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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: wily who wrote (103705)5/26/2000 11:21:00 PM
From: deibutfeif  Read Replies (1) | Respond to of 186894
 
Wily, re:...since OUM is nonvolatile, it can replace SRAM...

Maybe some definitions would be useful here:

a) Non-volatile means the memory retains state even when power is removed.

b) Static (as in Static RAM or SRAM) means the memory retains state even in the absence of periodic refresh signals.

c) Dynamic (as in Dynamic RAM or DRAM) means the memory requires periodic refreshing to maintain state.

Any memory can be a candidate to replace SRAM (or the SRAM used to implement on-chip cache, if you prefer), if its fast enough and compatible with the process technology, etc.

It doesn't need to be non-volatile, nor even static.

~dbf



To: wily who wrote (103705)5/27/2000 1:59:00 AM
From: Scumbria  Read Replies (2) | Respond to of 186894
 
Wily,

I'd like to know, is there a point at which increasing cache size yields diminishing returns?

Absolutely. A 32K L1 cache will hit about 97% of the time running Winstone. Doubling that to 64K will raise that to about 98%. Double it again to 128K, and you get about 98.5%.

These differences are significant for performance, but the returns are definitely diminishing.

Scumbria


HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2026 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News