I am surpised no one notice this in Miyamoto interview. He said the GPU and CPU are STILL in development. Development for Revolution GPU and CPU started before Xbox360's, yet it seems like it will not finish development until around a year after Xbox360's. If PS3 GPU and CPU are finished being developed then Revolution would be the furthers behind in terms of development.
Revolution GPU and CPU STILL in development.
- Thread starter SoVos20
- Start date
- Status
I'm not sure whay your getting at Sovos. I don't see why its suprising that the GPU for a system that won't be released for another year isn't finished yet to be honest. There's no reason to complete the GPU for Revolution at around the same time as the XBox 360 GPU when the systems will be released a year appart. That would waste a year of possible technology advancements.
BTW I don't think we know if PS3's GPU is finished or not, in fact its almost certainly not finished either.
BTW I don't think we know if PS3's GPU is finished or not, in fact its almost certainly not finished either.
You guys are only setting yourself up for further disappointment if you think the CPU in Revolution is going to be some remarkable feat of engineering that is both as powerful or more powerful as the Xbox360/PS3 CPUs at the same process, with a lot less heat generated.
IBM has had huge incentive to produce such a processor for years in the cluster/blade market. It's not like Nintendo's the first person to ever ask them to make a cool, but fast processor.
IBM has had huge incentive to produce such a processor for years in the cluster/blade market. It's not like Nintendo's the first person to ever ask them to make a cool, but fast processor.
That tidbit from DeanoC about Nintendo licensing that multi-processor C compiler (XLCPP) is very interesting too. However it's just a rumor so who knows if it's true or not. Multicore isn't rocket science anyway. The quad core ARM/NEC MPCore could be made using the Gekko cores or three lower clocked G5 cores maybe running at 2GHz.
Oh and for all of the people that keep saying you cannot make a console as small as the Revolution one year after Xbox 360 and be as powerful, you haven't looked at all the factors. I'll list a few feasible possibilites from an engineering POV.
1. Have the same cores as the Xbox, but more of them and lower clocked. It's larger but will run a little cooler and would also benefit 65nm which will be available next year.
2. Have two of the same cores plus a third PPU like core all at lower clocks.
3. Use NEC's compact low profile piezo electric pump with coolant as a the cooling solution.
Oh and for all of the people that keep saying you cannot make a console as small as the Revolution one year after Xbox 360 and be as powerful, you haven't looked at all the factors. I'll list a few feasible possibilites from an engineering POV.
1. Have the same cores as the Xbox, but more of them and lower clocked. It's larger but will run a little cooler and would also benefit 65nm which will be available next year.
2. Have two of the same cores plus a third PPU like core all at lower clocks.
3. Use NEC's compact low profile piezo electric pump with coolant as a the cooling solution.
Glonk said:
You grossly underestimate how much time and money has been spent on Rev CPU and GPU. The reality is you have no idea about what you are talkiing about. You keep saying things with out any idea if it is true or not. It is your opinion that Rev won't be as powerful as Xbox360 not FACT. You should learn the difference between an opinion and a FACT, because you keep spewing your opinion as fact. I hate people who call their opinion factual statements. Your opinion that Rev will be weaker than Xbox360 is no more and no less valid then someone else opinion that it will be more powerfull.
Okay, how much time and money have Nintendo spent on Rev's CPU and GPU? After all, money has more bearing than time. If 5 people spend 12 months building a car for £500, or 25 people spend 6 months building one for £20,000, I know which I'd expect to be a better motor. How many man-hours and $millions have gone into Rev development? (with links please)
Shifty Geezer said:
Wait is that the cost of the materials going into the car or the cost to develop the car???
If that is the budget for the materials to build the car then your analogy is skewed not to mention 25 is 5x as many people you have to pay but you've only cut the time in half to 6 months? If you have 5 x the number of people then it should be 5 x faster to complete the project.
I pointing out that the time given over to a project isn't the only factor. Man hours and finance have a bigger bearing.
Like my car analogy, 2 GPUs are designed. One is given five years, has five people working on the project and a research budget of $50,000. The other has less time, only one year, but 200 people working on it and a budget of $50,000,000. Which would you expect to be the better developed GPU?
Just one chip being in development longer != better chip. There's other factors to consider.
Like my car analogy, 2 GPUs are designed. One is given five years, has five people working on the project and a research budget of $50,000. The other has less time, only one year, but 200 people working on it and a budget of $50,000,000. Which would you expect to be the better developed GPU?
Just one chip being in development longer != better chip. There's other factors to consider.
The thread seems to be SoVos20 saying because Rev's chips are still in development, having been in development the longest, they are likely better. I'm just clarifying that's not a fair assumption which SoVos20 doesn't appear aware of.
- Status
