Personal Servers. Authors and/or their companies shall have the right to post their IEEE-copyrighted material on their own servers without permission, provided that the server displays a prominent notice alerting readers to their obligations with respect to copyrighted material and that the posted work includes the IEEE copyright notice as shown in Section 8.1.10A above. An example of an acceptable notice is:
"This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder."
Sure, it is still a somewhat lame policy to transfer copyright to IEEE, but the dictator is not as malevolent as some here would make him out to be.
The reason for this story is that this article is supposed to be a review, whereas all the information about it before was marketing information from the manufacturer.
But of course, this article is clearly not a review, and is just more of the same.
Good point. Some companies like Adobe are finally looking at using GPUs for accelerating image processing as well. If the PPU API is general enough, perhaps it could lend itself to another problem domain.
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.
I've worked with the Lin-Cannny algorithm. It's great for tracking closest distances between non-colliding, convex polyhedra. It becomes a bit more problematic when the objects are actually interpenetrating, and of course the convexity restriction probably makes it incompatible with most games.
This is actually a debatable point. When I used to work on collision detection, we generally acknowledged that the actual detection of collisions was easier than the generation of accurate collision responses. There are lots of software speedup techniques (based on space partitions and hierarchies) to reduce the work in figuring out where the intersections occur. Computation of "penetration distance" as well as doing the physics with realistic friction forces makes the problem harder.
(see software packages like I-Collide, V-Collide, RAPID, etc. for computing collisions, distance compuatations, penetration distances, etc.)
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).
You're way off base. No one is being sued for damages, only for the information regarding the identity of the person who actually broke the contract. So if you repeat some rumor told to you by someone breaking a contract, you may be required to identify the person. Big deal. Get over it. (Again, there may be whistleblower exceptions, etc.)
Do people actually use the grammar checker in MS Office? I find that it usually suggests that I change something that is grammatically correct to something totally wrong.
I hope they will fix the problem with inserting java applets into the presentation software some day. If it actually worked, that would be a neat feature that PowerPoint does not have.
You are just wrong. Consumer cards have much better price/performance. In fact, the workstation cards by NVIDIA, for example, do not offer much beyond the consumer cards. They operate at lower clock rates for much more money. Other vendors' workstation cards may offer more VRAM and certification, but generally lack the cutting edge programmability features (which are definitely used for scientific visualization).
Slashdot Mirror
You never owned an Apple ][, then.
We didn't have buttons on those either. Just a manual slot cover.
Why would I want to print it? :-)
That is incredible. No blathering commentary - just the article.
My baby is not a big fan of your word.
One expects Slashdot readers would at least read the linked article before commenting on the topic...
New around here, aren't you? (and I thought you were such a frequent poster...)
...I started filing patents...And the U.S. Government peer-reviews...
Hah!!!
Sure, it is still a somewhat lame policy to transfer copyright to IEEE, but the dictator is not as malevolent as some here would make him out to be.
The reason for this story is that this article is supposed to be a review, whereas all the information about it before was marketing information from the manufacturer.
But of course, this article is clearly not a review, and is just more of the same.
Good point. Some companies like Adobe are finally looking at using GPUs for accelerating image processing as well. If the PPU API is general enough, perhaps it could lend itself to another problem domain.
This is truly a great achievement. I am happy to offload Clippy to a dedicated processing unit!
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.
What are the significant business applications of the GPU?
I've worked with the Lin-Cannny algorithm. It's great for tracking closest distances between non-colliding, convex polyhedra. It becomes a bit more problematic when the objects are actually interpenetrating, and of course the convexity restriction probably makes it incompatible with most games.
This is actually a debatable point. When I used to work on collision detection, we generally acknowledged that the actual detection of collisions was easier than the generation of accurate collision responses. There are lots of software speedup techniques (based on space partitions and hierarchies) to reduce the work in figuring out where the intersections occur. Computation of "penetration distance" as well as doing the physics with realistic friction forces makes the problem harder. (see software packages like I-Collide, V-Collide, RAPID, etc. for computing collisions, distance compuatations, penetration distances, etc.)
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).
Perhaps you would only need to read back the actual collisions and not the non-collisions.
I guess that agent can say goodbye to his commission.
The site says they are available in three sizes, all for free download.
You're way off base. No one is being sued for damages, only for the information regarding the identity of the person who actually broke the contract. So if you repeat some rumor told to you by someone breaking a contract, you may be required to identify the person. Big deal. Get over it. (Again, there may be whistleblower exceptions, etc.)
Do people actually use the grammar checker in MS Office? I find that it usually suggests that I change something that is grammatically correct to something totally wrong.
I hope they will fix the problem with inserting java applets into the presentation software some day. If it actually worked, that would be a neat feature that PowerPoint does not have.
2 2661
http://www.openoffice.org/issues/show_bug.cgi?id=
Why not get a credit card with your photo on it?
You are just wrong. Consumer cards have much better price/performance. In fact, the workstation cards by NVIDIA, for example, do not offer much beyond the consumer cards. They operate at lower clock rates for much more money. Other vendors' workstation cards may offer more VRAM and certification, but generally lack the cutting edge programmability features (which are definitely used for scientific visualization).
too bad about the accompanying drivel about each item.
They're not "ads" they're "infomercials"