onetimeposter
Banned
GREEN: My Questions
BOLD: The Answers
Print
Expand all
Clarification of Comments made by your executives concerning Xbox 360 limitations Inbox
Andy Keane
to me, Kim
Hide options 2:55 pm (3½ hours ago)
From: Andy Keane <akeane@ageia.com>
To: hasanahmad@gmail.com
Cc: Kim Stowe <kim@kimstowe.com>
Date: Sep 3, 2005 2:55 PM
Subject: RE: Clarification of Comments made by your executives concerning Xbox 360 limitations
Reply | Reply to all | Forward | Print | Add sender to Contacts list | Report phishing | Show original
Here is the section of the ExtremeTech article in question. The response was to answer this section and your email requesting information on limitations in the Xbox360 that prevent it’s being able to run specific algorithms.:
From ExtremeTech:
“Although the PhysX libraries accelerate a host of technologies, from physical object interactions to fluid-based particle effects like water and smoke, only the PlayStation 3 and a PhysX PC will have the horsepower to process all of the technology's features, Ageia executives said. The Xbox 360 will not be able to process the fluid-based technology, because of the limitations of its architecture.â€
This statement is not accurate. We have no data that the Xbox360 cannot “process fluid-based technologyâ€. Additionally, we have no data on the relative performance of the two platforms that reveal limitations in either console platform. The specific question is about the limitations and the specific answer is that we don’t have an indication that there are limitations regarding physics algorithms in our SDK. If this was said during the presentation by an AGEIA person, then this was in error.
I have answered other points below.
From: Hasan Ahmad [mailto:hasanahmad@gmail.com]
Sent: Friday, September 02, 2005 5:49 PM
To: Andy Keane
Subject: Re: Clarification of Comments made by your executives concerning Xbox 360 limitations
Thanks for the detailed response. Although I find it rather interesting the response that is given.
I find it highly unlikely that your technical staff did not benchmark your software on all these platforms including the available Playstation 3 and Xbox 360 development kits (and the PC architectures). Am i right in thinking that you are not willing to give out explicit performance comparisons on the kits just to sound more diplomatic?
The two console platforms are both new and the software on the platforms is new. We have not benchmarked the consoles for two reasons. First, there are no physics benchmarks that represent the performance you’ll see in a game. We do have engineering code that tests operation of the software, but these are not benchmarks. Second, benchmarking physics is very difficult. It is not meaningful to measure a number that dedicates the entire resources of a platform to one operation. This is a peak case that never exists in a game. In addition, the performance of physics in a game is dependent on the game code. Physics is a simulation where the game provides data to the physics software, a simulation is run and the results are handed back to the game for use by the game code.
In the long run, we will not be publishing benchmarks on the relative performance of these platforms. Our engineering dollars are spent putting physics features into our SDK and our PhysX chip that increase the range of effects in games. Our investment of time and energy is to make our physics software useful on all platforms. Publishing performance results of my own software across platforms does not add to AGEIA’s business.
also this qoute:
The implied difference in performance was from assumptions about the number of compute elements and memory architectures, and how well these might fit to various simulation algorithms as enabled by typical game developers.
are you clearly stating that whatever the limitations the 360 will have are due to its lack of resources compared to the playstation 3?
Any comparison was done from specifications. Like I said in the previous email, we don’t have data that says one platform is better than another. There have been numerous paper analyses comparing the capabilities of the two consoles and PC. The performance of a game on a console will be determined by the developer and how that developer allocates the resources in the console. I think it still remains to be seen how well the game developers utilize each console.
Also in the presentation you guys made it clearly stated "relative performance" which clearly seems to be indicating you are distances yourselves not from the extremeTech article but the presentation itself.
Let’s break this in 2. I answered the first part regarding ExtremeTech’s statements above. In the previous email, I talked about what point the section of the presentation was intended to convey. The presentation was aimed at developers. The message was that we want physics in game play but that you might have to scale the physics depending on the resources of the platform or what resources they choose to dedicate to physics. This entire discussion is about one line, on one slide that did not put Xbox360 and PS3 in the same sentence. I will repeat again that this opinion was an inference from compute and memory resources detailed in specifications, not hard data. This was an incorrect inference because we don’t have any data nor do we have data from game developers.
All these questions are important to TeamXbox so we can get the facts right that this seems to be facts that you are saying Extremetech reported it wrong or your slides in the presentation were wrong because the slides explicitly (if the translation is to be believed) state whatever Extremetech was assuming about the Xbox 360 not being proeffient enough to compete with a Playstation 3 or a PC sloted with a Physx card. This is in regards to physics performance.
1. ExtremeTech reported what they heard and saw. If an AGEIA person stated that the Xbox360 has technical limitations that prevent the Xbox360 from running algorithms in our SDK, then that statement was not correct. If we weren’t clear and this was implied by the presentation, then likewise we should have been more clear. I will not fault a reporter for writing what they heard. It’s up to us to be clear in our message.
2. We did infer in a slide that the PS3 has more resources that could be dedicated to physics. This was an error since this statement needs real data as backup. Any conclusion on relative performance was built from specifications released publicly by the console vendors. As I stated earlier, the performance of physics in a game is dependent on so many factors and resources.
In simple words, will the Xbox 360 (according to Xbox 360 and PS3 available documentation) perform on par with the Playstation 3 when it comes to real in-game performance?
To this specific question I can only say that we don’t know. We have a physics SDK that is ported to the platforms. Performance on this one piece of code will not give you any indication of the performance in a real game. We are one piece of a complicated software system in the game. Our software is also very interrelated to the core of the game code where our results cause internal and external paths to change within the game. In addition, there are no physics benchmarks that give you an indication of the performance in a real game. As you have probably seen, even benchmarks in mature technologies like graphics and CPUs cause raging debates about how to apply the results to real world applications.
I do know that eventually PC based Physx cards will give higher performance than even the PS3 and Xbox 360 but im explicitly requesting an honest answer between some real data from numbers available between the 2 consoles.
Thanks
Hasan
Regards, Andy
ReplyReply to allForwardInvite Andy to Gmail
akeane@ageia.com
BOLD: The Answers
Expand all
Clarification of Comments made by your executives concerning Xbox 360 limitations Inbox
Andy Keane
to me, Kim
Hide options 2:55 pm (3½ hours ago)
From: Andy Keane <akeane@ageia.com>
To: hasanahmad@gmail.com
Cc: Kim Stowe <kim@kimstowe.com>
Date: Sep 3, 2005 2:55 PM
Subject: RE: Clarification of Comments made by your executives concerning Xbox 360 limitations
Reply | Reply to all | Forward | Print | Add sender to Contacts list | Report phishing | Show original
Here is the section of the ExtremeTech article in question. The response was to answer this section and your email requesting information on limitations in the Xbox360 that prevent it’s being able to run specific algorithms.:
From ExtremeTech:
“Although the PhysX libraries accelerate a host of technologies, from physical object interactions to fluid-based particle effects like water and smoke, only the PlayStation 3 and a PhysX PC will have the horsepower to process all of the technology's features, Ageia executives said. The Xbox 360 will not be able to process the fluid-based technology, because of the limitations of its architecture.â€
This statement is not accurate. We have no data that the Xbox360 cannot “process fluid-based technologyâ€. Additionally, we have no data on the relative performance of the two platforms that reveal limitations in either console platform. The specific question is about the limitations and the specific answer is that we don’t have an indication that there are limitations regarding physics algorithms in our SDK. If this was said during the presentation by an AGEIA person, then this was in error.
I have answered other points below.
From: Hasan Ahmad [mailto:hasanahmad@gmail.com]
Sent: Friday, September 02, 2005 5:49 PM
To: Andy Keane
Subject: Re: Clarification of Comments made by your executives concerning Xbox 360 limitations
Thanks for the detailed response. Although I find it rather interesting the response that is given.
I find it highly unlikely that your technical staff did not benchmark your software on all these platforms including the available Playstation 3 and Xbox 360 development kits (and the PC architectures). Am i right in thinking that you are not willing to give out explicit performance comparisons on the kits just to sound more diplomatic?
The two console platforms are both new and the software on the platforms is new. We have not benchmarked the consoles for two reasons. First, there are no physics benchmarks that represent the performance you’ll see in a game. We do have engineering code that tests operation of the software, but these are not benchmarks. Second, benchmarking physics is very difficult. It is not meaningful to measure a number that dedicates the entire resources of a platform to one operation. This is a peak case that never exists in a game. In addition, the performance of physics in a game is dependent on the game code. Physics is a simulation where the game provides data to the physics software, a simulation is run and the results are handed back to the game for use by the game code.
In the long run, we will not be publishing benchmarks on the relative performance of these platforms. Our engineering dollars are spent putting physics features into our SDK and our PhysX chip that increase the range of effects in games. Our investment of time and energy is to make our physics software useful on all platforms. Publishing performance results of my own software across platforms does not add to AGEIA’s business.
also this qoute:
The implied difference in performance was from assumptions about the number of compute elements and memory architectures, and how well these might fit to various simulation algorithms as enabled by typical game developers.
are you clearly stating that whatever the limitations the 360 will have are due to its lack of resources compared to the playstation 3?
Any comparison was done from specifications. Like I said in the previous email, we don’t have data that says one platform is better than another. There have been numerous paper analyses comparing the capabilities of the two consoles and PC. The performance of a game on a console will be determined by the developer and how that developer allocates the resources in the console. I think it still remains to be seen how well the game developers utilize each console.
Also in the presentation you guys made it clearly stated "relative performance" which clearly seems to be indicating you are distances yourselves not from the extremeTech article but the presentation itself.
Let’s break this in 2. I answered the first part regarding ExtremeTech’s statements above. In the previous email, I talked about what point the section of the presentation was intended to convey. The presentation was aimed at developers. The message was that we want physics in game play but that you might have to scale the physics depending on the resources of the platform or what resources they choose to dedicate to physics. This entire discussion is about one line, on one slide that did not put Xbox360 and PS3 in the same sentence. I will repeat again that this opinion was an inference from compute and memory resources detailed in specifications, not hard data. This was an incorrect inference because we don’t have any data nor do we have data from game developers.
All these questions are important to TeamXbox so we can get the facts right that this seems to be facts that you are saying Extremetech reported it wrong or your slides in the presentation were wrong because the slides explicitly (if the translation is to be believed) state whatever Extremetech was assuming about the Xbox 360 not being proeffient enough to compete with a Playstation 3 or a PC sloted with a Physx card. This is in regards to physics performance.
1. ExtremeTech reported what they heard and saw. If an AGEIA person stated that the Xbox360 has technical limitations that prevent the Xbox360 from running algorithms in our SDK, then that statement was not correct. If we weren’t clear and this was implied by the presentation, then likewise we should have been more clear. I will not fault a reporter for writing what they heard. It’s up to us to be clear in our message.
2. We did infer in a slide that the PS3 has more resources that could be dedicated to physics. This was an error since this statement needs real data as backup. Any conclusion on relative performance was built from specifications released publicly by the console vendors. As I stated earlier, the performance of physics in a game is dependent on so many factors and resources.
In simple words, will the Xbox 360 (according to Xbox 360 and PS3 available documentation) perform on par with the Playstation 3 when it comes to real in-game performance?
To this specific question I can only say that we don’t know. We have a physics SDK that is ported to the platforms. Performance on this one piece of code will not give you any indication of the performance in a real game. We are one piece of a complicated software system in the game. Our software is also very interrelated to the core of the game code where our results cause internal and external paths to change within the game. In addition, there are no physics benchmarks that give you an indication of the performance in a real game. As you have probably seen, even benchmarks in mature technologies like graphics and CPUs cause raging debates about how to apply the results to real world applications.
I do know that eventually PC based Physx cards will give higher performance than even the PS3 and Xbox 360 but im explicitly requesting an honest answer between some real data from numbers available between the 2 consoles.
Thanks
Hasan
Regards, Andy
ReplyReply to allForwardInvite Andy to Gmail
akeane@ageia.com
