Carmack: Hardware Physics A Bad Idea

But PhysX is one of the reasons I like nvidia.

Are we allowed to breathe now? I mean, John Carmack is done talking, right?

how much of a difference do they make?
and how many people actually use PPUs?
I have seen them in rigs from Falcon Northwest and stuff but are they worth it?

Well, where I agree that "Physx cards" were a bad idea, since you payed for hardware that only showed performance increases in select games *like 4 lol*, now that new series of GPU's come with Physx support built right into the GPU, I think it's great!

Lets face it, id software is a little behind the time anywho, even when they released Doom3 and quake 4, the games were pretty stale. The games had decent graphics for the time they were released, but the games were nothing special, and for the part pretty boring. And what have they given to the gaming community since then? Nothing. While other game companies are pushing forward with fresh new ideas, id is sitting around talking trash about how bad physx is.

I think this can all be summed up in 1 sentence. Id is sore that they can't invent anything new and fresh, so they attack new ideas that they wish they had come up with.

i think carmack is right to a degree, a seperate ppu unit does suck, but what nvidia did by intergrating it into the video card was the way it should have always been, and maybe with a combo of using an available cpu core that isnt being used by by a game or application it would maximise what we can do with physics without the need to spend extra money, use and available slot and consume more power

and to answer Upendra, No one buys PPU's because all Geforce cards from 9 series and up come with built in PhysX support And those have been out well over a year now.

I believe there is potentiality in physic cards but its really up to the developers such as id to decide to actually make it worth while or not the hardware is in the complete mercy to people like Carmack. However since the merger of physics cards and GPU's this certainly makes it much more affordable to the consumer and developers that take advantage of it have a edge over those who do not.

I do however believe that Carmack's recent love for consoles influences his opinion on this subject. If a game was released today that had ground breaking graphics and to top it off a very advanced physics system that really took advantage of physics capabilities the game simply wouldn't be capable of running on the consoles we have today. With more developers looking to release on both consoles and PC this wouldn't be in their best interest.

Totally right, I think the same thing. PhysX is impressive, but it was far from really more impressive than Havok was when we first saw Half-Life 2.

I am sorry, but why do we need hardware to run physics? Havok was doing a fine job and it never required any additional physical hardware to be run.

What will happens with DX11 around the corner?

I think that with gpu having physx like my 8800gt it will be good, now i don't have to sell it or just throw it out i cant turn it in to a dedicated physx card when i buy a 300 Geforce cars. then keep moving the last gen card to physx or as needed. i think physics done right could change gaming a lot and make it more fun as well as help with graphics by helping with different effects to make them look more real.

Nvidia's Physx on graphics cards is the right thing I think. You don't actually have to spend extra money on additional hardware taking up vital PCI-E slots. Nvidia's Physx actually makes a pretty big impact on games that support it. So his argument is pretty late to the game, we saw PPUs die a very quick death over a year ago.

Maybe Carmack is just angry at physics 'cos his rockets aren't working.

http://www.armadilloaerospace.com/ [...] ness/About

Ok people, this proves it! The iPhone does NOT have a physics chip.

Well, dedicated physics cards are pretty much gone so it's kinda like a pointless rant. After all GPUs are more flexible than the ol PPU which was designed to do a specific set of calculations. As long as there isn't a crossplatorm GPU accelerated Physics API, ie doesn't rely on hardware from just one vendor, it will be just a nice thing to have rather than something mainstream.

physX is awesome with games that support it. Makes the game feal that much more immersive.

Finally some big name says PhysX is crap. As long as ATI won't touch it, it will eventually die.

When will dedicated graphics hardware be gone too?

I don't understand why all this hate geared towards Carmack latelly (or all the media focus to begin with)... Tim Sweeny from Epic Games has been saying things like "in the future real-developers won't use DirectX or OpenGL, they'll write their own low-level APIs" and "in about ten years the GPU will be gone, replaced by massive CPU arays"... so either the true gurus are right in their insanity or just insane.

BTW, ID not inventive... right... first 3d engine, first FPS, first deathmatch game, first real-time shadows game... yeah... I suppose. How many GOOD PC games came out this year anyway? How many Inventive oones then? How many years since you've seen something you could say "wow, this is something totally new"?

It's not Carmack's job to make good games, he's a programmer, he's not responsible for story line, modeling, level design, or whatever other aspect of development.

All Physic effect on PhysX i had seed On old games Like Straglehold2 and TomClancy Rainbow six Vegas 2(yep)
^^BTW all that games work perfectly with my IGP
.

Yawn...Yeah, we all know how well four cores are being used in games. LOL!



Carmack is sore because he chose not to play and now doesn't get invited to the parties very often.

Really it comes down to what kind of games do you want to play. Highly submersive details require hardware to do it properly, you want to add that to the CPU?

Modeled Physics presents a challenge to be convincing after repeated gameplay, if you dont care about those kind of details than Carmack is right, if you want theose kinds of details then doing the physics has to take place somewhere.

When we end up with 28 cores this will be all mute. From a developer perspective it is cheaper to develop a game to maximize CPU resources than have to pay license fees to use some extra technology that may not be available to most users.

Any of these "physics demonstrations" I have seen have no impressed me one bit. Like that game where you are running around from rooftop to rooftop, supposedly the whole game is based on physics. MAYBE the flags being shot apart is kinda cool but that does NOT make a game COOL or even GOOD for that matter. Jeesh, talk about a FAD.

Wow, this is so stupid. First, he is a programmer and as programmer, we are very lazy and prefer quick and dirty solutions, like emulating everything. YES, havok is very nice because you can compute all physics with your CPU but at a very big cost... no matter how well optimized and how many core you have, a PPU will pwn havok on both cost and performance. Intel owns Havok so they want you to buy over sized and over priced CPUs from its monopoly even if a small cheap PPU could do the job. That's why it's very bad that intel acquired havok. So, why not just throw all our GPUs to the garbage and replace them with crazy 10 ghz CPUs to compensate? We must focus on GPU and GPGPU because they are the key to achieve computing power at low costs for the masses. OpenCL is very promising.

Hardware Physics acceleration sounds great on paper, but I don't know. I actually would say that Carmack has a lot of expertise in this field as he is one of the highest regarded game programmer. I do believe him since CPUs still have a lot of game in them and with the advent of multicore processors, I think physics can still be calculated on the CPU fine. I think NVIDIA may be overhyping physics on GPU ... then again, maybe because NVIDIA wants to move more units so they keep pushing for physics on GPUs? Both AMD and Intel back Havok, the only company that backs AGEIA is really NVIDIA here. Sure, Havok is owned by Intel and NVIDIA owns AGEIA, but I think there's a reason why both AMD and Intel, two powerhouses in CPU industry backs Havok which does perfectly fine for physics on the CPU. Most games out there are Havok-enabled compared to PhysX enabled, so I think NVIDIA is just trying to overhype things.

I'm going to get thumbed down to hell here but this just has to be said.

What year does Carmack think were in? 2007? Hardware physics never took off and never will and the one company that tried it got bought out and it's IP was integrated into mainstream video cards.

It's really sad to see a person like Carmack saying these things it makes him look foolish and out of touch which is far cry away from the ground breaking and innovative games this guy once made.

In the era where we have netbooks selling for $300 we still see Developers obsessing about the extra cost of a extra piece of hardware to the consumer(ie a $60 physic supported card).

You know honestly I like the idea that you can make some quick collision physics with Ageia. Granted the lack of x64 support is really a bad idea(I here this could be fixed??). However to think the GPU will replace the PPU in the immediate future means you ignore the fact that customers are always clammering for better looking graphics. The upside of putting physics on the GPU is you do get higher bandwidth since you wouldn't be using the PCI bus as often. However I would argue due to the cheap cost of physic support cards at the moment) while on the PPU/CPU side you can have more branching that the PPU is needed. Branching is really the key here since to do something random access allows for a hybrid of collision/custom physics emphasis for the specific genre.

Now on the flipside the ray tracing folks can argue you want raytracing on the cpu with the branching and physics on the GPU. Well to that I say good luck sending all that bandwidth across the PCI Express bus.

In sum would be nice if the implementation for all this game related SDK's would be cleaner but hey I'll take what I can get.

Ignore my first post needed some grammar corrections ;-)

In the era where we have netbooks selling for $300 we still see Developers obsessing about the extra cost of a extra piece of hardware to the consumer(ie a $60 physic supported card).

You know honestly I like the idea that you can make some quick collision physics with Ageia. Granted the lack of x64 support is really a bad idea(I here this could be fixed??). However to think the GPU will replace the PPU in the immediate future means you ignore the fact that customers are always clammering for better looking graphics. The upside of putting physics on the GPU is you do get higher bandwidth since you wouldn't be using the PCI bus as often. However I would argue due to the cheap cost of physic support cards at the moment) while on the PPU/CPU side you can have more branching that the PPU is needed.

Branching is really the key here since to do something, random access is needed. Random access allows for a hybrid of collision/custom physics emphasis for the specific genre which is a good thing since we are nowhere near real world physics(try to do a realworld blackhole simulation))..

Now on the flipside the ray tracing folks can argue you want raytracing on the cpu with the branching and simple physics on the GPU. Well to that I say good luck sending all that bandwidth across the PCI Express bus with the hd images which must be transmitted across the bus(1920x1080x32+ bitsx 30fps eats away very quickly the bus).

In sum would be nice if the implementation for all this game related SDK's would be cleaner but hey I'll take what I can get.

Carmack is ok but have you guys noticed the developers are talking up CPU's over GPUs. I think it's just easier for them to code with & saving money is the bottom line. I think it would be taking a step backwards to do cpu only some day.