Japanese article about PS3 backwards compatablilty

[TheChefO]

$5 * 20 million consoles could be the difference between profit and loss for the division. When you're looking at such large numbers of units, every penny saving amounts to a lot.

True - does anyone have figures for the new vs old chip costs (or diesize cost estimates)?

 

[leechan25]

tomatoes tomahtoes .
What I meant was that I agree thats alot of power for one task. Im not sure what kind of tasks either. physics? particle effects?(which the ps2 was very good at ircc) background gui? who knows.

I agree, having the ee/gs managing one task like particle effects would mindblowing because it PS2 ability to handle such effects is outstanding. Maybe developers would off load multiple task to the ee/gs because that chip really is powerful.

 

[Crossbar]

The 'nearly' part is unclear in the article, anyway the accent is clearly on 'EE+GS as is' and if it's different from that of PS2 it'd be trivial.

This leaves a lot of open questions if it is indeed true. Where would you put a discrete 'EE+GS as is'? I find it unlikely that they would also add 32 MB RDRAM and that means it would piggy-back the the RAM of either the Cell or the RSX.

Would it be possible to add the EE+GS to the south bridge chip and use Cells memory? The speed of RDRAM of PS2 is 3.2 GB/s and the SB bus is 2.5 GB/s write and 2.5 GB/s read. These are theoretical speeds. Does anyone know if they end up at about the same real-world performance?

If the EE+GS was added to the SB it would follow the pattern of the PS2 where the PS1 circuitry was addded to the I/O chip. I guess it would make a lot of sense to keep the Cell and the RSX as clean as possible from old circuitry and piggy-back interface circuitry.