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: Petz who wrote (76558)10/22/1999 7:20:00 PM
From: Tenchusatsu  Respond to of 1574055
 
Petz, <How does going from 0.25 to 0.18 micron suddenly allow you to squeeze 8x as much cache on a chip?>

Try sticking even 512K of L2 cache onto a 0.25u chip. For Athlon, that could lead to a die size of over 300 mm2. Given that AMD's manufacturing capacity is limited, a 300 mm2 die size is unacceptable.

On the other hand, 2M of on-die L2 cache even for a 0.18u Athlon is kind of a stretch, unless AMD feels their Dresden fab would give them capacity to spare.

Tenchusatsu



To: Petz who wrote (76558)10/22/1999 9:00:00 PM
From: Paul Engel  Read Replies (1) | Respond to of 1574055
 
Petz - Re: "How does going from 0.25 to 0.18 micron suddenly allow you to squeeze 8x as much cache on a chip? "

If your defect densities are LOW, good yields can still be achieved with 2 MegaBytes L2 cache, even though the die size is essentially HUGE.

Remember - Intel has been manufacturing HP's PA8500 Risc Chip with 1.5 MegaBYTES L1 cache for over a year - using the OLDER 0.25 micron process.

Proof positive that it can be done !

Paul


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