@ Squilliam

Or the game engine they have to work with and adapt (for example other people's code) or usage of middleware technology which still hasn't been adapted enough (the developers being far more dependent on 3rd party coders).



Lots of info can be found on CellPerformance.com, it's not as easy as a simple recompile for the SPEs.



Of course they intend to use as much as possible over time, but redesign takes time and effort. They are still using only 4 SPEs but they moved a lot of workload from the PPE onto these SPEs, so they are using more processor time on these SPEs than they did before. It could well be Killzone 2 when it is released uses all 6 SPEs, but the SPEs will have a lot of processor time to spare. Rome wasn't build in a day.



Adding additional effects or onscreen activity for Killzone 3, taking more GPU workload off the RSX to improve framerates, take more workload from the PPE onto the SPEs for performance gains, etc.

@ mike as much as I would like to give real world examples to counter your points. I think it can wait until later.

MikeB said: but the SPEs will have a lot of processor time to spare.

Welcome to the reality of multi-core. Unlike with single-core, it's very hard/impossible to attain peak power usage when you're doing multi-core (for any complex application with lots of different serial workloads to process).

Can they use more later? Sure... Will there still be plenty of untapped CPU time? Most likely, yes.

 

NJ5 said: MikeB said: but the SPEs will have a lot of processor time to spare. Welcome to the reality of multi-core. Unlike with single-core, it's very hard/impossible to attain peak power usage when you're doing multi-core (for any complex application with lots of different serial workloads to process). Can they use more later? Sure... Will there still be plenty of untapped CPU time? Most likely, yes.

Near peak performance is far harder/impossible to reach with other kinds of multi-core CPUs due to various design bottlenecks, the Cell has been designed specifcally to tackle such problems, the SPEs can produce results far more independtly and thus are able to yield superior efficiency like demonstrated by scientists in various test settings. The key is to write the game engine as asynchronous as possible to harvest the most potential.

Any gamer here knows that you CANNOT do comparisons by screenshots alone. It ultimately falls back on the end results.

"MGS4 1024x768=786432 Pixels resolution Gears 1280x720=921600 Pixels resolution Gears of War runs in a higher resolution then MGS4 does." It's posts like these that really show fanboyism. You just take a piece of the pie while ignoring the big picture, the overall experience.

It's sad how some of you guys are ignoring the fact that there have been multi-platform games that look better on the PS3.

Jordahn said: FJ-Warez said: TheRealMafoo said:   One thing you don't see in that screen shot, is the motion, the sound, the fact that if you turn left, you will see a building 300 yards off collapse into smoke. You don't see the 30 other extremely high detailed gunmen shooting at you, all at a higher resolution then anything Gears could think about. Play the game, and then you will understand. I have both, and have finished both. Gears is not even in the same league graphically.   Both games feature a different kind of war, and as I said MGS4 looks better but not a lot better taken in account that GoW is just a 10 millions game vs a 70-80 millions game, I can´t see huge gap, and about the number of things happening, I will wait to GoW 2 to make a better judgement...   It's posts like these that really show fanboyism. You just take a piece of the pie while ignoring the big picture, the overall experience.

Except that this thread is all about what looks better, not whats the most fun or even technical stuff like which has the most impressive engine because those that are up to such a discussion would be drowned out by those who arent.

He has a point, you don't.

 

MikeB said: NJ5 said: MikeB said: but the SPEs will have a lot of processor time to spare. Welcome to the reality of multi-core. Unlike with single-core, it's very hard/impossible to attain peak power usage when you're doing multi-core (for any complex application with lots of different serial workloads to process). Can they use more later? Sure... Will there still be plenty of untapped CPU time? Most likely, yes. Near peak performance is far harder/impossible to reach with other kinds of multi-core CPUs due to various design bottlenecks, the Cell has been designed specifcally to tackle such problems, the SPEs can produce results far more independtly and thus are able to yield superior efficiency like demonstrated by scientists in various test settings. The key is to write the game engine as asynchronous as possible to harvest the most potential.

Which is (essentially) the core of the problem, and the reason why we will never obtain good performance out of the Cell processor ...

When you are programming something that can be broken up into tons of tiny (and independant) processes like a web application developing an asynchronous application is easy because the completion of one process is not dependant on the completion of another process. A videogame is an implicitly syncronized problem being that there is a clear process flow which has to be completed in order, and if you split up a process that makes up the larger process flow each of these sub-processes has to be broken up (roughly) equally in order to prevent waste.

The Cell Processor is the round peg to the Game Engine's square hole.

 

@Squilliam MikeB is talking about large dense matrix multiplication with single precision floating point elements.

@FJ-Warez AA isn't magic. All the techniques boil down to blur filters.