From: Randal Whittle (rwhittle_at_usa.net)
Date: Thu Aug 05 1999 - 16:09:21 EDT
At 09:43 AM 8/5/99 -1000, you wrote:
>There *is* a possibility that PEP and Maximum Upgrades are somewhat
>different.
Perhaps. But I'd like to see these claims of a multiplier of 2 or 3 being
the speed performance increase--and I mean overall, not just a CPU-only
benchmark. Those don't count.
In solo vacuum-style tests, PowerPC chips blew the doors off of Intel
chips in every way, but the reality was much more modest. The reality was
once you encumbered that PowerPC chip with the Mac OS, it was really no
better than the Intel chip it competed with.
I suspect even if Maximum Upgrades claims of CPU speed being a multiple of
2 or 3 over the original are true, the reality would be much like the
above--when put in the real world, that multiple of speed is nowhere near true.
>clock speed on one of the slower (DX2-50) 701s from 25mhz to 33mhz as part
>of the upgrade,
>realizing in the process an even greater speed increase than you noticed in
>the upgrade
>of your already-33mhz DX2-75 (*especially* on system-level tasks).
>Anyway, I wouldn't be at all surprised to see a 100% speedup on a DX2-50 to
>DX4-133 upgrade
>with proper chip tweaks, even for real world stuff. But, that DX2-50 was
>pretty slow to begin with!
I did specify the DX4-75 as being the platform for comparison, not the
DX2-50. If apples are to be compared to apples, that's what will be
necessary. The DX4-75 is what I had and tested. It stands to reason that
if the PEP upgrade was 20 to 25% better over a stock DX4-75, that the gain
>from a DX2-50 would be appreciably larger than that.
>When I upgraded one of our desktops from DX2-66 to AMD5x133 I saw
>substantial improvement, even
>on real-world apps.
"Substantial improvement". That's good. Did you do any testing to
compare before & after? Something a little more quantitative rather than
qualitative. Anyone can try it and say "hey, this is quicker!" but that
doesn't tell us much. I ended up selling my 701C to someone on this list
roughly 1 year ago and he noted it was quicker than a DX4-75 he already
had, but not by a whole lot.
To me, 2 to 3 times quicker would *be* a "whole lot". 25% quicker isn't.
- Randy Whittle
This archive was generated by hypermail 2.1.3 : Thu Jan 23 2003 - 09:55:13 EST