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.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: fyo who wrote (66150)7/19/1999 9:13:00 PM
From: Ali Chen  Read Replies (1) | Respond to of 1578436
 
Fyo, re <latency>
First, you asked whether the DDR will hurt performance,
but did not specify as relative to which design.
Since you were talking about K7, and K7
is already faster than P-III, it is natural
to assume that you were talking about reference
K7 SRAM. Therefore my remark: the latency
will not be worse:)

After all, why do you think that the Intel
SRAM latency is so good? If I recall correctly
from something like ISSC-98,
their SRAM utilizes multiple interleaved banks
of regular SRAMS, and they never disclosed
their actual latency.



To: fyo who wrote (66150)7/19/1999 10:44:00 PM
From: Petz  Respond to of 1578436
 
fyo, re: using DDR-SRAM in L2 cache. You're right, the latency would be same as 350 MHz SRAM so performance would be somewhere between true full-speed L2 and "standard" Athlon half-speed L2. It would particularly speed up DSP applications and graphics (including games) where long vectors of data are manipulated. In those cases, latency doesn't matter as much as throughput.

Petz


HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News