@ Deneidez

If you want to, you can just stick to using the PPE. Genji 2, a PS3 exclusive did not use any SPEs at all. The game still looked decent though.



It's not a difficult platform to develop for when designing your game engine from scratch.

To quote Media Molecule:

"We went multiprocessor from the beginning, went multicore, and not having legacy code to hamper that code was such a blessing."

"The PS3 has this insane quantity of power, which seems to take a cloth simulator, and then lots more you want to throw at it. It's just a really enjoyable machine to code for."



It's a most excellent basis for deep knowledge of more modern gaming hardware, likewise if you developed high quality games for the Amiga, NeoGeo, Snes, etc. At least you know how hardware works under the hood to some extend and your perspectives aren't limited to what middleware and high level development environments allow you to know.

@MikeB

And how well does x86 emulator work on PS3? You can even make programs that do not work 'at all' with it. Of course you can make programs on PS3 that PC can't emulate currently. Both platforms are a bit too different to make emulator work properly when they are about equal in power. Anyway the thing is you can make program that runs well with 90's pentium computers and doesn't run nearly at all with PS3 x86 emulator.

To put things into proper perspective. (As in showing the other side too, which you usually don't show. )

Just like half life 2 nearly ran on PS3 when using PPE only.

(HL2 & source engine were released already in 2004 ... )

So making the engine, game logic, everything from the scratch is a blessing instead of burden? Uhm... I beg to differ. I find it more costly even if there would be only code for one core. For example have you ever wondered why games today usually run on third party engine, which have been usually done with the burden of old code?

I would like to know what kind of budget Media Molecule had with LBP or any other company that has made fully their games. In the end its all about money always. Making games without profits just won't work, which actually reminds me this article...

http://www.forbes.com/2008/11/21/games-eedar-developers-tech-ebiz-cx_mji_1121eedar.html?partner=yahootix

@ Deneidez

I don't get your point, of course a modern PC or PS3 can emulate an 80's 7 Mhz Amiga 500. I just pointed out it took a lot of time and upgrades for this approach to be feasible on alternative platforms. A 80s PC, nomatter how well upgraded, could never emulate a 80s Amiga, vice versa is possible though through PC Task for instance.



Designing a new game engine is nothing new, it happens everywhere, also on the PC by ambitious game developers. Many demoscene coders even think it's fun to design new better engines from scratch, it's a hobby to most of them. Showing off their coding skills.



Good programmers are expensive and developing your own technology takes a lot of time and effort, if you can do with just some game designers, script writers and artists it saves the company money. It's the low level technology provider's job and responsibility in such cases to modernize their technology for modern hardware.

@MikeB

Looks like the thing is that you didn't even read my post or you just didn't understand it or what happened? There were nothing about amiga in it, but you somehow got it from somewhere. :)

Demoscene coders do it for free and with no expenses and for fun. Companies pay for their workers to do the same thing. I wonder what workers in companies would say if you would say that they are doing their work just for fun? As I said its all about money in game industry.

"Do it for fun only, because you are not getting paid." , hehe... That sounds funny tho. :D

And good PS3 programmers are even more expensive as there aren't many of them. The ones who knew how to do stuff well for heterogeneous platform are dead or retired. However you are saying that solution would be make less and make it better? I wonder what players would think if your game would be over in 1h...

 

Btw, I am still waiting for that concrete example what you could do with all the untapped power of PS3.

selnor said: MikeB said: @ selnor Well you can argue all you want with IBM. They are the figures and in real world terms the Cell as a whole gets nowhere near 200GFLOPS inside a PS3 for game usage. You can argue with IBM all you want. I have spoken to several IBM technical specialists, their perspective is the Xenon's theorectical peak is 76.8 GFLops (the equivalent of about 3 Cell PPUs) and the PS3's Cell theoretical peak is 218 GFlops, this while the Cell inside the PS3 is able to achieve much higher real world efficiency and the SPEs are much better suited at various tasks. For example in 2005 Mikael Haglund, technical specialist of IBM Sweden gave some Cell presentations at the AmiGBG fair in Sweden, which many DemoScene, game developers, OS and application developers attending. Our team did some life coverage, interviews, I wrote a show report and we had our own stand at the event. I have seen it all now. You argue that IBM's own website provides the cells Theoretical peak performance and you dont believe it? You seem miss informed to. 218 GFLOPS is the speculative theoretical performance of the cell. To get this number they test in a static environment 1 SPE on it's own. to get 25.12 GFLOPS. They than X that by 8. That is not theoretical it's speculative thoretical. There is a difference. Now Theoretical performance of the Cell testing is done as follows. The entire Cell is tested in the same environment. So thats all 8 SPE's and PPE (as SPES cannot run without the PPE it is full theoretical Cell performance). So testing the Cell in the same environment provides us with 155 GFLOPS according to IBM themselves. Sony chose to take the speculative theoretical performance of which the numbers come from a single SPE test at 25.12 GFLOPS. Now again, 2 SPE's in the PS3's Cell are NEVER used for games. That is fact. So that leaves 6 SPE's for games. So immediately 155.5GFLOPS in the PS3 Cell is not available for game programmers. Now the Xenon unfortunately the Theoretical performance peak numbers are not available from IBM's website. But many Tech sites claim that the theoretical Performance is 115 GFLOPS tested as the whole CPU (rather than testing on thread and adding them up). We know as fact that only 3% of the xenon is used for OS so the rest is available to Programmers for games. With this information, many people do not choose to show this info and continue to use Sony's claims (which is based off Speculative Theoretical performance). Now go ahead and argue with IBM's official Theoretical performance of the CEll in PS3 if you want. But it makes you look extremely silly.

I believe as you have boycotted this post that you admit your wrong and IBM's official numbers are right, yes?

 

@ Deneidez

I already said I don't understand your point, you replied to a comment which fully regarded the Amiga in reply to Crumas2's Amiga comments.

Yes it wastes a lot of CPU cycles to emulate x86 on PPC and vice versa, if that was your point...



I just pointed out it can be fun, game development can be fun, game engine development from scratch can be fun. There are enough demosceners out there who would like to make a job out of their hobby.

The devs behind Max Payne for instance also made a cool award winning Amiga demo back in 2001 using old 1992 AGA technology. It seems they like to code for fun as well.



There's still plenty of talent out there. It's however true many schools nowadays are doing a bad job (maybe they get their computers and class material from Microsoft for free).



The options are too many to name, it ranges from being able to have more action on screen to better artificial intelligence, from better graphics in a very broad sense to having otherwise too demanding OS background functionality.

@ selnor

No offence, buddy. But my replies to you turned into book-works in the past, it's useless, there's no end. Our perspectives, insights and conclusions are too wide apart on almost any issue. You want to believe the 360 is the most powerful console out there, that's OK with me.

MikeB said: Crumas said: The problem was that the architecture wasn't easily scalable like the PC/Mac architectures. Of course it was, you could upgrade 80s high end Amigas in the 90s with graphics cards which used the same chips as their PC or Mac equivalents. Classic Amigas can even be upgraded to run AmigaOS4.x from 2008. Amigas from the 80s could be upgraded like no PC or Mac from this era could.

The Amiga hardware wasn't scalable Mike.  Switching Amigas to use the same graphics hardware as a PC or Mac wasn't scaling Jay Miner's masterpiece of architecture, it was changing the architecture to the PC/Mac graphics architecture.

Now the OS... Dale Luck and RJ Michaels wrote that in such a way as it WAS scalable... the APIs were fantastic.  No, a work of art.  Just those incredibly complex C includes that were frustrating until I got used to them...

Off topic... did you ever hear the story of Joe Pillow and the Dancing Fools?  I was at RJ Michael's last AmigaWorld keynote in Chicago, and he told the story.  A testiment to those who knew they were doing something great, and were not going to allow something like the limits of human endurance stop them. 

 

@MikeB

Still hobbies != work , just having fun => no money, having fun while working => better job, but it still when you work you must get paid. Whatever you throw at me you can't null the fact that work means that you get paid too.

This shows how biased and paranoia you are. Behind every corner theres a bad MS guy lurking and waiting for opportunity to strike... :D

Anyway, I actually asked is there anything about heterogeneous platforms in parallel computing course and answer was that nearest we get is CUDA programming, because theres no indication that heterogeneous platforms will be back. Somehow you seem certain that they will take over the world soon. Afaik theres no indication about that happening.

Show me game that uses SPUs for most of its AI. As far I have seen games that use poor PPU only for AI. Why it should use mostly SPUs? Because that was the source of power of PS3. X360 has 3xPPU, which is a bit more than one PPU. And more action on the screen? Humm... I just made point about that one in other thread. Graphics... It might have slight advantage, but its still slight. :)

Btw, how much you get paid for defending SONY? In the end MS & SONY are just companies, which try to make money, which don't care a squat about you or any of their fans as long as they just get money.

@ crumas2

There was a ECS/AGA compatible better scaleable chipset under development at Amiga Inc. But Commodore invested into R&D far too little and far too late. However for the long run trying to do everything yourself would never have been feasible with the verge of many big heavily funded dedicated chip companies.

Dave Haynie although his Amiga 3000 design from 1990 was awarded as the best system available at the time, he really wanted to have this AAA chipset with Andrea, Linda, Monica and Mary custom chips in there to make the system as revolutionary like the Amiga was back in 1985.

'Commodore Engineering developed an entirely modular system of upgradeable pieces, consisting of the motherboard, Chipset module, host processor module and Open System Bus Slots. The system bus would also be processor-independent allowing it to be upgraded over time without any performance loss.'

'The Chip Module represented a move away from placing the custom chips on the motherboard. This would allow them to be upgraded as time progressed through simple plug-in expansion.'

Commodore management only acknowledged the importance of their Amiga division right before the end when nothing could be done about it (The Amiga engineers foresaw what would happen and stated in AmigaOS eastereggs that Commodore was killing the Amiga), the last sparkle of hope was the Amiga CD32 for them but with a big already produced US launch stock was blocked by court due to alledged patent violations, it was all too late.