[Info-vax] Some VMS/x86 perf test data from WASD maintainer.

[Simon Clubley]

On 2023-04-27, Volker Halle <volker_halle at hotmail.com> wrote:
> Mark Daniel schrieb am Donnerstag, 27. April 2023 um 15:31:26 UTC+2:
>
>> |$ cpu_multiplier = 10 ! VAX = 10 - Alpha/AXP = 40 
>> |$ if f$getsyi("arch_name").eqs."x86_64" then $ cpu_multiplier=8 
>
> Mark,
>
> we seem to be using different versions of the 'VUPS.COM' procedure ;-(
> You seem to be using the one from FreeAXP, I'm using the one from Stromasys.
>
> And therefore we cannot compare the results !
>
> How about trying the PRIME_SIEVE.C  example.
>

I've mentioned this before (and got some negative feedback for it
as everyone is focused on the CPU, not I/O), but I think some tests
that measure I/O performance using RMS (which will force a range of
mode transitions including through executive mode) may be interesting.

To make it a fair comparison, the I/O could be to a RAM disk so that
VMS overheads only are being compared.

Simon.

-- 
Simon Clubley, clubley at remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.