansion card for some sparks/explosion effects as in GRAW isn't going to justify their hard-earned money. I wasn't aware that Ageia has promised to support any new standard-APIs if they become available - this is a good step. People just want to be able to enjoy their games with the price/performance hardware of their choice: ATi, nVidia, Ageia or a multi-core Intel/AMD CPU. But it shouldn't mean they are locked out of features themselves.

A PPU should make things faster and deliver higher frame-rates...a general-purpose multi-core CPU should still be able to handle it but at a relative performance hit. Just the way it works in the single/dual GPU world. Currently, PhysX doesn't exactly offload tasks from the CPU to take your FPS from 25 to 35...it just calculates some superflous effects.

I understand that HLSL/SM 3.0 instructions that can be executed on GPUs offer higher performance advantages over CPUs for physics processing. Let's just say I was wishfully hoping that mutli-core would be the way of the future...maybe I am wrong. So I know a PPU will be inevitable in the future.

The main point is consumers DON'T want to be locked out of a major chunk of games that may/will not support the hardware they have. Why should I have to make a compromise whether I want Ageia for Unreal 3 or HavokFX for Source-powered games (or whatever) or DirectPhysics for Alan Wake/Halo 3 or whatever?

Unfortunately, since PhysX is already here...ATi/nVidia will want to capture a bit of the market - good that ATi is backing down. Otherwise, this usually means rushed-to-the-market products that have short shelf-lives and buggy support. The idea is out there...the developments will arrive...and like you said, if I don't like it, I don't HAVE to buy the f---ing card. That's not the point: the point is, I WANT to buy a f---ing card BUT when it is done right.


"I'd be pissed too, but you didn't have to go all Minority Report on his ass!" -- Jon Stewart on police raiding Gizmodo editor Jason Chen's home

Related Articles
AGEIA Responds To ATI's Triple Play
June 9, 2006, 6:00 AM
ATI GPU 2006 Roadmap
June 6, 2006, 3:20 PM













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki