Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.sys.intel > #56

Re: how can a newer processor have worse L1 cache?

Date Thu, 07 Apr 2011 00:38:22 -0400
From Yousuf Khan <bbbl67@spammenot.yahoo.com>
User-Agent Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2.15) Gecko/20110303 Lightning/1.0b2 Thunderbird/3.1.9
MIME-Version 1.0
Newsgroups comp.sys.intel
Subject Re: how can a newer processor have worse L1 cache?
References <2ea0be0202d503e5e069233f5585410b@rip.ax.lt>
In-Reply-To <2ea0be0202d503e5e069233f5585410b@rip.ax.lt>
Content-Type text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding 7bit
NNTP-Posting-Host bas11-ottawa23-1128668341.dsl.bell.ca
Message-ID <4d9d3fc6@news.bnb-lp.com> (permalink)
X-Trace news.bnb-lp.com 1302151110 bas11-ottawa23-1128668341.dsl.bell.ca (7 Apr 2011 00:38:30 -0400)
Organization Send abuse or DMCA complaints to abuse@bnb-lp.com
Lines 18
X-Authenticated-User vfa100
X-DMCA-Complaints Send abuse or DMCA complaints to abuse@bnb-lp.com
X-DMCA-Complaints The subject line should contain only the 4 letters DMCA
Path csiph.com!x330-a1.tempe.blueboxinc.net!newsfeed.hal-mli.net!feeder1.hal-mli.net!news.glorb.com!news.bnb-lp.com!not-for-mail
Xref x330-a1.tempe.blueboxinc.net comp.sys.intel:56

Show key headers only | View raw


On 06/04/2011 9:17 AM, Anne Onime wrote:
> I have seen many peoples castigating the Prescott compared to
> the Northwood P4. We had a stretch of rainy days, so I tested
> some old PCs - I had a draw full of Pentium IVs to swap in them.
> One benchmark for L1 latency came out thus:
> Northwood 2 cycles
> Prescott 5 cycles
> Cedar Mill 4 cycles
> How did Intel bugger up the L1 cache?

Thinking back to my ancient history, it had to do with the fact that 
Intel was aiming for 10GHz back then, so they stretched the instruction 
pipeline stage on the newer P4's, and this affected their latency. 
Higher latency in exchange for higher throughput. They expected that the 
greater clock rate would make up for the increased relative latency, by 
keeping the absolute latency the same.

	Yousuf Khan

Back to comp.sys.intel | Previous | NextPrevious in thread | Next in thread | Find similar


Thread

how can a newer processor have worse L1 cache? Anne Onime <anonymous@rip.ax.lt> - 2011-04-06 15:17 +0200
  Re: how can a newer processor have worse L1 cache? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2011-04-07 00:38 -0400
  Re: how can a newer processor have worse L1 cache? "Jim" <jim@wtf.invalid> - 2011-04-07 05:48 -0400
  Re: how can a newer processor have worse L1 cache? Robert Myers <rbmyersusa@gmail.com> - 2011-04-10 15:19 -0400

csiph-web