bayes@hpislx.HP.COM (Scott Bayes) (01/17/90)
> I would reply via Email, but this is a VERY bad comparison. The idea > of the mythical "MIP" is an idead propagated by computer sales(wo)men. > A MIP is ONLY useful when comparing proccessors with the same instruction > set. For example, in my Mac, I have a 16mhz 68020/68881 which can put Actually a "MIP" is useless for almost any purpose, meaning, as it does, "Machine Instruction Per". The singular term (e.g. 1 MIPS) is "MIPS": "Machine Instruction Per Second". This is not to say that the plural is "MIPSs" :-) [...] > Mike Schechter > Institute for Sensory Research > isr@rodan.acs.syr.edu Scott Bayes Hewlett-Packard Company
bmartin@uhccux.uhcc.hawaii.edu (Brian Martin) (01/18/90)
a digression...
I recall reading somewhere (probably some IBM sales literature for
one of their "mainframes") that the user's perception of a machine's
speed is directly dependent on the time it takes a message to
appear on screen after the user presses a key. They made the point
that on a heavily loaded processor, printing a "processing your request"
message within 1/2 second of the user pressing a key would fool
the user into thinking that the machine was not loaded down. That's
one of the reasons they put I/O processors on their "multi-channel" machines.
Here's the point: my Apollo, which is an old 12MHz 68020, feels a whole
lot faster than the Mac IIci running Multifinder, even though the Apollo
runs various benchmarks much slower than the mac. With a backup across the
network to a tape drive running in one window, a compile running in background
in a another window, and an editor in the foreground, the machine
still gives me the perception of responsiveness. With a couple of
programs loaded under Multifinder on the mac (word, spelling coach, wingz,
and statview) the mac feels real sluggish, and tends to crash at very
inopportune moments.
-- Brian
====
Brian K. Martin, M.D.
Assistant Research Professor (hat #1)
University of Hawaii/ Cancer Research Center of Hawaii,
and
CEO (hat #2)
Martin Information Systems, Ltd.
1103 9th Avenue, Suite 203
Honolulu, Hawai`i 96816-2403
Voice (808) 733-2003
Fax (808) 733-2011
ARPA: uhccux!bmartin@nosc.MIL
UUCP: {uunet,dcdwest,ucbvax}!ucsd!nosc!uhccux!bmartin
INTERNET: bmartin@uhccux.uhcc.hawaii.educonsp22@bingsune.cc.binghamton.edu (Darren Handler) (01/20/90)
-> Actually a "MIP" is useless for almost any purpose, meaning, as it does, -> "Machine Instruction Per". The singular term (e.g. 1 MIPS) is "MIPS": -> "Machine Instruction Per Second". This is not to say that the plural is -> "MIPSs" :-) I think you made a mistake. It stands for Million Instructions per Second. If my computer was rated a 6 Machine Instructions per Second it would be running as slow as Peanut Butter flowing uphill in February. ------------------------------------------------------------------------------- | Consp22@Bingsuns.pod.binghamton.edu | SUNY-B Computer Consultants - | | Consp22@Bingvaxu.cc.binghamton.edu | Trying to keep the world safe from | |---------------------------------------| the SUNY-B Computer users. | | System Consultant - World Computers |-------------------------------------| | Computer Cons. - SUNY Binghamton | Darren `Mac Hack' Handler | |-----------------------------------------------------------------------------| I don't know if I am going to heaven or hell, I just hope God grades on a curve
datta@cpsc.ucalgary.ca (Slarti) (01/21/90)
In article <2600001@hpislx.HP.COM> bayes@hpislx.HP.COM (Scott Bayes) writes: >Actually a "MIP" is useless for almost any purpose, meaning, as it does, >"Machine Instruction Per". The singular term (e.g. 1 MIPS) is "MIPS": >"Machine Instruction Per Second". This is not to say that the plural is >"MIPSs" :-) > Actually, the term MIPS stands for "Million Instructions Per Second". A 6 MIPS machine would be useless if it executed six instructions per second. That's where the problems arise. A 68030 instruction is worth more than a SPARC instruction because it can do more. So, just going by MIPS is not a valid comparison unless you are comparing two computers with the same CPUs. The only real way to compare the speeds of machines is to do benchmarks.