Nvidia Fermi is recommended for Metro 2033 game

Page 10 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0

Havok.


Havok.

http://www.youtube.com/watch?v=6o6YVLmOs74#t=50s
Modified Havok.

Interesting uses of physics. Three games which don't have hardware acceleration of physics available (since they all run on consoles).
And what do we get on PC? "realistically waving flags and realistic volumetric fog".

And this is why we don't give a damn.


You sure post a lot for a guy that dosn't care
AMD soon putting out anything?
You know this is 2006:
http://www.youtube.com/watch?v=-x0S0b_eG_M

Still working on it I guess :hmm:

Like I said, NVIDIA, Intel and AMD (the 3 key players) are all going this way...the only ones mad about though seems to be the red boys...since they have no toys.

Like it or not GPU physics won't be going anywhere.
And PhysX can be made to run on CUDA, OpenCL or DirectCompute.

(will people please get a grasp between CUDA/OpenCL/DC and the PhysX/Havok/Bullet for fraks sake...)
 

badb0y

Diamond Member
Feb 22, 2010
4,015
30
91
Like I said, NVIDIA, Intel and AMD (the 3 key players) are all going this way...the only ones mad about though seems to be the red boys...since they have no toys.
Wait, who's the red boy and what company are they associated with?
 

zerocool84

Lifer
Nov 11, 2004
36,041
472
126
Like it or not GPU physics won't be going anywhere.

Exactly what most of us have been saying. Thanks for saying that yet again. It's still going no where and has nothing going for it right now and doesn't seem to have anything going for it in the near future.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Exactly what most of us have been saying. Thanks for saying that yet again. It's still going no where and has nothing going for it right now and doesn't seem to have anything going for it in the near future.

Tell that to NVIDIA.
Tell that to AMD.
Tell that to Intel.

Only one not seeming to get it is you

But I have no worries for you, in a few years time, when AMD actully does more than just talk and comes with GPU-physics...you will like it.

(post bookmarked for future reference aka When AMD wakes up)
 

zerocool84

Lifer
Nov 11, 2004
36,041
472
126
Tell that to NVIDIA.
Tell that to AMD.
Tell that to Intel.

Only one not seeming to get it is you

But I have no worries for you, in a few years time, when AMD actully does more than just talk and comes with GPU-physics...you will like it.

(post bookmarked for future reference aka When AMD wakes up)

I guess you haven't read the thread then cus it's more than me who aren't getting by what you're saying if you actually paid attention but we all know you don't.
 

PingviN

Golden Member
Nov 3, 2009
1,848
13
81
Tell that to NVIDIA.
Tell that to AMD.
Tell that to Intel.

Only one not seeming to get it is you

But I have no worries for you, in a few years time, when AMD actully does more than just talk and comes with GPU-physics...you will like it.

(post bookmarked for future reference aka When AMD wakes up)

I'd prefer AMD to keep on focusing on OpenCL instead of developing their own standard. It's better for me, as a consumer. Don't really get why you hold Nvidia in such high regard, but I'm sure the explanation isn't sensible enough to be put in words.
 

MrK6

Diamond Member
Aug 9, 2004
4,458
4
81
Tell that to NVIDIA.
Tell that to AMD.
Tell that to Intel.

Only one not seeming to get it is you

But I have no worries for you, in a few years time, when AMD actully does more than just talk and comes with GPU-physics...you will like it.

(post bookmarked for future reference aka When AMD wakes up)
When you stop getting off to your fanboy witch hunt and actually read the thread, you'll realize we all left on the logic train awhile ago.
 

Ancalagon44

Diamond Member
Feb 17, 2010
3,274
202
106
The problem is that OpenCL is just a heterogenous computing language - its not a physics library at all. PhysX is a physics library that can use either a PPU or an Nvidia GPU, or possibly an AMD gpu with a hack.

What AMD needs is a physics library that is free, well supported, full featured, and supports CPU or GPU acceleration using OpenCL, and is vendor agnostic. That would put the nail in the coffin of PhysX, which is already struggling for acceptance. Sadly, the possibility of that happening, even if AMD teams up with Intel (and we all know how much they love each other) is slim at best.
 

Lonyo

Lifer
Aug 10, 2002
21,939
6
81
The problem is that OpenCL is just a heterogenous computing language - its not a physics library at all. PhysX is a physics library that can use either a PPU or an Nvidia GPU, or possibly an AMD gpu with a hack.

What AMD needs is a physics library that is free, well supported, full featured, and supports CPU or GPU acceleration using OpenCL, and is vendor agnostic. That would put the nail in the coffin of PhysX, which is already struggling for acceptance. Sadly, the possibility of that happening, even if AMD teams up with Intel (and we all know how much they love each other) is slim at best.



For 2010 we have some exciting plans towards Bullet 3.x with support for OpenCL acceleration
http://bulletphysics.org/Bullet/phpBB3/viewtopic.php?f=18&t=4495
 

VirtualLarry

No Lifer
Aug 25, 2001
56,452
10,120
126
Also- the nV1 was a console chip that ended up getting released for the PC. It was the Sega Saturn's 3D chip.

No, it was not the same chip that ended up in the Saturn, although the two were similar enough that the Diamond Edge 3D cards shipped with a few games ported over from the Saturn that ran on the Edge 3D board (NV1).
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
The problem is that OpenCL is just a heterogenous computing language - its not a physics library at all. PhysX is a physics library that can use either a PPU or an Nvidia GPU, or possibly an AMD gpu with a hack.

What AMD needs is a physics library that is free, well supported, full featured, and supports CPU or GPU acceleration using OpenCL, and is vendor agnostic. That would put the nail in the coffin of PhysX, which is already struggling for acceptance. Sadly, the possibility of that happening, even if AMD teams up with Intel (and we all know how much they love each other) is slim at best.


Nice to see 1 person has figured it out.
What really annoys me are people bashing stuff...beyond their techincal knowlegde.

Eg. "OpenCL will destroy PhysX!!!"
A statement like that only goes to show ignorance.

Physcis libraies:
PhysX
Havok
Bullet

Platforms:
CUDA
OpenCL (very limited on DX10/DX10.1 cards)
DirectCompute (very limited on DX10/DX10.1 cards)

Before people realize this there isn't much hope.
 

Kuzi

Senior member
Sep 16, 2007
572
0
0
What AMD needs is a physics library that is free, well supported, full featured, and supports CPU or GPU acceleration using OpenCL, and is vendor agnostic. That would put the nail in the coffin of PhysX, which is already struggling for acceptance. Sadly, the possibility of that happening, even if AMD teams up with Intel (and we all know how much they love each other) is slim at best.

There you go

AMD Open Physics Initiative Expands Ecosystem with Free DMM for Game Production and Updated version of Bullet Physics:

"March 8, 2010 — AMD (NYSE: AMD) today announced that, along with partners Pixelux Entertainment and Bullet Physics, it has added significant support to the Open Physics ecosystem by providing game developers with access to the newest version of the Pixelux Digital Molecular Matter (DMM), a breakthrough in physics simulation. In addition, to enabling a superior development experience and helping to reduce time to market, Pixelux has tightly integrated its technology, DMM, with Bullet Physics, allowing developers to integrate physics simulation into game titles that run on both OpenCL- and DirectCompute-capable platforms."

http://www.amdzone.com/index.php/co...duction-and-updated-version-of-bullet-physics

"All of the Bullet Physics implementations described above can be run on any OpenCL- or DirectCompute-capable platform."
 

NoQuarter

Golden Member
Jan 1, 2001
1,006
0
76
Nice to see 1 person has figured it out.
What really annoys me are people bashing stuff...beyond their techincal knowlegde.

Eg. "OpenCL will destroy PhysX!!!"
A statement like that only goes to show ignorance.

And yet I believe you're the only one who's made it.. I don't think anyone has confused OpenCL/DirectCompute/CUDA/Stream with Physx/Havok/Bullet...
 

evolucion8

Platinum Member
Jun 17, 2005
2,867
3
81
Nice to see 1 person has figured it out.
What really annoys me are people bashing stuff...beyond their techincal knowlegde.

Eg. "OpenCL will destroy PhysX!!!"
A statement like that only goes to show ignorance.

Physcis libraies:
PhysX
Havok
Bullet

Platforms:
CUDA
OpenCL (very limited on DX10/DX10.1 cards)
DirectCompute (very limited on DX10/DX10.1 cards)

Before people realize this there isn't much hope.

You forgot Stream which unleash more potential to the HD 4x00 series than the current implentation of their OpenCL/DirectCompute
 

Lonyo

Lifer
Aug 10, 2002
21,939
6
81
And yet I believe you're the only one who's made it.. I don't think anyone has confused OpenCL/DirectCompute/CUDA/Stream with Physx/Havok/Bullet...

They have confused them.
From early on in this thread multiple people have talked about physics in terms of APIs instead of middleware, and multiple people have refuted it.
ViRGE was the first to point out that physics is done by a middleware product, and that was before Lonbjerg had even posted in the thread.
I've also pointed it out in the thread. As has Lonbjerg (obviously).

Some people have also been confused by hardware vs software PhysX.
 

NoQuarter

Golden Member
Jan 1, 2001
1,006
0
76
They have confused them.
From early on in this thread multiple people have talked about physics in terms of APIs instead of middleware, and multiple people have refuted it.
ViRGE was the first to point out that physics is done by a middleware product, and that was before Lonbjerg had even posted in the thread.
I've also pointed it out in the thread. As has Lonbjerg (obviously).

Some people have also been confused by hardware vs software PhysX.

Meh, perhaps, but it's easy enough to fill in the blanks to know what people are getting at.


The main problem is simply that nvidia owns PhysX, and thus it makes no business sense for ATI to license it because nVidia would then control ATI's future through licensing fees and threats once hardware PhysX became established. If you think Rambus is evil this is how nVidia would sink ATI..


So the two are at an impasse, and the only thing that will break the deadlock is a physics middleware that is open source (Bullet) or at least a 3rd party (Intel - Havok, not the perfect business solution but w/e), and preferably with extensions for an API developed by a consortium (Khronos - OpenCL) or again, at least a neutral 3rd party (Microsoft - DirectCompute).

So Nvidia is paradoxically holding back the industry while advancing it - by forging off on their own they show us a little of what can be done but slyly make sure no one else can follow, so no one gets on board.

Honestly though.. DirectCompute is going to have a lot of weight behind it once MS pushes it through their next Xbox console, so even if CUDA is better represented in hardware now I don't think it will matter much. Maybe we'll see a physics engine take advantage of that in a couple years as the video cards will all support it well by then too.
 

BenSkywalker

Diamond Member
Oct 9, 1999
9,140
67
91
Ok, clearly you aren't talking about Activision Blizzard, which is what I, and presumably most other people, would think of when hearing the words 'world's largest game publisher'.

Nintendo's games sales dwarfs Activision's.

Some people don't like nVidia's handling of PhysX not because they're ATI fanboys or anti-PhysX but because they aren't blinded by bias and can see how PhysX, in its current form, can fracture the market.

Of course it will fracture the market until their is something to take its place. What I can't comprehend is why people will lame the blame at the feet of someone who is trying to do something instead of those that refuse to. It make no sense to me. If PhysX didn't exist, ATi users wouldn't see anything different in their games. As of right now, the difference is that nV users get to turn on some extra graphics effects as a bonus- and this leads people to lament nVidia. Why? If it truly does upset people, why not push the other players to do something about it and come up with an alternative? Instead, what we see is people lamenting the one company that has actively been pushing advancement because they are pushing advancement.

Now lets modify your statement. PhysX supports more ATI based hardware used for gaming, by a huge margin, than Nvidia hardware (124M vs. 32M). That's kinda sad isn't it? You can add the PC numbers if you like, that won't help much.

PhysX capable cards on the PC cleared 100Million at the beginning of last year. Enthusiast boards may confuse things a bit, but nV outsells ATi by a lot.

Consoles run 'software'/CPU PhysX, as can all PCs, only NV cards can run hardware PhysX... why would you say PhysX runs on all consoles and 66% of the PC market when it either runs on all consoles and all PCs, or it just runs on 66% of the PC market. Youre being rather fuzzy here.

My point in only using 66% was that those are the PCs that can benefit from GPU accelerated physics, so a developer can use a baseline physics engine that will still run on the consoles and used more advanced features on the PC. Pointing out software physics on the PC side does nothing as Havok will do the same there, or any other physics middleware. Pointing out the PhysX will run on the other platforms was strictly about it being portable which like it or not is a very real requirement for most game publishers today.

UnrealEngine runs on PS3, Xbox 360, PC etc, and it runs on ATI, NV, whatever. And that's how it should be. You make middleware, and it runs on systems using the standard API(s) for that system.

Back in the proprietary API days I was a fan of Glide because it was the only truly viable API to get the task done. Right now it is the same with PhysX. Certainly you can get things to work using something else, but certainly not with the same results in terms of performance. There are several other physics middleware packages that have been talking about supporting GPU acceleration and being more open, when they hit I will rather quickly switch support. Until then, something>>>>>>>nothing.

Honestly though.. DirectCompute is going to have a lot of weight behind it once MS pushes it through their next Xbox console, so even if CUDA is better represented in hardware now I don't think it will matter much. Maybe we'll see a physics engine take advantage of that in a couple years as the video cards will all support it well by then too.

You think CUDA will be excluded from the PS4? I don't know one way or the other, mainly I'm just pointing that out as it comes back around to MS is a company with a vested interest in not being ideal for portability. The same can be said for nVidia, it's just they actually support a larger percentage of gaming platforms with their software then MS currently does.
 

NoQuarter

Golden Member
Jan 1, 2001
1,006
0
76
You think CUDA will be excluded from the PS4? I don't know one way or the other, mainly I'm just pointing that out as it comes back around to MS is a company with a vested interest in not being ideal for portability. The same can be said for nVidia, it's just they actually support a larger percentage of gaming platforms with their software then MS currently does.

Everything I've heard says Sony won't be using Nvidia in the PS4 so I have no idea. But if they design the PS4 like they did the PS3 everything on it will be rather esoteric, where Xbox has always had a development platform that was easy for devs to transition back and forth from the PC with.

Current games are largely DX9 because of console portability (among other reasons) and I would expect hardware physics to follow this trend. The next Xbox won't have CUDA but will have OpenCL and DirectCompute. If some physics middleware comes along and runs on top of either of these API's that is supported on major consoles then those games and their ports will easily do GPGPU physics.

It's for that reason that I don't think including Nvidia's broad software based PhysX support as any way relevant because it doesn't influence the inclusion of hardware based PhysX. But it does depend on what the PS4 ends up as, and if it shares OpenCL or DirectCompute with Xbox or only includes Cuda. It may end up being OpenCL as the common API.
 
Last edited by a moderator:
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |