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
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Elmer who wrote (70035)8/27/1999 11:02:00 AM
From: Ali Chen  Read Replies (1) of 1572969
 
Elmer <shows a real lack of understanding of the obvious problems AMD has already demonstrated,>

Yes, it's about yourself. You still did not get it,
do you? Sure problems are obvious, but the cause is
not, especially for you. You need to turn your brain
by 90 degrees.

Everything you write shows your lack
of understanding.

You say:
"AMD has major problems with large die and added L2"
In reality, AMD has made tremendous inroads
in process/design technology to get K6 to 500MHz
point. With new K7 microarchitecture, the pay-back
is still unbelievable for some.

You say:
"Intel is the only company shipping product on .18u while AMD hasn't shown any ability to significantly manufacture K7s on .25u"

In reality Intel 0.18 products did not get
even to 500MHz yet while AMD is shipping Athlons
on 0.25 um that are overclockable to 750MHz,
see Tomshardware.

You say:
<Off die full speed L2 ain't gonna happen above 600MHz. It's unmanufactureable and untestable. It's on die L2 (which AMD can't manufacture) or it's off die at 1/2 core speed.>
All 90-degree wrong again. Yes, 600MHz SRAM will not
happen soon. But it does not matter, Fudd! With proper
L2 design the 1/2 or 1/3 or even 1/4 speed for L2
does not impact performance significantly (on
small-size workloads) while bigger caches
vastly reduce miss rate on modern ever increasing
workloads. On-die caches will never catch up with
application demands because it is applications who
drives decisions to increase caches, not opposite.
The Ahlon flexible L2 cache design will allow
to employ ANY off-shelf cache no matter how
slow it is - still with performance benefits.
And your P-III can't get above 600 exactly due
to the lack of this flexibility - such a
short-sighted design it is.

Now, do you have more fudd for your short
thoughts?
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

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