21 Oct
2004
21 Oct
'04
8:03 a.m.
I seem to remeber there being a VUPS-measuring DCL script which I was
able to find a few years ago, which was probably mentioned on this list.
Does anyone have a copy they could send me, or point me to where I could
find it?
My attempts at using google and looking through past email have been
fruitless so far.
Thanks much,
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
21 Oct
21 Oct
8:08 a.m.
The one at
http://www.astro.multivax.de:8000/phillip/vup.com
Looks like the one I've used before.
Brian
On Thu, 2004-10-21 at 15:03, Patrick Finnegan wrote:
I seem to remeber there being a VUPS-measuring DCL
script which I was
able to find a few years ago, which was probably mentioned on this list.
Does anyone have a copy they could send me, or point me to where I could
find it?
My attempts at using google and looking through past email have been
fruitless so far.
Thanks much,
Pat
8:15 a.m.
I seem to remeber there being a VUPS-measuring DCL
script which I was
There was a script (produced by some 3rd party, I forget exactly
which party) that *claimed* to measure something that approximated
to VUPs.
My attempts at using google and looking through past
email have been
fruitless so far.
You could try dejanews, specifically in comp.os.vms - it's bound
to have floated through there at some stage.
Antonio
--
---------------
Antonio Carlini arcarlini(a)iee.org
8:16 a.m.
$! Newsgroups: comp.os.vms
$! Path:
news.rrz.uni-hamburg.de!cs.tu-berlin.de!fu-berlin.de!news.mathworks.com
!howland.erols.net!EU.net!main.Germany.EU.net!Dortmund.Germany.EU.net!ns.knirsch
.de!weller!ifwb!vax3.wupper.de!aw
$! From: aw(a)vax3.wupper.de (Andreas W Wylach)
$! Subject: Re: What dp VUPs really measure ?
$! Sender: news(a)ifwb.wupper.de (Net News Admin)
$! Message-ID: <E0o2yE.CA7(a)ifwb.wupper.de>
$! Date: Sun, 10 Nov 1996 18:20:38 GMT
$! Lines: 86
$! Reply-To: aw(a)vax3.wupper.de
$! References: <55vfv1$fje(a)news.NetVision.net.il>
<56006q$pt5$1(a)nntpd.lkg.dec.co
m> <5618vk$pl0(a)paperboy.ids.net>
$! Organization: CDrei - Wuppertal - Germany
$!
$!
$! In article <5618vk$pl0(a)paperboy.ids.net>et>,
mike(a)kronos.egr-ri.ids.net (Mike Um
bricht) writes:
$! |>In article <56006q$pt5$1(a)nntpd.lkg.dec.com>om>,
lionel(a)quark.zko.dec.com (Stev
e Lionel) writes:
$! |>>
$! |>>In article <55vfv1$fje(a)news.NetVision.net.il>il>, Uri Raz
$! |>><s2845543(a)t2.technion.ac.il> writes:
$! |>>|>
$! |>>|> After reading some articles trying to compare SPEC measures to VUP
$! |>>|> measures, I've started wondering what VUPs really measure.
$! |>>
$! |>>VUPs are similar in concept to the SPEC suite in that there is a specific
$! |>>collection of benchmark programs that are run and a performance range is
$! |>
$! |>Where can I get the code for the VUP benchmark programs? Do you have a
$! |>reference that details how VUPs are defined?
$! |>
$! |>>The VUP rating is mostly of historical interest now.
$! |>
$! |>I am with a group called the Retro-Computing Society of RI - a
user group f
or
$! |>individuals that collect older computer systems. I am
interested in runnin
g
$! |>the benchmark on our 750. (Yes, I know what the answer is, but
I'd like to
see
$! |>how it is calculated.)
$! |>
$! |>-mike
$! |>
$!
$! Here we go, the dcl code for VUPS :
$!
$! CALCULATE_VUPS:
$!
$ set noon
$ orig_privs = f$setprv("ALTPRI")
$ process_priority = f$getjpi(0,"PRIB")
$ cpu_multiplier = 10 ! VAX = 10 - Alpha/AXP = 40
$ cpu_round_add = 1 ! VAX = 1 - Alpha/AXP = 9
$ cpu_round_divide = cpu_round_add + 1
$ init_counter = cpu_multiplier * 525
$ init_loop_maximum = 205
$ start_cputime = f$getjpi(0,"CPUTIM")
$ loop_index = 0
$ 10$:
$ loop_index = loop_index + 1
$ if loop_index .ne. init_loop_maximum then goto 10$
$ end_cputime = f$getjpi(0,"CPUTIM")
$ init_vups = ((init_counter / (end_cputime - start_cputime) + -
cpu_round_add) / cpu_round_divide) * cpu_round_divide
$ loop_maximum = (init_vups * init_loop_maximum) / 10
$ base_counter = (init_counter * init_vups) / 10
$ vups = 0
$ times_through_loop = 0
$ 20$:
$ start_cputime = f$getjpi(0,"CPUTIM")
$ loop_index = 0
$ 30$:
$ loop_index = loop_index + 1
$ if loop_index .ne. loop_maximum then goto 30$
$ end_cputime = f$getjpi(0,"CPUTIM")
$ new_vups = ((base_counter / (end_cputime - start_cputime) + -
cpu_round_add) / cpu_round_divide) * cpu_round_divide
$ if new_vups .eq. vups then goto 40$
$ vups = new_vups
$ times_through_loop = times_through_loop + 1
$ if times_through_loop .le. 5 then goto 20$
$ 40$:
$ new_privs = f$setprv(orig_privs)
$ set message /nofacility/noidentification/noseverity/notext
$ ASSIGN/SYSTEM/EXEC 'vups' MACHINE_VUPS_RATING
$ set message /facility/identification/severity/text
$ write sys$output "Approximate System VUPs Rating : ", -
vups / 10,".", vups - ((vups / 10) * 10)
$ exit
$!
$! Have fun.
$!
$! see ya,
$! Andreasw
$! --
$! IfwBOnline Wuppertal, Maerkische Str. 28, D-42281 Wuppertal,
F.R.Germa
ny
$! Andreas W. Wylach, | AW(a)IFWB.WUPPER.DE
$! AW(a)VAX1.wupper.de | AW(a)VAX3.wupper.de | Local DECNet: VAX3::AW
$! Information system from students for students, WupperNet
$! @CDrei Wuppertal (FRG) /Miami (USA) - Competence, Consulting, Connectivity
$! --
$! This is comp.os.vms, not alt.read.my.mind (Carl J. Lydick)
$! --
$! This is WBIG New York, listen yo', there are a lot of dogs out
there. (NOT. B
IG)
--
John A. Dundas III
Director, Information Technology Services Infrastructure, Caltech
Mail Code: 014-81, Pasadena, CA 91125-8100
Phone: 626.395.3392 FAX: 626.449.6973
8:32 a.m.
On Thu, 21 Oct 2004, John A. Dundas III wrote:
Subject: Re: VUPS measuring utility
For pointless hilarity, how about someone with pentium-class
assembly knowledge coding up the Gill Test?
"The Gill Test
A measure of the arithmetic speed of an automatic
computer. It is defined as the time in milliseconds required
for a computer to:
A + B = C {Store C}
C + D + E = F {Store F}
G x H = K {Store K}
when all access is to the fastest internal storage used in
the computer in question. For a one-address computer only
with one constant storage access storage device {such as
magnetic cores}, a gill is equal to ten times the average
operation time.
{Named after Stanley Gill of Cambridge University}
"
from
http://www.users.nwark.com/~rcmahq/jclark/gill.htm
26 Oct
26 Oct
8:54 p.m.
New subject: RS6000 diagnostic tapes and loopbacks
I missed the RS6000, but did snag the diagnostic kit that was inside the
cabinet. Its two tapes, one for the 8 mm drive, and one for the RS6000, and
three loopback thingies (two plugs and short cable with weird rectangular
connectors on each end). Email me if its something you want.
7361
days inactive
7367
days old
5 comments
6 participants
participants (6)
-
a.carlini＠ntlworld.com -
bdwheele＠indiana.edu -
dundas＠caltech.edu -
mikeford＠socal.rr.com -
pat＠computer-refuge.org -
tomj＠wps.com