Slashdot Mirror
World's First Physics Processing Unit
Duane writes "Gamers Depot has an exclusive interview with the team behind Ageia - the maker of the world's first Physics Processing Unit (PPU) - which was just announced today.
"Sure we've all heard about the CPU and GPU - that's old hat by now and as most hardware reviewers will tell you, it's about time we got something that's truly revolutionary. Yeah, Pixel shaders are cool, and can do a lot of really nice things; however, pale in comparison in scope to what the PhysX chip from Ageia has the potential to bring to gaming.""
All I know is that I want to throw the dead hooker down the stairs and have her head split open... or whatever that anti-violent game ad says I can do.
It won't be /. worthy news until the first Linux port is up and running on it!
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Nerds around the world rejoice!
Go outside.
Sweet, now I can find out how fast my PC is going out the window.
...that my suspcions were correct. All this 3D stuff with pixels and texels and blah blah blah is just test runs before we create physical augmentation with nanotech replacing pixels and texels. (Wow that was one sentence!) Holodeck anyone? ;P
-"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
...right here. It doesn't really say anything, though - just a few pages that recap physics usage in games, and then a paragraph about how they're going to change all that, etc.
Didn't white papers use to be heavy on technical content? Now it seems that "white paper" just means "nicely formatted eight page PDF advertisement"....
The Army reading list
Note that no-where in the press release does it say that this is a shipping product. Before you get all excited about the promise of this product, realize that this chip may never see the light of day. A press release does not a product make, regardless of how cool the product might be.
i don't have an Advanced Physics Port on my motherboard!!
PPU: Pr0n Processing Unit.
An Indian-American Hindu committed to non-violent thought/speech/action alarmed by the global explosion of radical Islam
And I thought Slashdot "editors" had poor grammar skills! Damn. I guess they're starting to farm the technical report writing and gaming reviews to India now too!
Hmm, this could give some use to boards with multiple PCI Express slots. So, in addition to buying that new $500 graphics card, you also have to buy an additional $500 physics card to play the latest and greatest....
Call me when it works like in the Matrix movie.
Screams on this baby!!!!! 50,000 frames/second! 500x faster then my screen refreshes!
Dashboard Widgets
I just finished reading the article, and this actually has some potential.
The biggest problem they're going to have to deal with, and granted, I'm not a game developer so someone can feel free to fill in the details, is that I would believe that most developers have their own method for dealing with physics - from simple collision to ragdoll and the like. The idea is "How do I tell the computer these things are touching each other' (like bullets - these are "instant shot", so the developer just says "if there's a straight line between the direction the Player A is facing, and if that line would intersect Player B, then it's a hit. If not, then miss." And algorithms like that are done by matrixes, if I'm not mistaken. Other "hits" deal with actual objects (rockets moving, goops from the goop gun, etc).
But the difference between Quake III and Unreal Tournament is more than just 'draw the graphics", it's also in how each engine deals with how those collisions are managed.
So with a PPU, you have to decide on a common library of collisions. Good news: more objects you can play with and let the PPU decide what's getting hit. Bad news: everybody's game will react basically the same and they'll have to decide if that's a good idea.
Either way, I'll wait a year or so and see what happens. Best of luck to the developers - looks like they're at least shooting for something unique.
52 Weeks, 52 Religions with John Hummel
This includes things such as Rigid Body Dynamics, Collision Detection, Fluid Simulation, Soft Bodies and Fracturing of objects.
This will be useful for all those pr0n sites out there!
Attention all planets of the Solar Federation! We have assumed control! - Neil Peart
Get ready for another *phucked* patent. The CPU in a computer is *VERY* capable of doing simulations. Someone explain to me why I need to purchase a *slower* processor with less ram to do it for me? The money would always be better spent on a faster processor...
Religion is a gateway psychosis. -- Dave Foley
I always imagined we would have dedicated AI cards by now...
...but coincidentally the 1,000,000,000th computer accessory that will be a complete failure in the market.
My unit is a wonder of fluid dynamics.
Duke Nukem Forever!!
If this thing can do physics homework, I'm getting two.
Mark A. McBride -- OmniNerd.com
This processor is designed to simulate dynamics of systems other than the inside of a semiconductor chip.
then why play a game?
So which physics model are they using? I'm pretty sure it's not the RealLife(tm) one. That being the case how the hell are they going to upgrade it.
We have some very sophisticated software for newtonian physics modelling. We already have very fast general purpose hardware. Is this add on any more than just neon lighting for gamers?
Yet another reason for Duke Nukem Forever to be delayed.
Considering that most games routinely defy the laws of physics, I would think that such a processor would actually make the games more dull.
Proverbs 21:19
snip
GD: Faith Hill or Shania Twain? (this has nothing to do with their music)
Curtis: Really when you have physical scale as we do it is Jessica Simpson, Beyonce, Mariah, Jewel, Britanny, Shakira, Christina and the other 2.5 billion singers in addition to Faith Hill AND Shania Twain - we want them all.
If you have a game like Unreal Tournament 2004, it is the physics processing that really kills your framerate, no matter how good your GPU. You can see this by simply swapping between the Deathmatch and Onslaught gametypes. The Onslaught world is filled with vehicles which run off the Karma physics engine, and they KILL your framerate, so that the game effectively becomes CPU-throttled, instead of GPU-throttled (which is what we are used to). A PPU is a genuinely brilliant idea, and relatively easy to implement. It will be interesting to see what the programming interface is... and whether the board runs an engine like Karma or something they've invented all for themselves. Prepare to be amazed, I think.
I'm not wrong. You haven't thought about it hard enough.
Now seriously, why would I want to play a game where I need to sit down and rest after running 3 flights of stairs?
The Internet is full. Go Away!!!
I also wonder how it compares to the Cell processor's dedicated units.
For a console, sounds like someone could really steal a march on the rest of them...
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Whereas Havok (http://www.havok.com/) seems to be doing a great job , software wise, I don't find it too strange to let those calculations be done on a seperate unit ; and take the strains/limits off that we have with current hardware.
"Nobel prize-winning physicist, he found the energy source of the stars and campaigned for nuclear power - and nuclear disarmament"
He also worked on the Manhattan Project, as head of the Theoretical Physics Division.
Best Slashdot Co
The Pyschic Processor Unit, PPU for short, is able to know what type of multi-tasking you will be doing before you do it. This was the first great invention of mankind that eventually led to mankinds extinction.
Hopefully once these fall into common usage they'll re-release the "classics" such as Dead or Alive Beach Volleyball. You know, games that could really benefit from this technology.
The Inquirer had a big article on this yesterday. But then again they are always good at getting info out early :)
I think it would be interesting to PPUs in use in robotics. Maybe these devices can give better balance sensors - or provide intrinsic abilities for robots to know/sense how to navigate and interact within the world.
Shouldnt mechs use this to create highly mobile bipedal motion with a good ability to balance in chaotic environments (fast paced combat)
???
The universe is going to be processing physical laws, then BAM! this processor will hit it with a function call. The only thing God'll be able to do is reboot and start over.
If only scientists would be a little more concerned with nested loops.
In what ways do you think your PPU will revolutionize gaming?
For example, imagine you being part of a battle scene similar to the ones depicted in the Lord of the Rings: Return of the King, Troy the movie.
You might want to cut the last 3 words of that sentence.
Pictures of boards are all well and good, and the martketing hype is fun, but we need to know.
Stick Men
This game has dazeling grahpics that look like a film, physics so good you can really blow the walls apart, but an AI that sucks.
I don't want a game that looks, sounds or 'acts' real if it's AI isn't much better than snake's.
thank God the internet isn't a human right.
Take Two would love this if they have a dedicated set of instructions which cover the finer aspects of running over hookers.
Feed the need: Digitaladdiction.net
I'm not a professional developer... that said I'm developing both a robotic simulation API/framework and a game, both in my free time, both *heavily* use the open dynamics engine for physics.
The Open Dynamics Engine is free, & open source. It's not the best physics engine, by any margin. However, being open source I can afford it... and most importantly I can use it on my Mac ( hell, I actually provided some patches to get it to correctly use single-precision trig when OS X.3 came out ). Plus, I want to release my game and robot simulator under an open source license... can't expect people to *buy* novodex or havok just to build the apps.
This PPU looks like a *wonderful* thing, but reading their site, and the interview, it sounds like to use it you've got to use Novodex. That said, Novodex is awesome -- and many games use Novodex already for physics.
(Perhaps I missed something, maybe Novodex is just an API wrapper. Maybe they'll have a low-level API which you can bind to as you want. )
But the thing is, I'd like to be able to buy one of these boards and *not* have to shell out for a developer license for an API which isn't even available on Mac ( maybe it is ). Also, both my simulator and game are intended to be released under an open source license at some point. So, no novodex for me. So, no PPU for me.
Perhaps we're just a little short on data at the moment.
lorem ipsum, dolor sit amet
let's not let the rest of the world know there's something in use called a PPU. I just don't think they'd understand.
Your video card's GPU runs at a slower clock rate than the CPU, but because its pipelines are completely optimized for T&L and triangle filling, it can do those tasks faster than your CPU ever could. Likewise, a physics processor is optimized for simulating the dynamics of a mechanical system.
So nVidia and ATI will have one more product to sell to gamers. If someone is willing to pay $600 for a GPU, you know damn well they'll pay a few hundred for a PPU.
It's now all about the frame-rates. Remember the old days when it was about the game play?
If someone says he and his monkey have nothing to hide, they almost certainly do.
So far I feel this is stupid current for a few reasons.
1) Games do not use Real Physics, they fake it. If they didn't fake it, you wouldn't want to play it.
2) Processors are currently faster then what programs can use(If programmed correctly). It is going to take a few years before games keep up with Processors.
3) Why not just have two general purpose processors. Multithreading is getting pretty common. What would the added advantage between having a seperate processor just for Physics,then having two general usage processors?
Things could change in the comming years, but right now, or in the future I dont think something like this is needed.
mnewberg.com
Can I use it to do my quantum mechanics homework due in 30 minutes?
... I guess this means I won't be allowed to run upside down on the ceiling anymore :-(
Unless they can make a deal with ATI or NVidia and have their PPU work with a GPU, it will be a very difficult thing to sell to people. They also need to get Direct X support and maybe have it work transparently with it (if that is possible). I can see this being a part of a video card, not a standalone PCI card unless the results are incredible and can be shown as a huge benefit to gaming, othewise only the hardcore framerate junkies will buy it.
I'd like to know how important physics in games really are. Personally, rag doll effects are cool but I'd rather have more eye candy than seeing how a body falls down a flight of stairs.
If it came down to running Doom3 at 1024x768 with rag dolls turned on VS 1280x1024 with rag doll off, I would easily go for the higher res.
Yes I know the physic calculations are performed on another chip but that's money I could have saved to get the next highest video card.
It's fortunate that AGEIA didn't release this product on April 1st. I never would have believed it for one second.
If someone says he and his monkey have nothing to hide, they almost certainly do.
Now thats using youre noodle, but it begs the question, is it a limp noodle an aldente noodle, of an uncooked noodle, all of which have different physical properties.
Does a bug in the PPU mean that in fact a Eorpean swallow COULD carry a cocunut from point A to point B , when everyone knows it would require an African swallow ?
In jest though would a bug in the PPU create a situation that could not be escaped in programming a game/simulation based on its "Laws" ?
Oh, Great.. Now I need to get the lastest and greating HPU for my computer. I want the most realistic Hookers, but I dont think I can afford the $500 for the Chip..
mnewberg.com
While a PPU would doubtless be great for almost any type of game (even strategy/sim games could profit), and while gaming does drive a large part of computer innovation, I wonder how limited the success of this chip might be given that there are probably very few uses outside of gaming. No doubt it could help people in technical fields doing theoretical research, maybe even meterology, but up until this point, most of these addons and so forth have had significant (albeit longterm) business applications. Can a chip without somewhat more practical uses make it into mainstream motherboards? I'll be interested to see.
I would be far more excited if they had any information on how this great new chip was going to be distributed. Are they working with ATI or NVIDIA to integrate this new chip onto existing graphics cards? Might be the only way they will make use of the full bandwidth on the PCI-E slots ;)
Or are they working on getting a new type of card out onto the market? And if so, have they talked to the motherboard manufacturers on the type of requirements that it might need?
Nice announcement. And I hope that it pans out, but I would appreciate a little less "Look at our big announcement on how we will change everything" and a little more real information on how they plan to do it.
physical characteristic in that PPG, or it supposed to be provided by the OS (Windows in particular) ?
After all of this time, Havok has been straining to build a beautiful physics engine using the traditional CPU. Now they've got a whole new toy to play with. I'm sure that for a while, we'll see a huge supply of games come out with ragdoll physics, seeing as how that's all we know how to do (that's said "insane backbends and odd positions")... but as soon as we get that "reality" thing down, we'll be all set over here on the gamer side. I just hope we can eventually come out with a decent Leisure Suit Larry. Damn cellshades.
I didn't RTFA (this is slashdot, what did you expect?) but what I'm wondering is how this is different from a CPU being dedicated to running the physics engine? Is this chip somehow optimized for physics? Right now, I can't think how it could be better than, say, having a box with 2 64bit CPUs and dedicating one of those CPUs to physics calculations when playing the game. To me, physics calculations should just be a bunch of relatively normal calculations (probably performed on a matrix) which just about any general purpose CPU should be good at.
SIGFAULT
Games need an A.I processing unit (AIPU). The enemies in computer games are still as dumb as a box of hammers (despite all the developer lies about how they work together, adapt to you, etc). All the graphics and ragdoll gimmicks in the world don't make up for boring enemies.
I read a better article about the AGEIA 's PPU. They claim their card will be available by year's end and will be able to compute interactions of 100x more rigid bodies than a current generation CPU. They also claim that they're delaying release until most of the fifteen games being developed based on this card are ready for shipment.
Also, some nay-sayers posting in this thread have claimed that such a card will make all games alike or that it will force reality where reality isn't desired. That's simply not the case. This is just network in, network out. You could use it for anything that requires a dynamic mesh computation. Games will be free to define their networks/meshes as they please, and they're of course free to ignore the "normal" laws of physics altogether.
The main benefit here is that these games will be able to compute movements of individual hairs or blades of grass in a complex scene. What you do with that ability is up to the developer.
p.s. Personally I expect to see people using this card to accelerate stuff like encryption and (de)compression within a few months of its release.
This product is already dead since everything is moving to multithread/multi-processors. You will just be able to dictate cpu #4 to performing physics calcuations.
I see a new era for the FPS genre ... ::wink::wink::
would be a better comparison than a PPU vs. a single cpu. Assuming the game was multi-threaded so it could expoit multiple cpus. I wouldn't short Intel just yet.
I thought it was funny, but I don't have any mod points.
netbsd could run on this ages ago.
And it is definitely worth a full-on article Slashdot article of its own, but here it is anyway (Coral cache) and (Google cache with appropriate highlighting). :)
This PPU is from October 2004, and is from the creator of the Devtendo and Grand Theftendo. (both = Coral cache)
Buy Steampunk Clothing Online!
with this they could make the HalfLife 3 http://img24.exs.cx/img24/2053/HalfLife3.jpg parody picture real
This reminds me of the Tom and Jerry chips from the Sega Saturn. One chip could great graphics while the other did great music. Alone they couldn't do much of anything.
So how does this relate to this new chip? Simple: This new chip will do great physics processing, but with low-end graphics. I could see nVidia or someone else buy them out and stick the PhysiX chip into one of their card along with (or inside of) their current vector anti-alice-the-maid powerhouses. This would create a dual architecture that would make awesome graphics, but be hard as hell to program for a while.
You say things that offend me and I can deal with it. Can you?
Upgrades to continue as long as we continue to gain knowledge and make new discoveries. Can't guarantee a ship date though. With the Hiesenberg UP module the enemy can't do anything until sighted, detected or measured. Stand still, close your eyes, shoot, winner everytime.
Just look at the file name they gave the article: http://biz.yahoo.com/prnews/050308/sftu104_1.html
This is another example of bussiness folks trying to bullshit investors. They don't even have any specs on their website. How can people be so stupid to fall for this?
You know, this kind of screams of snake oil, and I'm kind of surprised Slashdot posted it. As far as I can tell, there is no technical information on the actual product.
I grabbed the whitepaper, and was disappointed to see nothing about the "PPU's" design. Here's the thing: managing a physics model basically involves a lot of floating point math, which CPU's do fairly well already. It can be parallelized a fair bit, if you know what you're doing, so you could build a processor which can execute more floating point instructions at once (something like what's in the Itanium or the new Cell processor). However, it's worth noting that this is an area that the CPU folks work very hard to excel at. It's hard to imagine that these guys have some kind of breakthrough that allows them to compete, particularly since they seem to avoid discussing it at all. If nothing else, economies of scale heavily favor the CPU guys.
The second thing is that the way this works makes it seem like it could actually end up being slower. Keep in mind that the interaction model is CPU -> PPU -> CPU. So, the PPU is going to have overhead of communication both to and from the CPU. To make this work they'll either have to have a huge level of abstraction (which will limit applicability) or the PPU will have to be dramatically faster than the CPU to make up for the latencies from the hand offs.
sigs are a waste of space
A lot of you don't seem to quite get the picture here, and maybe I don't either, for sure, but what I'm seeing is that this chip will provide a standardized system for how games process their physics situations.
Having a separate processor to take care of an important aspect of gaming such as the physics portion seems to be a good thing. Just think of how many old games(Windows) from the 90's use directx drivers to run their games and how great they were.
This new PPU can make it so that instead of everybody having different computations for their physics processing, now game developers can have that half step already taken care of. You(the game designer) can still decide whether you want your walls to explode when you shoot 'em, you just don't have to worry so much about coding it up.
Of course, I RTFA, but I'm not 100% sure I'm right. IANACE (I am not a computer engineer)
They'd become a physics monopoly... seriously, it would allow Ageia to survive against ATI and NVIDIA, who will pull an EA "lets buy half of you"!
They'll be coming out with a revision in a couple of months to simulate Star Trek physics. It will basically let you change the laws of physics at any point in a game, and will also spit out some technobabble explaining how you did it. (Reversing the flow of the barrion reverter, Captain!)
It's good to use your head, but not as a battering ram.
Soon, we could compartmentalize each separate part of the game code to a seperate processor.
In the ideal world, would the CPU be responsible only for keeping track of ammo and managing the save-game dialogues?
No, this PPU chip will revolutionize gaming if it really work like it is marketed. You play Doom for example, the background really isn't part of the game, just you and the aliens.
With this chip, the background CAN be part of the equation automatically. You can bring the whole level down, collapse bridges that doesn't need to be designed to be collapse-able. Seems too good to be true.
A massively parallel physics engine. I can see this being very useful in the fields of analytic science, where they currently use supercomputers to do similar physics calculations.
I'm curious about this. Admittedly I'm no expert in either graphics or physics, especially where processors are concerned, but doesn't it make sense to have both combined instead of separate?
Now I don't know how a physics processor works but I'm assuming it's something like the following. You fire bullet (x) and player's head (y). The game feeds the PPU all the physical properties of the objects in question and the PPU figures how the bullet interacts with the player's head (snaps back, tugs the body over, etc). Sounds great, but with games using things like bump mapping more and more to distort the shapes of basic 3D objects, shouldn't this come into play?
For instance, if I roll a ball down a hill and it bounces off a jump at the bottom, that's pretty basic. If I apply a bump mapping to the ball that makes it all spikey, well now the GPU is showing me the spikes, but will the PPU take that into account as well? Wouldn't it make sense to combine the two, maybe making it easier for the end developer to work with one single product instead of two?
"I'm a leaf on the wind. Watch how I soar."
-Hoban Washburn
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
This is MY galaxy...go find your OWN!
Sorry to reply to myself (it's okay, I'm AC, just pretend I'm another person), but:
PLEASE do not hammer the original website I linked to in the previous post. I did not warn the webmaster that I would be posting links to his awesome site on Slashdot, albeit in the form of Coral cache and Google cache links.
Moderators and editors: If the parent post gets moderated up, please also mod this post so people are adequately warned. I'm just trying to give a responsible heads-up here. And if an editor wants to post an article about any of Mr. Provinciano's NES-based projects that I linked to here, please feel free to cite AC as the source of the story. Thanks!
This thing really belongs in a console. Dedicated hardware is wasted in most PC's.
Reading the results of the "PPU" is going to be the stumbling block. Graphics accelerators work because you compose the geometry then send it off to the other processor, and from then on you don't worry about the data. You don't have to worry about reading anything back.
I assume having a seperate "physics processor" will mean the app has to send the data off to be processed (say, a couple thousand points to collision-detect against a couple thousand planes), but then your app needs to read the results back across the bus! Is the time saved off-loading these computations going to be worth all this IO?
Today I'm announcing the WPU - the Word Processor Unit; optimised for word processing this will be complimentary to your PC's CPU, steering in the next generation in document preparation. Scroll down for a picture of me smiling and a mocked-up circuit board...
MVI $0018, R0
... er ... slightly more sophisticated. :-)
ANDI #$2, R0
BNEQ @@kaboom
Back in 1979, that's how you would have 'asked' the STIC (Standard Television Interface Chip) of the Intellivision whether MOB (Moveable OBject) #0 collides with MOB #1. Typically, MOB #0 could be a cool 8x16 pixels character and MOB #1, say, a funky 2x2 pixels bullet.
Yup: that's a hardware collision detection, commonly used to drive some important 'physics' of the game and save quite a lot of precious CPU cycles (remember that I'm talking about a 895Khz machine of the early 80's).
And no: this thing is not the World's First Physics Processing Unit! It's just
Just an example among others, of course.
The problem with Slashdot memes is that YOU INSENSITIVE CLOD!
Im not as much into games as i used to be, but i would really be interested in a addon card that could be (ab)used to do some more intersting stuff, like accelerating inverse squareroots for nbody calculations or other stuff.
Those things can be done incredibly fast if you glue them in hardware.
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
I read some nay-sayers, and some praise in the comments so far...
First off, as somebody mentioned earlier: This is a press release, not a released product. Thus, it's still vaporware. No real exitement until I see some demonstrations of it's power.
Second, This _is_ a good idea. What makes a game more fun is how physical objects interact in ways we can't duplicate in real life, like firing a rocket at a helicopter.
Next question: Will GPU Vendors embrase, or destroy this technology? nVidia or ATI could just buy the company and use the tech, or reproduce the functionality themselves in-house.
I'll be looking forward to see where this goes. But, unfortunetly, I'm reminded of Aureal's A3D products, and the way it dissapeared. I fear the same for this technology.
... not the client, in MMORPG's.
The reason is the client can be hacked to allow the player do violate the rules, like walking through walls, etc.
A PPU on the server might help the server support more complex environments, more users per server, etc.
For stand alone games, and scientific applications, this is a pretty neat idea. But I don't think the power-gamers are going to be purchasing extra hardware that just sits there, while the server does the physics.
I can't wait until I have a PPU that can model 200 universes per second.
I can see it now:
PPU Emulates Grand Unified Theory. Physicists surrender.
Vivin Suresh Paliath
http://vivin.net
I like
why will it be a failure?
Interesting project, for sure, but as usual, one should be skeptical of claims of being the first to do anything thse days. First dedicated physics chip for games, maybe ...
... A pulls on B, but B pulls on C, who also pulls on D, who are all pulling on A, etc).
There has been a long-term project involving the devlopment of special-purpose hardware for accelerating N-body gravitational calculations (imagine lots of planets A, B, C
http://www.astrogrape.org/
I'm a bit hard pressed to think about what purpose people may find for this outside of astrophysics, but I just wanted to remind everyone that contrary to popular belief these days, games by no means the only driving force in computer development these days.
Everybody here talk about games... HL2 felt just fine, but oh my god if they could sell some basic physics to hollywood as well! Maybe next Hulk sequel will have gravity pulling the green ape down!
Try measuring frame by frame if the laws of gravity apply to YOUR favorite movie before you say it is SOOOO realistic..
love slashdot. populate it. use it. abuse it. hate it. kill it. miss it. stop following links, they only kill servers.
Normal cpu's already have pretty good fpu units, which are very fast for scalar code. Also, we have things like SSE2/Altivec for vectorizable code. And then there's things like gpgpu.org looking at using the massively parallel fpu capacity of modern gpu's for general purpose physics calculations (linear algebra), i.e. vector processing on a budget.
So where does this thing fit in? As expected, the "article" was nothing more than a thinly veiled marketing blurb, so no info there. Personally, I find it hard to believe that the PPU is competetive with FPU's and GPGPU for general purpose FP calculations. That leaves a chip optimized for certain operations, a bit like MDGRAPE. Or what am I missing?
The next logical step would be to integrate AI processors on graphic boards, so that the graphics card can focus solely on generating those pixels, and the processor can worry solely about running the engine.
Since AI is becoming more and more complex, a chip dedicated to computing its algorithms and calculations would be very useful, although, i think the industry has yet to make AI that powerful...
Where's my *&#%%!$@ cyberware dammit!?! I didn't spend my college years playing Shadowrun for nothing.
This doesn't seem like a good sign to me. Ever since the death of the 16-bit days, game developers have been focusing more on realistic, eye-catching graphics, than making a game that's actually fun to play. Not to say there aren't exceptions, but that seems to be the case for the most part. Theoretically a device like this could be used to improve on creative ideas, but I have a feeling that if something like this catches on, it will create more of an excuse to make a game that aims to sell because of its realism or whatever, rather than because it's actually fun to play.
Doesn't a PCI-E slot have enough bandwidth to include a PPU on the GPU?! If a PPU is a good idea, then it would seem that nVidia or ATI could simply slap a chip on their own cards and sell them for more money.
BTW, never in my life did I ever think I'd say the phrase, "PPU on the GPU"!
If someone says he and his monkey have nothing to hide, they almost certainly do.
But I'm looking forward to racing games using this.
It could enable tires being worn out based upon how you drive, no more stupid signs/bushes stopping you dead when you hit them, maybe even Carmageddon will be redone with crowds of people flying through the air as you jump the curb and take a short cut through a mall?
See my art -> http://herbevore.deviantart.com
I have to agree here. I am finding too that the coding effort necessary to work with a high end (video) adapter is already almost beyond any one person's ability to wrap their head around. All of the instructions and features are already pushing the limits. To add another physical processor that needs even more coding is not going to go very far at this point. The complexity is already close to being to great. I see this as easily pushing things over the edge, so to speak. The arguement to leaving the physics processing in code is that you can modularize responses, and feed the modules with simple information, in the formats that you need, not necessarily the processor. One could retort that you have a choice, either code for the 'PPU' or code for the software physics module, six of one half a dozen of the other. It doesn't seem to be apples and oranges though, from a coding standpoint. The hardware calls are much harder integrate into code in that you have to be aware of things like timing and the coder is left responsible for making sure that the objects layers are rendered in the correct order and filters applied correctly. When done in a software module, those tasks can be hidden from the coder to allow the concentrate on the other important items rather than timing and rendering.
Maybe I am overeracting to having to learn another new code set, but moving physics to a hardware platform scares me to think of the work involved. I think they are going to have an uphill battle getting this thing accepted by the coding houses.
We've already pushed off sound, graphics, and now physics onto seperate processors. Why not just craft an entire game console onto a single card and be done with it? Jeesh.
Oh wait ... I guess that's one more euphamism I can stop using.
Comment removed based on user account deletion
So what happens to the Havoc engine, is it going to be muscled out of business because of this hardware? Their engine has been the most fun to play with in games so far.
Personally I'd rather see engines geared around multiple-CPU systems and with the ability to offload both physics and AI routines off to the extra processors but I guess this PPU (WOW! A PHYSICS CHIP!) is a much more marketable idea.
Isn't anyone afraid of this becoming the new Glide. Games that support nothing but this chip/physics engine. Hopefully an open standard emerges quickly or hardware accelerated physics will go through a lot of growing pains.
That's all pretty thin stuff. I can't help but wonder, though, whether this physics coprocessor will be of any use in doing, um, physics...
mt
If this weren't tied to a single API, I'd buy it in a heartbeat. As it stands, I'll wait. It's like 3Dfx and Glide, giving way to OpenGL and Direct3D. Time for OpenPL and DirectPhysics....
When I can get a general-purpose coprocessor that games can access through a multi-vendor API, I'll be first in line - I'd shell out $100 or so for such a device.
(And yes, I realize I just described a SMP system. Most games aren't multithreaded yet, however, I hope the rise of dual-core changes that.)
With this chip, the background CAN be part of the equation automatically. You can bring the whole level down, collapse bridges that doesn't need to be designed to be collapse-able. Seems too good to be true.
Does that sound like a good thing to you? It sounds to me like a designer's nightmare... far, far too easy for the player to miss their target with a rocket and accidentally blow up the bridge they needed to cross to advance in the game.
Personally, I'd rather buy a board that does super AI, voice recognition and voice synthesis. But this looks interesting still.
If you by specific-purpose chips mean that we will have a few types of chips that each deal with a whole area of different tasks, then you are possibly right.
But I don't think we will have one extremely specialized chip for every single task in a computer. It will most likely be cheaper to produce 20 semi-general-purpose chips than 10 very specific chips, since the volumes would be much smaller in the latter.
In addition, the more general tasks a chips can handle, the easier it will be to balance varying load between different tasks. With specialized chips, half of them will waste time idling, while others have too much work.
All you need for decent physics processing (which basically means core mechanics nothing more)
is a decent fpu good vector processing for matrices and thats basically it.
Are they going to sell, yet another DSP?
Or what is the real difference in this new PU?
To me the new processor by IBM and Sony sounds like the right step, a good general purpose processor (PowerPC) as normal processing frontend and a powerhouse collections of DSPs in the backend which work parallely.
GameSpot has a news story on the Ageia PhysX PPU and a Q&A with Epic Games lead programmer Tim Sweeney on the PPU.
I wonder how long it will be before Pat-Right sues them?
Would it not be simpler to drop all these specialised processors and just produce add-on CPU cards and a good standard way (like OpenGL/DirectX is to graphics etc) of games supporting these devices? Generic CPUs are cheaper and if the whole thing could be sorted out well enough then people would be able to use their old processors and even old GPUs, it just fits in with the whole idea of PC's being general purpose?
This comment does not represent the views or opinions of the user.
Having a dedicated physics processor seems like an interesting idea, even if the press release here is just some vaporware.
For such a processor to succeed, we probably need some similar properties that have driven the success of the graphics processor.
1. The problem should be embarrassingly parallel.
2. We need to reduce the most common physics processing problems to a simple API and data flow model.
3. The above data flow model should offer some advantage over just crunching the data on a second or third CPU.
We can do various types of physics simulations on the CPU and the GPU today. In some cases, we can get significant speedup using the GPU, especially if we can minimize readback and redundant computation. To drive adoption of a separate PPU, it had better be possible for a more customized architecture to significantly outperform the GPU or cost much less (otherwise we could use a dual-GPU solution instead).
With dual and quad based CPUs on the way, I doubt this type of thing being popular. An extra CPU can very likely deal with most physics required for next generation games.
Considering that most of the next generation games will be already programmed for Xbox2/PS3 consoles supporting multiple CPUs, all the extra work required to get the PPU support in games might not be all the rage among game developers.
endless posibilities,,,my head hurts!
I did RTFA, and didn't see an answer to this one. I was wondering if this could be used to accelerate simulations/calculations for real-world problems, as well as for games.
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
The fact that there's a physics API (NovodeX) that games are already using made me immediately think of the cell processor. This is exactly the kind of thing vector CPUs were originally developed for, back in the Mainframe/Supercomputer days.
When people started talking about the cell processor and the fact that it was PPC based... and tying it in to Apple... the only think I could think of for Apple to really do with it would be to use it for speeding up Quartz and OpenGL. But how about a Powermac with a cell-based coprocessor doing NovodeX acceleration. The Mac could be the gamer PC all of a sudden.
That sounds cool and all, and I can't wait to play with it, however I'm wondering about how free (as in both beer and freedom) their SDK will be. I hope they won't be charging some insane price for it. Open source would be nice too, but I'm not holding my breath. At the very least, I hope they'll publicly release enough specs to communicate with the PPU outside of their SDK (and targeted systems).
Games become more successful the more they are based on reality.
Politicians become more successful the more they avoid reality.
Games: better than politicians?
FredRated
Stupidity: it's a renewable resource!
First thing that crossed my mind was that ol' Wile E. would buy one of these, integrated into a helmet to allow him to run through tunnels painted on walls and allow him to walk through the air and catch the elusive RR...
-- If god wanted me to have a sig, he'd have given me a sense of humor.
This reminds me on when I used to do computer-aded chemistry, and by far the most time consuming operations was computing second derivatives on non-bonded interactions. Somebody found out that there was a bunch of NVRAM on one of the microvax processors, and decided to experiment with putting the calculations in there. I guess this was like using a FGPA approach. It never happened as far as I know, but it was an interesting approach nonetheless...
Call me old fashioned, but I like a dump to be as memorable as it is devastating - Bender
You don't need specialized hardware to run 85MPH in a first person shooter, and unless they come out with some benchmark for some number of "physics operations per second" that makes software physics engines seem slow, they'll never get any fanboys.
As I understand things a GPU has a purpose because graphics is very heavily vector oriented and most of those are interger vectors. And as CPUs are not designed to be vector rockets because its not practical the GPU was born. But from my understanding Physics modeling is quite fpu intensive and also is not able to be parallised as easily as graphics so will this realy provide such a huge benift or would spawning a new thread for the physics and placing it on a different CPU to the game be better in the long run with the advent of wide spread multi core machines?
I just read the article and am now thinking "why wasn't such a device conceived of much earlier?" It sounds so natural to develop the physics processing into its own dedicated hardware.
I'm hoping the Linux community can come up with an API that's attractive for game developers to use. It would be best if it could be integrated into a set of APIs to do 3D graphics, 3D sound and music, hardware access to input devices and have a common programming paradigm for its use. Perhaps SDL or then OpenAL/GL group could come up with something.
Hi there, The chip works. It will be in stores this fall. We already have compelling content that will ship with the chip. See you at Christmas!
Today's kids have measureably less creative skills than in the past, because corporate america is doing an ever better job of shrink wrapping imagination.
Back in the day, there were Lincoln Logs, Erector Sets, Legos, and many other toys that had an inherent creative context. Legos are still around, but the most popular kits now have some kind of licensing tie-in.
In fact, the most popular toys for most of the last 20 years have been licensed. In the last few years, witness the Pokemon and Spongebob crazes.
Video games are most directly damaging to a child's imagination, because now the child doesn't have to imagine anything. Their mind's eye is transplanted onto a screen. Not to mention that video games today have increasingly less replay value, having become more like interactive scripts and requiring less and less problem solving. If the hero dies, go back to the last save point and try again. Video games have been sacrificing game play for more visual, aural, and physical believability.
Today's kids have computers to think for them, tomorrow's kids will have computers to imagine for them.The chip will be a stand alone PCI PCI Express card. It will be significantly less than $500 US.
did you hear? duke nukem forever release is timed to match this chip's so it can use all t3h new features
See pictures of tits
What is the biggest difference between having a building break into 10,000 peices or just faking the effect and have it break into 100? I am a geek and I love thinking about doing things like this, but I also am a realistic about what people want. Is this concept such that people will actually be interested in putting money towards this? I clearly understand that there could be improvements in games becuase of this chip. I just feel better designed general purpose processors would be more worthwhile, then a processor design just for Physics. I have heard people workign with AI are interested in a seperate processor. At what point does one say, enough is enough, how many different processors need to be in a computer.
mnewberg.com
They came over to our office and did a demo a few months back.
There's no hardware yet so it was more like a software showcase of what COULD be happening. You have your basic explosion and rag doll stuff. The ones that showed best are a corridor flooding with liquid fire, a dense system of cogs and gears that worked flawless, and a tall building that collapse with thousands of pieces of debris bouncing off each other. The physics can be turned on and off in real time and will in turn generate different outcome.
We questioned the process of integrating the chip into the market. It will be the chicken and egg conundrum. The manufactures and developers will both wait for the other to create the demand for it.
The most straight forward solution seems to be to convince a console developer to include it in their next gen console.
Yeah, but can I SLI two PPU boards ...
Let's compare all your responses to the benefits of having a GPU for graphics...
1) Games do not use Real Physics, they fake it. If they didn't fake it, you wouldn't want to play it.
But they should fake it in a way that is believable. Traditionally, computer graphics is a bunch of hacks for image generation. Even the hacks benefit from hardware support, and people desire more and more impressive looking graphics, whether or not they are trying to be "realistic".
2) Processors are currently faster then what programs can use(If programmed correctly). It is going to take a few years before games keep up with Processors.
That's a load of crap (IMHO). Game developers have to dumb down the visuals, physics, and AI of the game enough to make it reach their target frame rate. Frame rate takes precedence in today's market. What makes it even harder is that the users all have different speeds of machine, so in many cases, the game must be dumbed down enough for the slowest machine to execute (or they need various computational levels of detail for different machine speeds). If you see that your CPU is not busy enough, it is because the game developers dumbed down the physics and AI enough to make it so. Similarly, if your GPU is not working hard enough. It's actually a bit of a chicken and egg problem, because the game developers will not let the games use more resources until there are machines that actually have them. This is a bit similar to the PCI-Express problem.
3) Why not just have two general purpose processors. Multithreading is getting pretty common. What would the added advantage between having a seperate processor just for Physics,then having two general usage processors?
Presumably because the PPU should be faster at performing these specialized operations than adding a similarly priced additional CPU. Again, a dual processor machine does not eliminate the need for a GPU.
Its funny how the PC is gradually turning into and a jucied up Amiga computer.
//Amiga4Ever
So this is what all the amiga crackpots ment when they said that the Amiga would rise again. =P
http://www.theinquirer.net/?article=21648
But without some sort of common API standard, I think that PPU technology, interesting though it is, will not catch on.
The actual problem here is a chicken-and-egg one, unless they get this built into a next-gen console. If the performance gain of the hardware (and I have no idea, so I'll pick two values and speculate on them for illustration) is only 2x or 3x, then you can drop it in and get a performance boost, and maybe that's cool, but it's going to be hardcore-only to bother with. If the performance gain is 10x or 100x, then you can do a lot more crazy stuff--but only on machines that have the hardware, and normal machines without the hardware just aren't going to get anything like the same gameplay experience, unless the physics is used entirely as a special effect and doesn't feed into gameplay.
Of course, the PC confronted the same chicken-and-egg problem with hardware in the past; we had PC speaker sound, then AdLib FM music, then Soundblaster sampled audio. We had CGA, EGA, VGA, mode X, SVGA, and then hardware-accelerated graphics. But, in fact, audio and graphics are output-only; they're close a feedback loop between the game and the player, but they never feed back into the game logic, so it was always possible to compromise on them... and we were never looking at a factor of 100 difference.
Whereas the console industry is littered with failed add-on peripherals--Sega CD, 32X, I don't even remember the ones from Nintendo. Karaoke Revolution comes with a microphone; AntiGrav comes with the eyetoy camera--because trying to sell the peripherals separately is so totally doomed. I doubt any games are going to be sold with a so-called PPU, so it remains to be seen whether, chicken-and-egg-wise, the physics chip (purely computational) ends up more like a peripheral (input device) or sound/graphics (output device).
Maybe the better analogy is to the inclusion of an FPU. On games that had to work on computers without FPUs (you remember the 386?), the software simply didn't use floating point. It didn't make sense to write two versions of the code. Instead, you just target the lowest common-denominator and optimize for that. If someone has a faster/more-capable machine, well, it will be faster than the machine you optimized for, and hopefully that's good enough.
Now people can start doing numerical simulations really fast!
Evolution of Language Through The Ages: 6000 BC : ungh, grrf, booga 2000 AD : grep, awk, sed
This will surely improve the realism of boobie bounce. :)
"I'm a leaf on the wind. Watch how I soar."
-Hoban Washburn
"Ageia's PPU is an important step in the right direction as it can take the current limit of 30-40 bodies of today's high-end CPUs to a maximum of 40,000."
So its going to 10000 times stronger then a high end cpu, you think they would try to sell it to businesses/schools/goverment agencies that need REAL PHYSICS analysis (aka NOT Counter Strike)..
Am I missing something here, Wouldn't this thing take over disributed computing for physics simulation?
Crah testing, Vehicle dynamics, Flight characteristics, Blowing crap up, High End 3D Animation, etc.
I don't understand the reasoning for going after bunch of dorks...
May be they have some other Business Plan, one that makes sense.
Next, they will develop specialized chips for enemy AI. Hope that your enemy doesn't hack it's way out of the chip and set your toaster on fire.
I just checked the US Patent Office website, and there are no patent applications for ANYTHING from Ageia. It's possible they've filed very recently, but I would think that if you had a hardware platform about to be released, you'd have had IP in the pipeline for a while. Sounds like they're just announcing their plans to try something. It's a cool idea, but lots of ideas sound cool before anybody tries to do the layout.
Everyone's favorite "analyst", rob enderle, has a choice quote at the end of the article here
thanks for the insight (and grammar), rob. i dont mean to flame, but charlatans like this being quoted incessantly in all manner of technology-related articles disgusts me - and he rules this blowhard-for-hire tech analyst faction on high.
cogent, pertinent, well-written, intelligent - rob, just apply ONE of these attributes to any of your published ramblings. please.
Actually, I think the current fixation on FPS isn't all that exciting. Amazingly enough, there are other game types out there. I can spend 10 to 15 hours a weekend flying planes. Put PPU assisted realistic flight dynamics and scenery to make the plane's environment interactions even more real, and I'll likely cry just waiting for the next version of FlightGear or Flightsim that will work with this. Or baseball, by being able to add realistic movement based on current wind conditions. Golf, simulating every blade of grass the ball hits... There are other examples I suspect.
Also, I think the Cell processor might be the first attempt at a single chip encompassing multiple processing units. Maybe they didn't think along these lines exactly, but from all the blurbs, I don't see why it's not possible.
Overall, cool idea really. If it a $200+ addition, I won't buy it. If there isn't an open implementation, I will probably ask why we can't do an open hardware implementation in an FPGA or somesuch. It's going to follow market forces, especially in the game industry where the only thing that mattes is how cool is the next big game and how important is this PPU to making the next big game as cool as it could be.
Bah
"A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
I guarantee that this thing will fail if they market it as the PPU, also known as the "who farted?" chip.
I don't this taking off because no one would buy it. Not "no one" but really so few people that it won't make a difference. With a graphics card upgrade to a better card you get prettier pictures. People always want prettier sh.t so they keep on buying newer and newer graphics cards. I just don't see people overclocking their PPU and upgrading it. Will a game developer offer a slider for how much physics the game offers like eye candy now? Well, frankly the physics affect the game play more than graphics, for example if I can't jump as high or my grenade bounces differently then it really changes the game. Maybe if they attached it to the GPU so you'd have a GPU+PPU combo card (w/ 2 power connectors) you'd have people getting newer and never ones. Maybe this will be a hit in consoles though...
Your CPU is not doing anything else, at least do something.
http://www.gpgpu.org/ says that the latest GPUs can already accelerate physics a lot. If that's so, how much is a dedicated physics chip needed?
Does this mean I'm going to have to buy a PCIX Physics Card on top of the 500 dollar Graphics Card to play the next generation of games. That would suck a lot. Here's hoping the ppu becomes standard on Graphics Cards.
"At what point does one say, enough is enough, how many different processors need to be in a computer."
:)
Never - you are clearly not a gamer
I could see it being very useful for all sorts of simulation and CAD, and various scientific applications.
I'm not a professional game developer but I'm a techie who has spent some time working with game engines and such. I like to play around. Anyway where I see a huge benefit of this is the amount of dynamic objects a game could have at this point. Right now there are lots of tricks in game development to achieve realism without sacrificing peformance. Just look at any FPS and most of the world is a static mesh, there are very little movable and/or dynamic objects in the world. This is because it requires CPU power to keep track of everything all at once and you still have to reserve processing power to handle the players and the AI players. So where I see this coming into play is having thousands of dynamic objects within a game. Imagine being able to blast a whole in any wall in the map. Having anything visable have properties that are changable depending on what is happening. Say your playing a monster truck game and you crash into the stands, well then every spectator would have its own ragdoll property that would apply there. These are just a few examples, theoretcially you could do this today but your framerate would crawl at that point. Anyway these are the types of applications I see being used with it.
Perhaps this is the Final Solution to Digital Rights and Content Management problems.
Have no content.
Does it run MATLab?
Granted FS are used by few people but it is one type of game which is CPU-bound currently, no need to invent new types of objects: flight dynamic, bullets, etc.. are enough!
:-)
In IL2 Sturmovick, IA's opponent use a different physic engine which enables them to do figures that you can't imitate, which is a bit frustrating..
But I'm not sure that if this accelerator allowed using realistic physic for IA's opponent, the IA would be good enough to fly a plane with this complex physic..
Ok, who is going to make an IA accelerator board?
On a more serious topic, I'd be interested to see what they're going to implement, while I don't really beleive that a standalone PPU will be a success, if the HW is not too different from a GPU, maybe a combined GPU-PPU would be interesting: a bunch of 'normal' FPU for doing vertex|physic process + some specialised unit dedicated to the pixel processing part.
When these become commonplace, I bet we'll be hearing press releases announcing premature failure of a space rover after incorrectly installing the proper physics chip.
"Nasa is sorry to announce that the latest lunar rover had a Mars physics chip installed which resulted in behavior akin to a Nascar hot rod. After breaking the moon's gravitational pull, the rover struck the Hubbel telescope. South Korean residents are instructed to saty indoors in case the re-entering space junk is mistaken for a missle attack by the North Korean army."
This chip would be an amazing benefit towards the physical modeling of sound propagation environments. Processor lag in time-based audio effects could be almost eliminated. This would be fantastic for audio processing!
At what point does one say, enough is enough, how many different processors need to be in a computer.
When I run out of things for those processors to do. I have a CPU, because liek duh, I need one!
I have a processor on my soundcard, because it assists in sound processing, something I do a lot (in games, of course)
I have a GPU because it assists in graphics processing, something I do a lot (in games, of course)
I can imagine having a physics processor, because physical simulation is something I do a lot (in games, of course, although I do do a little of my own coding in this area from time to time too)
Why would I want all of these things in a single processor, if multiple independent units can do it better? Why would I want to swap out a single, hideously expensive unit to upgrade one aspect, when I could just swap out that one component?
The way I see it, we've gone from having a single CPU to having a CPU and graphics hardware and sound hardware; why not physics hardware, AI hardware, encryption hardware, etc? The point is that a CPU, by its very nature, cannot ever be as fast at a given specialised task (eg rendering graphics) as specialised hardware. The best you could do would be to stick seperate cores into a single package; but why bother? That raises all sorts of problems (especially heat - the GeForce 6 line runs very hot, for example), for what gain?
It's official. Most of you are morons.
First of all, nice comment!
With modern games, GPUs take care of the graphics, which probably means that the biggest task the CPU is performing is the physics. If that were offloaded to a PPU card, that leaves the CPU to basically handle the "playing" of the game. So, I'm thinking that if/when these PPU cards take off, the CPU could be used to implement much more advanced AI algorithms than is currently possible. With visuals and physics taken care of, I'd speculate that intelligence and "interactivity" might be the next big challenge - enemies that can plan intelligently and adaptively, allies that can follow (complex verbal) orders, etc.
So wonder how long it will be before instead of getting an Intel processor with two general purpose CPUs on one die, we will be able to get a CPU and a PPU on one die or a CPU and a GPU on one die. Maybe we will eventually have the choice of having all 3 cores integrated into a single die. I suppose this would provide greater bandwidth between the various processors and would allow us to use one bank of easily upgradeable memory. The reason I add the memory thing is because I am not aware if there are video cards out there where it is convenient to add memory capacity, I know my cheap card doesn't have an easy way to add memory. I suppose I could solder on some chips to the open pads, but I'm not sure if the GPU would automatically recognize the memory or not.
I meant AMD or Intel of course. Sorry, the audience slipped my mind. I don't want to be modded down to flamebait for throwing Intel out there like that :)
It seems that it would be pretty unlikely that this product is pure vaporware seeing as how there's some pretty substantial developer support. (link) It's pretty difficult for these guys to get developers on board - ATI and NVIDIA both have issues, and they're industry standards. I highly doubt Epic, Ubisoft, Sega and the others that have come on board this early signed deals based on pure speculation. Add to that the reports that the product is to be available this year (link) They must have some sort of prototype already functioning.
I don't get what is so different between a GPU and this PPU thing. A GPU is mainly multiplying vectors and matrices and dot products and division, physics simulation is not very different, although the result is handled differently.
Quite differently. You are suggesting that one draws the matrix results and the other just stores the matrix results, but there is a more important factor than that. All the data that is pushed onto the grpahics card is essentially on a one-way trip. After going through the T&L pipeline, it guts pushed into a video buffer, drawn and then overwritten. I'm not even sure that there is capacity to write back data to general memory from the AGP cards, since this incurs a big performance hit. Any physics chip will need to be able to matrix math and then save the results for further useage, since you need to maintain things like mass, velocity, heading, etc.
Using a GPU chip as the starting point for a PPU wouldn't be an entirely bad idea, though. Interestingly enough, I think the rendering community would love to have a 3d accelerator that could use specialized GPU rendering speeds and the ability to write the image generated to the hard drive, so it's not like you are the only person thinking along those lines.
HA! I just wasted some of your bandwidth with a frivolous sig!
"Why would I want all of these things in a single processor, if multiple independent units can do it better? Why would I want to swap out a single, hideously expensive unit to upgrade one aspect, when I could just swap out that one component?"
Because a very large percentage of the costs are in the circuit board, support capacitors, clocks, shipping and marketing of all the individual items. A single processor (with multiple cores) may cost you $400 or so with all of that, but the other way you would add a $250 CPU, a $200 graphics card, a $200 physics card, a $100 sound card. If you want to pay a cost of $750 in comparison to the $400 so you can mix and match, you're welcome to, but I suspect many people may prefer the $400 price tag to the $750.
If this is going to be successful at all, I highly suspect it will at least have to be included on or with a graphics card by one of the big graphics card companies.
...something more interresting than GAMES ? ...something we could use ?? like calculating a comets way thru the solar system...
...or calculating a tsunamis way, speed and size faster than the tsunami moves... so we know where and when the tsunami strikes and how big it will be...
I would assume that the gaming market is more lucrative than elsewhere, or that the company is betting on this assumption. As long as there is developer support for PPUs, especially with more popular games, then many gamers will feel the need to buy PPU cards on the knowledge that the game will run better with one in the box.
As it is, with the real physics simulations chances are that distributed computing will stay, perhaps further bolstered by any advancements offered by the new PPU technology. Something says to me that a 40k body limit would still be restrictive for massive simulations run by government agencies and the such.
Christopher S. 'coldacid' Charabaruk -- coldacid.net
Admittedly soundcards are total pieces of shit (yes even that top of the line creative), but if sound is supposed to be 30% of a game, why is ALL the horsepower spent of graphics and physics??!!
Anyone else see multi-core processors as a possibility here?
One core - CPU
One core - PPU
One core - GPU
One core - SPU (sound?).
It's a 4 core gaming system (not upgradeable, but in the not so distant future, likely quite cheap...potential for handheld gaming devices?)
What'chu lookin' at Willis?
Second Life does all the physics using the Havoc engine on the servers. I can see where using a dedicated hardware-based physics engine could improve Second Life server performance.
MAC | A polar bear is a cartesian bear after a coordinate transform.
Including exactly what performance will mean with this chip... Link.
CAD/CAM software, phys sim software, supercomputing........
All can benefit from this technology, greatly. May not sound so much like a benefit to all those in the industry, but as a person who is very much interested in CAD/CAM and Phys. Simulation software for the crackpot inventor at home......this is a VERY fortunate upcoming.
Of all the Universal Constants, here's one I know: Nice guys finish last
Macintosh? At 9.8 meters/second squared?
Tag lost or not installed.
In the SDK Documentation PDF on their site (available from http://www.ageia.com/novodex_downloads.html) section 5.2.9 on page 33 mentions the solver used. Does anyone know if the solver is an n-th order Runge-Kutta method, or do they simply subdivide each step into n euler steps?
The question of whether a computer can think is no more interesting than the question of whether a submarine can swim.
First off, as somebody mentioned earlier: This is a press release, not a released product. Thus, it's still vaporware.
So that's it. They're pre-announcing it so they can get response to it from Slashdotters, and add in all the stuff they forgot, such as one thing mentioned several times in these comments:
"OMF'inG!!! We forgot Gravity!"
And while they're putting physics ona chip, perhaps they could take a hint from my tagline...
Tag lost or not installed.
...how this improves my Solitaire experience!
Anakin Simpson: If you're not with me, then you're my enemy--ooh, donuts!
Because Newtonian Physics is old-fashioned.
Seriously, it would be neat to have a video game in which one goes at a significant fraction of the speed of light. I recall seeing a movie clip many years ago (perhaps on NOVA) of a simulation of relativistic travel down a road in which the telephone poles appear bent over and such. Realtime simulation would of course take more calculations per 'atom' in the scene than Newtonian physics, but it would be cool, and We Have The Technology...
Tag lost or not installed.
This PhysX seems ideal for a console:
*it would dodge the chicken/egg thing, developers can know that all of the consoles will have physx
*it would give the console a massive leg up on competitors without PhysX
*it would give people time to get used to new impressively accelerated physics
*could use cheaper cpus and still have superior game performance
Oh if only they could get it into a console!
There's no one key item that bottlenecks a rigid-body physics engine, and it's not a simple pipelining problem like graphics, so the main thing special-purpose hardware can provide there is parallelism. And plenty of double-precision floating point power. (In a single-precision system, you have to take great care to never try to do physics far from the origin.)
Collision detection is a minor CPU load if you do it right. If collision detection is using more than 10% of your physics time, you're doing it wrong. This may seem counterintutive, but the good algorithms are incredibly fast, even in complex environments. It's more of a data structure issue.
Deformation, i.e. finite element analysis, is more of an inner loop crunch problem than rigid body physics. Finite element analysis has been parallelized for decades in engineering applications, and the problem is well understood. It's localized; you can divide the problem up into cells. So I'll bet that's what they are focusing on.
Everyone remember to keep their mouths shut up about this to M$. Otherwise they'll tack on some new "feature" to Longhorn or Blackcomb that not only makes windows 3d, but also simulates realistic collisions between those windows:) It won't be a resource hog, of course....
"As you say - certain behaviors minimize the HIV risk and writing Slashdot tripe on Friday night is by far the most secu
Really to simulate accurate physical behavior needs floating-point processing, which I doubt this hardware will be anywhere close.
I'll be waiting for the specs.
Well, if this technology really takes off, I guess I'll have a secure job (in physics) once again.
The system is already been cracked; some Jesus freak named Yeshuah said that all you gotta do is file a death certificate and denial of corporate existance and you'll realize that you have alreadyy been born again as a spiritual manifestation of God and not under some alien's statute clause. Don't accept any ploy, because all ploys are of man! It's a paper and pencil idol conspiracy of idolaters!
Or, you can take the red pill and liquidate your corporate sole CITIZENSHIP in a savings to suitor ruling, returning gracefully onto soil from your CITIZENSHIP vessel; under God the Father and you can judge and file default judgments and site drafts and liens. And you get free energy, free medical, and none can be given or taken to marriage.
Instead of all of these dedicated CPUs for processing graphics, physics, audio all on separate pieces of hardware, I would like to see a consumer media expansion card that has a generic array of CPUs that can be dynamically configurable by the operating system for specific tasks.
For example, suppose I have an array of 5 simple CPUs. In a configuration for games, I can have two CPU for graphics processing, two for audio processing, and one for physics. Or, in a configuration for advanced audio-video processing, I can have three processors for video and two for audio processing. The possiblities are endless.
But AFAIK, there is really no technical reason why this can happen. Maybe I am expecting too much for my relatively cheap PC.
Coderz 4 Life
..Because Sony has so much more cash to throw into optimizing the CPU architecture and the production process. It is difficult to imagine a better "physics processing unit" than the Cell. It would be possible to implement specialized hardware for a set of common processing tasks, but then you'd lose the versatility that a fully programmable design has..It's a bit different from 3D graphics where almost the only thing that matters is drawing textured polygons. Physics simulation tasks are more varied.
Is a specialized hardware raytracing processor on graphics boards. That or built into the GPU but I mean, doing actual real raytracing.
I saw a webpage where some guy made a raytracer with a FPGA or similar and got quake or quake 2 running and using his hardware tracer and the shadows etc were doom3 like (it didnt look that good, but it did look great for quake), at a decent fps. And this was just some guy. If ATI/Nvidia actually developed a nice hardware raytracer onboard their cards, plus this physics processor, I think we'd be pretty damn close to the goal of movie-like 3d in real time.
Ah well, one day.
It's easier to fight for one's principles than to live up to them.
The problem isn't necessarily one of processing power (although specialized is definitely faster than generalized), but of bandwidth. Separate processors and addons (usually) have dedicated memory, cache and lessen the calls to system memory and hard drive access. Even new system memory doesn't have as much bandwidth available total as the specialized memory for graphics cards.
Also, your example of pricing does not reflect that by not having a separate graphics card you are giving up the ability to play most games. Lacking a separate sound card/physics card etc does not currently mean giving up this ability completely. Price per performance has to be equivalent for the comparison to be valid. If you don't want to play games then buy the cheapest thing you can find as it'll do office work, internet just as well.
I am not in anyway affiliated with Max Cannon