Congratulations to Vince

[Titanio]

I think some of you should tell deadmeat your sorry , as he said from the begining it would be no where near 1 tflop .

Anyway 256 is what i expected in terms of pr , in real world performance it will be much lower but still a nice cpu . Thank god they went with an nvidia gpu though

Did anyone ever predict a ITflop PE? I don't remember that being the case (and certainly would have been a little optimistic ). People were saying, though, that a 4-PE BE would yield 1Tflop. That would be the case today, if they could put 4PEs together - they are 256Gflops each.

 

[Farid]

Hello? I never stated Vince said CELL would be 1TFLOPs so I have no idea why you are saying I got owned. I see this thread is going to self destruct.....

Chill Acert...
I was owned by V3, because he (and Paul too, BTW) posted almost exactly what I posted, before me. Nothing' wrong about your or anything.

And about the "almost everybody", well, then let's change it to"almost everybody who aren't batshit nuts".

 

[Acert93]

Hello? I never stated Vince said CELL would be 1TFLOPs so I have no idea why you are saying I got owned. I see this thread is going to self destruct.....

Chill Acert... I was owned by V3, because he (and Paul too, BTW) posted almost exactly what I posted, before me.

And about the "almost everybody", well, then let's change it to"almost everybody who aren't batshit nuts".

Ok, I will chill Since you tagged it onto a reply to me I thought you were saying I was owned

Btw, if no one has said it yet..... CELL is looking pretty sweet. nVidia GPU + CELL + Rambus + BR = I had a PS, was not impressed by PS2 (combo of game direction and the fact my GCN runs in HDTV on my computer monitor), but PS3 is looking like a winner and I always like a good winner If they make ways for developers to keep costs down and develop products quicker and on budget, well, it is Sony's game to lose at this point. Next gen is going to rock...

 

[Spidermate]

I think some of you should tell deadmeat your sorry , as he said from the begining it would be no where near 1 tflop .

Anyway 256 is what i expected in terms of pr , in real world performance it will be much lower but still a nice cpu . Thank god they went with an nvidia gpu though

Um... Isn't 256 GFLOPS for referring to 1 PE? If so, wouldn't four PEs hit a teraflop? Unless I'm seriously missing something, he was DEAD-wrong.

Also, it is only on a 90nm from what I know.

 

[jvd]

I think some of you should tell deadmeat your sorry , as he said from the begining it would be no where near 1 tflop .

Anyway 256 is what i expected in terms of pr , in real world performance it will be much lower but still a nice cpu . Thank god they went with an nvidia gpu though

Um... Isn't 256 GFLOPS for referring to 1 PE? If so, wouldn't four PEs hit a teraflop? Unless I'm seriously missing something, he was DEAD-wrong.

Also, it is only on a 90nm from what I know.

Whenever I have talked about cell , I have talked about the cell chip in the ps 3 as this is a console forum , i don't care what a million dollar server can do .

Its allways been about the cell chip in the ps3 and it allways will be as we are on a console forum and that is the only area that we need to discuss performance

 

[Devourer]

PS3 will reach the teraflops figure. Just wait the March presentation. Cell chips in PS3 will be fabbed at 65 nm. Congrats to Vince and Panajev.

 

[Spidermate]

Fair enough, but I don't think neither of us know how many PEs will be going into the PS3 right now. We only know what one of them consist of. I doubt that's enough to predict anything solid just yet.

 

[jvd]

Fair enough, but I don't think neither of us know how many PEs will be going into the PS3 right now. We only know what one of them consist of. I doubt that's enough to predict anything solid just yet.

Oh i dunno a 1pe with 8 apus is

prototype chip is 221 mm(2), integrates 234 million transistors, and is fabricated with 90 nanometer SOI technology

So 2pe with 8apus would be 442mm(2) with 468 million transistors on 90nm ? If it could clock at 4.5 ghz that would give roughly 500gflops performance ? That would be a huge ass chip even at 65nm it will be a huge chip with crappy yields .

I really don't see them putting anything more than the 1pe 8 apu chip .

I don't see a need for it either .

Power , cost , yield and cooling are all factors . Remember this is going to be a 300$ console that at the most will cost sony 500$ to make .

I'm predicting with sony and nividia pr crap they will fudge a 1tflop number but its real world performance will be know where near that high

 

[Guden Oden]

Jvd, with all due respect, but you're just some guy living in f'n Jersey who according to himself works in a video game store. What would you possibly know about chip yields I ask?

 

[cthellis42]

I think some of you should tell deadmeat your sorry , as he said from the begining it would be no where near 1 tflop .

Feel free to re-read V3's post, as I think he summed it up best.

Whenever I have talked about cell , I have talked about the cell chip in the ps 3 as this is a console forum , i don't care what a million dollar server can do .

And of course that's exactly what hasn't been established yet, so on that point there is no "right" or "wrong" about any of that, but rather still conjecture as people await the official PS3 presentation itself.

 

[Acert93]

PS3 will reach the teraflops figure. Just wait the March presentation. Cell chips in PS3 will be fabbed at 65 nm. Congrats to Vince and Panajev.

1. Does it need to reach 1TFLOPs? If XeCPU is in the 80GFLOPs range, why would it need 1TFLOPs? I am not going to compare them yet (we still do not know enough about both designs... XeCPU seems to be a good at general processing with multiple PPC cores and if the CELL is doing vertex shading and XeCPU is not the CPUs are then really playing different designs goals), but lets say CELL is 2-4x more powerful. Why would it then need to be 8-16x as powerful? That power costs money and with Rambus RAM, BR, licensed GPU tech, and a lot of investment to make a return on plus a lot of work to be done on making development easier/faster/more affordable for developers that extra money may be best invested elsewhere. Also, smart design is better than spamming power everywhere. Will the PS3 have enough memory and memory bandwidth to feed 1TFLOP? If the memory bandwidth is shared you have a 1TFLOPs CELL and a GPU fighting for ~50GB/s. Considering current top of the line GPUs use ~32GB/s I am not sure there will be fast enough RAM, or enough, to feed 1TFLOPs. Maybe there is, but smart effecient design is important and there is no point having power sitting idly by.

2. Heat. Consoles are small so heat is a concern; a large console to accomodate heat would have a certain image issue. People like smaller systems in general. Throwing in 4 CELLs could produce a LOT of heat.

3. Power consumption. If A|SPU are 4W each, then add the PU, nVidia GPU, BR drive, and so on you have a machine sucking a lot of power. If you have 4 CELL cores you are looking at a high power requirement.

4. Production Yields and cost. Yields may be low if the system is hot which means higher costs. If a 256GFLOPs CELL is 2-4x as powerful as the XeCPU (if... the real question is how that power will translate into better games) why would they need to make it cost more? Cross platforming is important so spending a lot of money on CPU realestate that many developers (like EA) most likely wont use is an exepensive luxury. You add in the fact you would get lower yields and that just magnifies the issue.

5. Die size and cost. Sticking 4 CELLs on a single chip make large chips, even at 65nm. Add in the above reasons, well, I do not see why.

Power is nice, but smart design is better. If the PS3 ends up with 256MB of RAM then where is the PS3 going to get the memory to feed that much power? CELL + nVidia GPU + BR + Rambus is a killer combo. I would not be totally surprised if the PSE BE has 2 CELLS (although 1 looks like a safe bet), but I would be shocked at 4.

 

[PiNkY]

Jvd, with all due respect, but you're just some guy living in f'n Jersey who according to himself works in a video game store. What would you possibly know about chip yields I ask? Laughing

And what would be your point exactly? That you're looking forward to your twelvth birthday?

 

[Spidermate]

So 2pe with 8apus would be 442mm(2) with 468 million transistors on 90nm ? If it could clock at 4.5 ghz that would give roughly 500gflops performance ? That would be a huge ass chip even at 65nm it will be a huge chip with crappy yields .

I really don't see them putting anything more than the 1pe 8 apu chip .

I don't see a need for it either .

Power , cost , yield and cooling are all factors . Remember this is going to be a 300$ console that at the most will cost sony 500$ to make .

I'm predicting with sony and nividia pr crap they will fudge a 1tflop number but its real world performance will be know where near that high

Those figures may seem right, but we've gone through this a dozens of times before. I think it would be best to wait.

 

[jvd]

Jvd, with all due respect, but you're just some guy living in f'n Jersey who according to himself works in a video game store. What would you possibly know about chip yields I ask?

The point is we don't know and while you can assume anyhting there is no reason to go over board . That is all the cell threads do is go over board .

I'm sure if sony wanted to they could put a 1 tflop cell in the ps3 , but they would be eating a crap load of money on each one .


Btw , i work like 3 hours a month at the game store and its just to save money on video games , my family spends easily 300-400 a month on video games and getting 20% off that for working 3 hours a week at the store my friend manages is a good deal to me

Those figures may seem right, but we've gone through this a dozens of times before. I think it would be best to wait.

Well i'm not sure the numbers are correct , I just figure another pe and 8 apus should take up the same transistors as the first . There might even need more transistors than that for any kind of connections between the pes . So if anything it may go up .

I also don't know much about how to find the area and size of the chip. So i just figured doubling it would be okay as once again another pe and 8 apus should take up the same amount of die space

 

[DemoCoder]

Acert,
The goal shouldn't be to just build a console better than the X-Box2. If Sony was the only company (monopoly on video game market) would you suggest that they never increase power at all?

The purpose of having such a huge increase in power is to meet the needs of future games, which need massive amounts of processing to realistically simulate physics, AI, sound, etc -- to increase realism and believability.

It takes a man of vision, crazy, or creative genius to take a big financial risk, and go all out on something revolutionary, but that is how we make the big leaps.

 

[wco81]

Fair enough, but I don't think neither of us know how many PEs will be going into the PS3 right now. We only know what one of them consist of. I doubt that's enough to predict anything solid just yet.

If Sony does unveil the specs in March or at the latest, at E3, you would think the developers would know by now what kind of specs to target for the games right?

They might not know price and launch date but surely Sony has given them some idea of what kind of performance targets to shoot for?

Given that we're close to unveiling the final benchmarks?

 

[Titanio]

Fair enough, but I don't think neither of us know how many PEs will be going into the PS3 right now. We only know what one of them consist of. I doubt that's enough to predict anything solid just yet.

If Sony does unveil the specs in March or at the latest, at E3, you would think the developers would know by now what kind of specs to target for the games right?

They might not know price and launch date but surely Sony has given them some idea of what kind of performance targets to shoot for?

Given that we're close to unveiling the final benchmarks?

Some devs may know, but not a lot I would say. I remember back when PS2 specs was first unveiled, many of developers were in the dark, and it was the first they heard of final specs.

 

[Devourer]

PS3 will reach the teraflops figure. Just wait the March presentation. Cell chips in PS3 will be fabbed at 65 nm. Congrats to Vince and Panajev.

1. Does it need to reach 1TFLOPs? If XeCPU is in the 80GFLOPs range, why would it need 1TFLOPs? I am not going to compare them yet (we still do not know enough about both designs... XeCPU seems to be a good at general processing with multiple PPC cores and if the CELL is doing vertex shading and XeCPU is not the CPUs are then really playing different designs goals), but lets say CELL is 2-4x more powerful. Why would it then need to be 8-16x as powerful? That power costs money and with Rambus RAM, BR, licensed GPU tech, and a lot of investment to make a return on plus a lot of work to be done on making development easier/faster/more affordable for developers that extra money may be best invested elsewhere. Also, smart design is better than spamming power everywhere. Will the PS3 have enough memory and memory bandwidth to feed 1TFLOP? If the memory bandwidth is shared you have a 1TFLOPs CELL and a GPU fighting for ~50GB/s. Considering current top of the line GPUs use ~32GB/s I am not sure there will be fast enough RAM, or enough, to feed 1TFLOPs. Maybe there is, but smart effecient design is important and there is no point having power sitting idly by.

Power is nice, but smart design is better. If the PS3 ends up with 256MB of RAM then where is the PS3 going to get the memory to feed that much power? CELL + nVidia GPU + BR + Rambus is a killer combo. I would not be totally surprised if the PSE BE has 2 CELLS (although 1 looks like a safe bet), but I would be shocked at 4.

Prepare to be shocked. That's what Kutaragi is aiming for since 2000. He is so serious about his next console it's scary.

By the way, PS3 RAM will have 100 GB bandwidth, not 50

 

[Paul]

Prepare to be shocked. That's what Kutaragi is aiming for since 2000. He is so serious about his next console it's scary.

By the way, PS3 RAM will have 100 GB bandwidth, not 50

 

[Acert93]

Acert,
The goal shouldn't be to just build a console better than the X-Box2. If Sony was the only company (monopoly on video game market) would you suggest that they never increase power at all?

The purpose of having such a huge increase in power is to meet the needs of future games, which need massive amounts of processing to realistically simulate physics, AI, sound, etc -- to increase realism and believability.

It takes a man of vision, crazy, or creative genius to take a big financial risk, and go all out on something revolutionary, but that is how we make the big leaps.

I disagree. The issue is not pure power but priceerformance. Remember, these machines need to come in at ~$300 and hit ~$200 in 2-3 years.

I never said not to increase the power at all, but consider: Top of the line P4 is in the high 20's gof GFLOPs (that was off a website, IGN I believe) and XeCPU is rumored to be ~80ish GFLOPs (no on who knows is saying, so pure speculation). If 1 CELL is 256GFLOPs you are looking at 8x the performance of a top of the line desktop and 3x the performance of the XeCPU (although comparisons at this point are sketchy).

So I ask: Why would they need to make it even more powerful? Would not the cost for more CELLs be better spent on other areas? It is like the body builder with big arms and chicken legs. He might impress a few people, but when it comes to the competition or sports the well rounded athlete kicks his butt.

I agree we need more power for AI, physics and general believability but 1 CELL is a good step toward that. 1TFLOPs is a totally arbitrary number. Wny not 2TFLOPs? or 16TFLOPs? 1 CELL is substantially faster than what is going to be available, so why 4? It will cost more AND most cross platformed games wont take advantage of it.

If the concern was "massive amounts of processing to realistically simulate physics, AI, sound, etc" on a $300 machine I would look to spend some more resources on memory. This is just me, but a 256GFLOPs CELL and 512MB of XDR is more appealing to me than a 1TFLOP machine with 256MB of XDR although we do not know enough to compare pricing. A 1TFLOPs/256MB PS3 may cost a lot more than a 256GFLOPs/512MB CELL so the comparison would not be fair.

Anyhow, good design is more important than putting all your eggs in one basket. 1TFLOPs is just a number. The rest of the system needs to be able to feed that CPUs. And at a $300 price point, I would think if you were beating your competition by 3x in CPU performance you would look to put more money in other areas. You have to remember if they spend too much on CELL and neglect other areas that leaves room for competitors to move in. A great CPU with a weak GPU would be worse than a solid CPU and GPU. Balance is more important than pure power.