SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?


To: Ali Chen who wrote (79780)5/8/2002 4:20:51 PM
From: Harvey AllenRead Replies (1) | Respond to of 275872
 
Sony mulls memory options for next game machine
By Jack Robertson
EBN
(05/08/02, 01:56:40 PM EST)


TOKYO -- The projected DDR-II SDRAM memory is a memory candidate for the next generation Sony Corp. electronic game machine, but a top executive told EBN Wednesday that even that advanced new chip may not be fast enough.

Kenshi Manabe, senior vice president of Sony Computer Entertainment Semiconductor division, said the yet-unnamed successor to the current Sony Playstation-II "will need memory with incredibly high speed and tens-of-gigahertz bandwidth.I'm not sure that even what we have seen for DDR-II on the present (industry) memory roadmaps will be high enough performance for us.

"The next (game machine) may require an entirely new type of memory architecture. Right now we are discussing with our semiconductor development partners, IBM and Toshiba, how we can solve the memory technology issue," he said.

As previously reported, the three chip companies have a joint development program working at IBM's Austin, Tex., research center to design the next game machine's advanced processor and manufacturing process.

Separately, Sony and Toshiba have a joint development for the machine's highly-integrated graphics chip.

Manabe said it is certain that even the highest projected speed Direct Rambus DRAM or DDR-I memory aren't fast enough with high enough bandwidth for the next Sony Corp. game machine.

The Sony official also said the new machine will need an equally super-speed memory bandwidth interface. He said HyperTransport designed by an Advanced Micro Devices-promoted consortium "may be a candidate, but at this time we have made no decision."

ebnonline.com



To: Ali Chen who wrote (79780)5/8/2002 8:44:28 PM
From: hmalyRespond to of 275872
 
Ali Re...Why should anyone read this blurb? There is
no such thing as "necessary bandwidth it feed a xxxGHz"
processors, any bandwidth is not enough since times
when CPU core speeds started to exceed the
multi-cycle memory access time.<<<<


I don't believe either THG or Anands are saying it is enough. What they are saying is that P4 works faster with Rambus 800 or pc1066, rather than any current DDR solution. I am assuming they would know as they have already tested the P4 with each already.

I think the latency of PC1066 remains largely the same,
and this causes those weird discrepancies in
benchmarks scalability. Looks more and more like a dead
end, pawing the road to DDR+.<<<<<<<


That may or may not be true, as we haven't seen what DDR+ can do yet either. At any rate, DDR+ isn't scheduled until Q4 or q1-03, so Clawhammer with its HT buses should equal the bandwidth of DDR+ , as HT doubles the bandwidth to 3.2 g. And I am not sure, but Clawhammer may be able to address DDR+.



To: Ali Chen who wrote (79780)5/9/2002 5:53:42 PM
From: PetzRead Replies (1) | Respond to of 275872
 
re:anomalous RDRAM benchmark results. Perhaps the PC1066 parts have fewer open pages allowed on each chip. I believe a while back RMBS reduced the spec on #pages to increase yields. Perhaps the spec for 1066 was cut or manufacturers found they needed to cut corners to yield high enough at 266MHz FSB.

Petz