Re: CPU Speed
- To: firstname.lastname@example.org
- Subject: Re: CPU Speed
- From: Matus UHLAR - fantomas <email@example.com>
- Date: Sun, 2 Sep 2007 16:21:54 +0200
- Message-id: <[🔎] 20070902142154.GC12841@fantomas.sk>
- Mail-followup-to: firstname.lastname@example.org
- In-reply-to: <200707221343590256.00714B9E@smtp.yandex.ru>
- References: <20070722010834.GA14640@elp.rr.com> <46A2C108.email@example.com> <20070722043417.GA15066@elp.rr.com> <firstname.lastname@example.org> <20070722062729.GA15355@elp.rr.com> <email@example.com> <20070722173313.GC15355@elp.rr.com> <200707221343590256.00714B9E@smtp.yandex.ru>
On 22.07.07 13:43, Aenn Seidhe Priest wrote:
> Socket 7 motherboards used to have L2 cache on themselves (unlike P-IIs
> which had everything - L1 cache, L2 cache - in the processor cartridge).
Well - first PII's had its L2 cache on separate chip on the cartridge, but
L1 cache was on the CPU chip. L2 cache ran at half frequency, while K6-2's
cache ran on the same frequency than the CPU.
First Celerons had no L2 cache, while later Mendoccino's had 128KB L2 cache
on the same chip as CPU and running on the same frequency. Thus, Mendoccinos
could run some apps faster than P2's, while other apps were faster on PII.
(I am comparing CPU at the same frequency)
XEON had 512KB of cache on separate ship and half the frequency, but the
size could "fix" it (maybe still not everywhere).
The "missing" cache could still be located on motherboard, which was often
on (386?)486 machines.
Comparing diffetrent CPUs is not always easy.
Matus UHLAR - fantomas, firstname.lastname@example.org ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Spam = (S)tupid (P)eople's (A)dvertising (M)ethod