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

Forums - Gaming - If PS3 is hard to port to, then just make on the PS3 first....read

rubido said:
MikeB said:
Yes, to quote a multi-platform developer:

"Writing code optimized for the PS3 and using threading policies that are suited the SPUs is a given, because not doing so would not be acceptable at all. All our multithreading is done on PS3 first without exception, and other platforms emulate SPURS."

"Secondly, the matters of multithreading policies, the whole job queue architecture, encapsulation of jobs and their corresponding data packets, etc. that work on the PS3 are indeed more than applicable of the 360/PC. And as I've mentioned before, they work better than anything and everything that Microsoft recommends (so far without exception for us). The problems lie in the fact that that work is an absolute necessity on the PS3, whereas they're not entirely necessary on any other platform."

That's pretty much at the core of the issue. Writing PS3 friendly code, will also result in better performing code on other CPUs. Of course the Cell is by far the most powerful console CPU, so the gains here can be much bigger potentially, but that won't be the case for most multi-platform games as for example you can't have one version of the game have twice as many onscreen enemies than another version without considerable redesign.

 

Who the hell is this developer? I searched for those quotes in google and couldn't find anything.

And also, your conclusion from those quotes is wrong (the bolded part). It is true that it is easier to port a code from an architecture with more cores to one with less cores. It is also true that this forces you to work more on the code and might have something more organized. But in no way does this imply that the code will perform better. Don't confuse things as this is incorrect.

I found something on google (a MikeB post on neogaf):

http://www.neogaf.com/forum/showpost.php?p=9623440&postcount=606

Apparently it came from some PMs MikeB got over there.

 



My Mario Kart Wii friend code: 2707-1866-0957

Around the Network

In both my Computer Architecture and VLSI classes back in college, the professors always made it very clear that when designing some type of mass market programmable device, your architecture had to be easy (well, relatively easy, not like BASIC easy) to design for by the average engineer. It didn't matter how awesome your specs were; if your system were complicated, most of the system would go to waste. If the engineers designing it couldn't figure it out and design/program well in it, it wasn't their fault that they can't "get" your design; it's yours for making it complicated enough that your system is underutilized. There are more relevant engineering principles I could discuss, but I really don't want to turn this into a long, full-blown flame war.

This is not some fanboy ranting about the way things oughta be. This is a fundamental engineering principle that Sony chose to ignore. It isn't the programmers' faults that they're underutilizing the PS3. It's Sony's fault for forcing them to underutilize it. "Laziness" has absolutely nothing to do with it, and the fact that you're claiming that laziness is the problem tells me that you have absolutely no idea how product development works. Please stick to things you do know about.



TheNino85 said:
In both my Computer Architecture and VLSI classes back in college, the professors always made it very clear that when designing some type of mass market programmable device, your architecture had to be easy (well, relatively easy, not like BASIC easy) to design for by the average engineer. It didn't matter how awesome your specs were; if your system were complicated, most of the system would go to waste. If the engineers designing it couldn't figure it out and design/program well in it, it wasn't their fault that they can't "get" your design; it's yours for making it complicated enough that your system is underutilized. There are more relevant engineering principles I could discuss, but I really don't want to turn this into a long, full-blown flame war.

This is not some fanboy ranting about the way things oughta be. This is a fundamental engineering principle that Sony chose to ignore. It isn't the programmers' faults that they're underutilizing the PS3. It's Sony's fault for forcing them to underutilize it. "Laziness" has absolutely nothing to do with it, and the fact that you're claiming that laziness is the problem tells me that you have absolutely no idea how product development works. Please stick to things you do know about.

Speaking from experience, then?



TheNino85 said:
In both my Computer Architecture and VLSI classes back in college, the professors always made it very clear that when designing some type of mass market programmable device, your architecture had to be easy (well, relatively easy, not like BASIC easy) to design for by the average engineer. It didn't matter how awesome your specs were; if your system were complicated, most of the system would go to waste. If the engineers designing it couldn't figure it out and design/program well in it, it wasn't their fault that they can't "get" your design; it's yours for making it complicated enough that your system is underutilized. There are more relevant engineering principles I could discuss, but I really don't want to turn this into a long, full-blown flame war.

This is not some fanboy ranting about the way things oughta be. This is a fundamental engineering principle that Sony chose to ignore. It isn't the programmers' faults that they're underutilizing the PS3. It's Sony's fault for forcing them to underutilize it. "Laziness" has absolutely nothing to do with it, and the fact that you're claiming that laziness is the problem tells me that you have absolutely no idea how product development works. Please stick to things you do know about.

good post , the fault issue is debtable however "for forcing them to underutilize it."  well no Sony isn't forcing them to underutilize it , putting them  in a difficult position in terms of cost , time , know how etc is the case. the developer has the opportunity to under utilize which brings consequences unrelated to Sony,  bad reviews , p.r , sales etc.

"laziness is the problem tells me that you have absolutely no idea how product development works. Please stick to things you do know about."

if the developer has the opportunity to make a good game but chooses not to  because of the complexity of the architecture , that's a choice they make , one they should be largely be accountable for( i'm not saying Sony don't share the blame but let's be  reasonable). There are developers that have taken advantage of the ps3 and produced excelent title so it' onlly  fair that developers who can't even get  their game to run at a sready frame rate should be considered lazy.




rubido said:
MikeB said:


That's pretty much at the core of the issue. Writing PS3 friendly code, will also result in better performing code on other CPUs. Of course the Cell is by far the most powerful console CPU, so the gains here can be much bigger potentially, but that won't be the case for most multi-platform games as for example you can't have one version of the game have twice as many onscreen enemies than another version without considerable redesign.

And also, your conclusion from those quotes is wrong (the bolded part).

I think you misread what I stated, it means if not coded this way you will get bad results on the PS3, so the gains are much larger. On the 360 the gains are more marginal.



Naughty Dog: "At Naughty Dog, we're pretty sure we should be able to see leaps between games on the PS3 that are even bigger than they were on the PS2."

PS3 vs 360 sales

Around the Network
Million said:
TheNino85 said:
In both my Computer Architecture and VLSI classes back in college, the professors always made it very clear that when designing some type of mass market programmable device, your architecture had to be easy (well, relatively easy, not like BASIC easy) to design for by the average engineer. It didn't matter how awesome your specs were; if your system were complicated, most of the system would go to waste. If the engineers designing it couldn't figure it out and design/program well in it, it wasn't their fault that they can't "get" your design; it's yours for making it complicated enough that your system is underutilized. There are more relevant engineering principles I could discuss, but I really don't want to turn this into a long, full-blown flame war.

This is not some fanboy ranting about the way things oughta be. This is a fundamental engineering principle that Sony chose to ignore. It isn't the programmers' faults that they're underutilizing the PS3. It's Sony's fault for forcing them to underutilize it. "Laziness" has absolutely nothing to do with it, and the fact that you're claiming that laziness is the problem tells me that you have absolutely no idea how product development works. Please stick to things you do know about.

good post , the fault issue is debtable however "for forcing them to underutilize it."  well no Sony isn't forcing them to underutilize it , putting them  in a difficult position in terms of cost , time , know how etc is the case. the developer has the opportunity to under utilize which brings consequences unrelated to Sony,  bad reviews , p.r , sales etc.

"laziness is the problem tells me that you have absolutely no idea how product development works. Please stick to things you do know about."

if the developer has the opportunity to make a good game but chooses not to  because of the complexity of the architecture , that's a choice they make , one they should be largely be accountable for( i'm not saying Sony don't share the blame but let's be  reasonable). There are developers that have taken advantage of the ps3 and produced excelent title so it' onlly  fair that developers who can't even get  their game to run at a sready frame rate should be considered lazy.

I disagree that developers are accountable, I think it is more the shareholders.  You should maybe try to tell a shareholder putting up $5-30 million up front waiting for a game to be released that it may take 4-6 months longer because of the complexity of development for a given console.  The answer back may be something like, "it has to be done by the same date the other consoles are ready" or "screw it, let's not bother".  The bottom line is that games come down to dollars and cents (sense) and shareholder decisions NOT developers...  If it makes sense to spend the extra time they will but if there is no reason to they won't... 

Given the economic reality that has set in around the world, people are going to pay more attention to game building costs and make decisions based on that.  The PS3 needs to become easier to build for, otherwise a bunch of shareholders will make some real difficult choices on which platforms they will support.

 



MikeB said:
rubido said:
MikeB said:


That's pretty much at the core of the issue. Writing PS3 friendly code, will also result in better performing code on other CPUs. Of course the Cell is by far the most powerful console CPU, so the gains here can be much bigger potentially, but that won't be the case for most multi-platform games as for example you can't have one version of the game have twice as many onscreen enemies than another version without considerable redesign.

And also, your conclusion from those quotes is wrong (the bolded part).

I think you misread what I stated, it means if not coded this way you will get bad results on the PS3, so the gains are much larger. On the 360 the gains are more marginal.

 

 

I don't think I misread what you wrote. You wrote: "Writing PS3 friendly code, will also result in better performing code on other CPUs". That part is completely wrong. Also on what you tried to complement later, you will not get marginal gains on less cores. You will get a loss in performance as there will be more data passing between different threads and more swapping of threads for cpu usage. There are no gains in performance whatsoever... not even marginal. What you get is organization in the code and that is what the developer was talking about.

And still, where did you get those quotes?



@ rubido

NJ5 posted a link, there's another link to the e-mpire forums within this linked posting.

Also from this posting with this includes another multi-platform developer quote:

MikeB: Hopefully more many more developers will lead on the PS3, which could indirectly improve the quality of XBox 360 games as well, a multiplatform games developer posting at Beyond3D (admitted their initial games were just "quick & dirty ports from 360):

"We all agree given the time we'd like to architect for the SPU's first then work back... giving us cache-friendly algorithms by design "



Naughty Dog: "At Naughty Dog, we're pretty sure we should be able to see leaps between games on the PS3 that are even bigger than they were on the PS2."

PS3 vs 360 sales

Some more quotes:

A PC/360 developer responded to this presentation (developer for the PC/360 game Prey):

"Sure you can just about get away with bad code now on the 360"

"Regardless of managed memory or cached memory, the concepts and methods Mike has presented is highly portable. In the case of cached memory, that method results in optimized cache locality and cache utilization (something extremely important when multiple threads are sharing L1 on a single core, and multiple cores are sharing L2), and a predictable way to optimally prefetch. Good data locality, minimal sync points, branch elimination, and vectorization are all required to be able to extract great performance out of the 360 as well."

http://forum.beyond3d.com/showthread.php?t=47057

Nostromo (Neogaf, games developer)

"Using PS3 as a 'lead platform' is the right thing to do if you are going to make a game that has to run on 360 and PS3. The reason is very simple: on PS3 designing your data structure in the proper way is paramount to achieve decent performance (and to scale up..), while your PS3 friendly data will be also 360 friendly data in the vast majority of cases.
This is a big win cause you will definitely be able to get the most from BOTH platforms."

There are more you can find quoted in my Neogaf thread, but I think that's enough for now.



Naughty Dog: "At Naughty Dog, we're pretty sure we should be able to see leaps between games on the PS3 that are even bigger than they were on the PS2."

PS3 vs 360 sales

A game for both the PS3 and Xbox 360 cannot generally:

  • Be larger than 6.8GB
  • Use more memory than the PS3 has available
  • Be more than either the CPUs of the PS3 or Xbox 360 can handle
  • Use graphics techniques which are incompatible on either. Deferred rendering is one such nono.

Luckily then that they are so similar in performance.



Tease.