sv@v7fs1.UUCP (Steve Verity) (08/17/89)
Anyone out there know just what the heck Norton's SI measures? The utility tells us that it is measuring performance relitive to the PC. Still, what does *that* mean? _Steve ...!ames!vsi1!v7fs1!sv sv@v7fs1.com Quote ----->""
slimer@trsvax.UUCP (08/17/89)
I believe that this means that the machine is running at a multiple of the speed of the base for comparison. That is to say... If an AT index is 2.0, then it is twice as fast as the PC in the overall performance. This is the strategy used by many utilities that measure performance, and they all agree within reason. So I have always used the SI program with my interpretation of its meaning. **************************************************************************** * Thank You, texbell!letni!rwsys!trsvax!slimer * * Bill "Disk BUG'S for you!" - ComputerWorld * * George W. Pogue, 1300 Two Tandy, Fort Worth, TX. 76102 (817) 390-3562 * ****************************************************************************
Ralf.Brown@B.GP.CS.CMU.EDU (08/17/89)
In article <478@v7fs1.UUCP>, sv@v7fs1.UUCP (Steve Verity) wrote: } Anyone out there know just what the heck Norton's SI }measures? The utility tells us that it is measuring performance }relitive to the PC. Still, what does *that* mean? SI basically measures the performance of the multiply instruction.... The numbers have meaning for comparison only when comparing systems with identical processors, as different CPU types take different numbers of clock cycles for multiplication, and those different clock cycles are not representative of overall performance. SI overstates V20/V30 performance by about 60%, and 80286 performance by about 100% (i.e. a 286 with a 10.1 rating is really about five times as fast as a 4.77 8088 overall). -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=-=-=-=- Voice: (412) 268-3053 (school) ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/46 FAX: available on request Disclaimer? I claimed something? "Drama is life with the dull bits left out." -- Alfred Hitchcock
asrap@warwick.ac.uk (Sean Legassick) (08/18/89)
In article <478@v7fs1.UUCP> sv@v7fs1.UUCP (Steve Verity) writes: > Anyone out there know just what the heck Norton's SI >measures? The utility tells us that it is measuring performance >relitive to the PC. Still, what does *that* mean? > Norton SI is supposed to give you the peformance of your machine relative to a plain vanilla IBM PC - don't see many of them now! - as a figure, i.e. if SI returns a value of 2.0, then your machine should be twice as fast as an IBM PC. It does this by running benchmarks, comparing the time taken to peform certain mundane activities with the time it takes a PC. In this age of VGA displays and frequent hard-disk accesses, however, this figure bears little relation to actual real-time usage, and so is of little use. --- SSSSSSS K K L Sean K Legassick S K K L Coventry SSSSSSS KK L England S K K L "I'm not expendible, I'm not stupid SSSSSSS K K LLLLLLL and I'm not going!" Avon (Blakes 7)
stephen@ziebmef.mef.org (Stephen M. Dunn) (08/26/89)
In article <224@orchid.warwick.ac.uk> asrap@warwick.ac.uk (Sean Legassick) writes:
$Norton SI is supposed to give you the peformance of your machine
$relative to a plain vanilla IBM PC - don't see many of them now! -
$as a figure, i.e. if SI returns a value of 2.0, then your machine
$should be twice as fast as an IBM PC.
$It does this by running benchmarks, comparing the time taken to peform
$certain mundane activities with the time it takes a PC. In this age
$of VGA displays and frequent hard-disk accesses, however, this
$figure bears little relation to actual real-time usage, and so is
$of little use.
As far as the display goes, fine, but the more recent SI versions _do_
test your hard disk. They come up with three numbers:
1. Raw processing power relative to the XT
2. Some measure of hard disk speed relative to the XT's 10 Mb drive (I think)
3. Some sort of combination of the two
Of course, there have been many complaints about the fact that Norton's
SI seems skewed in favour of 286s (and higher) due to a (perhaps) inappropriate
choice of benchmarking algorithm. I don't know if any adjustments have been
made or not.
And, of course, any benchmark will only be an approximation based on what
the author thinks a typical user's usage patterns will be. If I use my
hard drive more than you do on an identical machine, we'll get different
performance levels even though SI won't find any difference because it uses
some assumed weighting in computing the overall performance score.
--
-------------------------------------------------------------------------------
! Stephen M. Dunn stephen@ziebmef.UUCP ! DISCLAIMER: Who'd ever !
!---------------------------------------------------! claim such dumb ideas? !
! I have become comfortably numb ... ! I sure as heck wouldn't !neighorn@nosun.UUCP ( SE Sun/PDX) (09/01/89)
In article <1989Aug25.162653.25755@ziebmef.mef.org> stephen@ziebmef.mef.org (Stephen M. Dunn) writes: > Of course, there have been many complaints about the fact that Norton's >SI seems skewed in favour of 286s (and higher) due to a (perhaps) inappropriate >choice of benchmarking algorithm. I don't know if any adjustments have been >made or not. "inappropriate" is an understatement. I shudder to think how many computer salesmen and advertisements and jokers who don't know what they are talking about quote that good old Norton SI number. It is just less than completely useless! The compute performance section of SI appears to check the multiply instruction in a loop. Now, how many programs do you know of that do nothing but multiply integers? For such programs, the SI number is accurate, but for anything else??? Take a 16 MHz, 1 wait state 80386 machine. It will get a SI rating of around 17. Now go and write your own little loop/mul program, and lo and behold, it runs approx 17 times faster as well. The oft-flamed Dhrystone benchmark is a magnitude better than this. Mis-information is worse than no information at all, because people who want answers will stop looking for them when they get *something*. > And, of course, any benchmark will only be an approximation based on what >the author thinks a typical user's usage patterns will be. If I use my >hard drive more than you do on an identical machine, we'll get different >performance levels even though SI won't find any difference because it uses >some assumed weighting in computing the overall performance score. And then there are really many "types" of hard disk use. Reading/Writing different size blocks, sequential/random, etc. Most of the cheap and dirty disk benchmarks only scratch the surface of what is possible to test. Of course the only real benchmark is the application(s) you really want to run! :-) -- Steven C. Neighorn !tektronix!{psu-cs,nosun,ogccse}!qiclab!neighorn Sun Microsystems, Inc. "Where we DESIGN the Star Fighters that defend the 9900 SW Greenburg Road #240 frontier against Xur and the Ko-dan Armada" Portland, Oregon 97223 work: (503) 684-9001 / home: (503) 641-3469
rogers@SRC.Honeywell.COM (Brynn Rogers) (09/01/89)
Okay, the norton SI is bad. How good is the landmark AT test (expressed in Mhz) that I see more and more advertisers quoting? BTW, I did a rough calculation (guess) and found that a 50Mhz 486 would rate between 200 and 330 on the Norton SI. (just for fun!) Brynn Rogers Honeywell S&RC rogers@src.honeywell.com work 612-782-7577 home 874-7737
pfratar@watdcsu.waterloo.edu (Paul Frattaroli - Department of Computing Services) (09/02/89)
In article <408@nosun.UUCP> neighorn@nosun.UUCP (Steven C. Neighorn - SE Sun/PDX) writes: >In article <1989Aug25.162653.25755@ziebmef.mef.org> stephen@ziebmef.mef.org (Stephen M. Dunn) writes: >> Of course, there have been many complaints about the fact that Norton's >>SI seems skewed in favour of 286s (and higher) due to a (perhaps) inappropriate >>choice of benchmarking algorithm. I don't know if any adjustments have been >>made or not. > >"inappropriate" is an understatement. I shudder to think how many >computer salesmen and advertisements and jokers who don't know what they >are talking about quote that good old Norton SI number. > >It is just less than completely useless! [ stuff deleted ] Tell me about it! I have a PS/2 mod 25, I get a rating of 1.9 unless I move my mouse around. When I move my mouse around while it is testing I get 1.6. I just don't think that a true benchmark should depend on what you also have running at the current time. Perhaps some TSR's could change the rating still further. I think a true benchmark comparison should take into account everything your system can do at once not just... how fast can it run when these other things are running? For example, how much faster would my system be if it didn't have to keep track of a clock? no interrupts while testing? 100% hit rate caching? etc. Just a thought. ....Paul F -- -------------------------------------------------------------------------- Paul Frattaroli - Department of Computing Services University of Waterloo < pfratar@watshine.waterloo.edu > < pfratar@watdcsu.waterloo.edu >