By using this site, you agree to our Privacy Policy and our Terms of Use. Close
MikeB said:

@ Squilliam

It obviously wasn't...


Why are you talking in past tense? The Unreal engine can be heavily adapted to tap into more processors. But the best used approaches in doing so are more beneficial to the PS3 architecture, yet beneficial to both platforms (more so to PCs sporting more cores).

I was talking in past tense because that claim have been proved to be false unless you make the proviso that it was all they could achieve with the budget/programing techniques of the time.

If the Spectrum of CPU design difference runs from PS3 --------> Xbox 360 -------> X86 PC its very unlikely that optimizations for the PS3 would even be applicable to the PC. They are completely different architectures.

For the Unreal engine on the PS3 to significantly outperform its 360 version only more stuff needs to be ported over to the Cell's SPEs (tapping into more processors than the 360 has cores available).

Only more stuff.... Well only more stuff needs to be ported from Core 1 to Core 2/3 on the Xenon.

"Microsoft Directions in Parallel Computing and Some Short Term Help:
This talk focuses on the native task scheduler being announced by the Parallel Computing Platform group in Microsoft this spring and offerings that are available in the XDK. The scheduling of tasks within games can improve resource utilization, load balancing, and performance. For games targeting the current generation of PCs and the Xbox 360 console, we discuss an interim solution. Previous talks given on this topic laid the foundation for using tasks to move work required by the engine from an over-utilized hardware core to an underutilized core. A progression of task and scheduler designs is presented that start with simple implementations and move to more complex designs that have low-overhead. The simple implementations are satisfactory for a small number of tasks but impose a prohibitive amount of overhead when the number of tasks grows. Finally, we present the work-stealing algorithm that pulls work from one core to another in the low-overhead scheduler."

"Xbox 360 Compiler and PgoLite Update:
The Xbox 360 compiler has changed dramatically in the last year, which changes the rules for how to write efficient Xbox 360 code. Many of the improvements automatically make your code faster, but others require you to change your code in order to reap the benefits. PgoLite has also improved and should be used differently, to get even better results. This talk summarizes the past year's developments, and gives simple rules for how to get maximum benefit from the changes.

I wonder if MS developpement could have met Intel's one, and if Ms is about to propose something close to Intel threading building blocks?
Which sounds like a effective coder friendly approach to multi threading.
(for reference:http://softwarecommunity.intel.com/i..._for_promo.pdf )"



You are speculating. Furthermore multicore development is still in its infancy so its a big statement to call the Xenon 100% tapped out.


Not really the case, look at the game engine design of 360 launch titles and CPU cycle usage. There are some important bottlenecks relating to shared system RAM bandwidth and shared L2 cache between all three cores (relevant to talk about if looking at theoretical peak performance).

Considering that many of these so called 100% usage scenario type games were ported to the PS3 and were blatently neglectful of much if any SPU usage I would have to call BS on that. If they were using 100% of 3 cores, the PS3 versions should have been absurdly terrible.

Considering Halo 3 went for a funky light engine that required the game to be rendered at 60FPS the game outside of the lighting was pretty good too.


First of all the lighting engine isn't really that impressive, secondly the game is rendering 30 FPS and slows down towards lower FPS in parts of the game.

Rendered 60 FPS then 2 frames were combined into 1 to do the HDR lighting. So runs at 30, renders at 60. Pretty much the same deal as COD IV considering the rendering rate and the resolution were the same.

Furthermore when you say easily, I hope you're talking about games with budgets under 20 million right?


Technically. The game budgets aren't entirely invested into technology advances, developing supreme game assets, hiring good actors, hiring game designers, hiring audio professionals, etc cost the bulk of the money.

Actually having only a small team of very talented coders usually yield far greater technical advancements than a huge team of mediocre talented programmers.

Once you gain advancements developers can share knowledge and even code with other developers, it will only improve for the PS3 over time (it will also become cheaper).

What you're doing is comparing games which Sony feels they have to invest heavily into to "prove" that the PS3 has huge gobs of power when the Xbox 360 running a generic engine with only some specific tweaks can pull uncomfortably close to a game which cost 2-3x to make and was meant to be the technical highlight of the PS3.

 

 



Tease.