Date: Thu, 14 Aug 1997 10:23:35 +0300 (IDT) From: Nadav Eiron <nadav@CS.Technion.AC.IL> To: Atipa <freebsd@atipa.com> Cc: Ada T Lim <ada@not-enough.bandwidth.org>, "Jamil J. Weatherbee" <jamil@counterintelligence.ml.org>, freebsd-hackers@FreeBSD.ORG Subject: Re: Alpha Benchmarks was: Re: FreeBSD --- ALPHA Message-ID: <Pine.GSO.3.95-heb-2.07.970814102152.10665A-100000@csd> In-Reply-To: <Pine.BSF.3.91.970813095831.17775A-100000@dot.ishiboo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 13 Aug 1997, Atipa wrote:
>
>
> On Wed, 13 Aug 1997, Ada T Lim wrote:
>
[snip]
>
> Not true. They are noticably different. The benchmarks show:
>
> +--------------+------------+-----------+
> | 533MHz | SPECint95 | SPECfp95 |
> +--------------|------------|-----------|
> | 21164 (LX) | 16.4 | 22.5 |
> +--------------|------------|-----------|
> | 21164PC (SX) | 14.0 | 17.0 |
> +--------------|------------|-----------|
> | 300MHz P-II | 11.6 | 7.2 | (Intel's closest to referece)
> +--------------+------------+-----------+
>
> Max programmable bus speed for the 21164 is 200MHz, and the highest for
> the 21164PC is 133MHz. NT (Linux/OpenBSD/NetBSD???) requires the "Windows
> NT Alpha BIOS Firmware" on the motherboard. You can get the BIOS on either
> the SX or LX motherboard.
Also, the 21164PC doesn't have the 96KB L2 cache on-chip cache that the
21164 has, which makes for another reason for the lower performance of the
21164PC.
>
> Both of these are tons faster than a 300MHz Pentium II. Our latest
> DEC catalog says to look for the next generation of Alpha to use a 0.25
> micron CMOS process, with SPECint95 over 30, and SPECfp95 over 50, and
> 2GB/sec memory bandwidth. That's over 700% faster floating point than a
> 300MHz Pentium II!
>
> Can't wait for FreeBSD to support it...
>
> Kevin
>
Nadav
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.3.95-heb-2.07.970814102152.10665A-100000>