[net.micro.pc] Norton Utilities 'si' program

john@moncol.UUCP (John Ruschmeyer) (09/25/85)

Does anyone know what heuristic is used by the Norton Utilities' 'si'
program to determine the performance ratio vs. the IBM-PC?

After installing a V20 in my Tandy 1000, the figure jumped from 1.0 to 1.8
(an AT&T 6300 with an 8mhz 8086 only rates 1.9). Meanwhile most other
benchmarks showed only about a 10% increase in speed. Odd!

Thanks in advance.


-- 
Name:		John Ruschmeyer
US Mail:	Monmouth College, W. Long Branch, NJ 07764
Phone:		(201) 222-6600 x366
UUCP:		...!vax135!petsd!moncol!john	...!princeton!moncol!john
						   ...!pesnta!moncol!john
Disclaimer:
	Monmouth College is a mecca for diverse opinions. It is, therefore,
	highly unlikely that the above opinions are those of anyone but me.

	"Are we gonna be starcrossed lovers or just good friends?"

bill@hp-pcd.UUCP (bill) (09/27/85)

I don't know how much I believe that performance index.

I've got a PC-AT with a 15 Mhz crystal (i.e., system running at 7.5 Mhz
instead of normal 6 Mhz) for which Norton gives a performance index
of 7.1 relative to a normal IBM/PC.  This AT may be fast, but I have
a hard time believing it's *that* fast!

bill frolik
hp-pcd!bill

faisal@smu (09/30/85)

Norton's "si"

x  = the amount of time Norton's IBM PC took to perform the loop below;

t1 = current time;
  do a bunch of times
   {
     mul
     div
   }
t2 = current time;

index = x / (t2 - t1);

I disassembled his code a few months ago; I don't remember the gritty
details, but this is pretty close.


Ralph Cramden tought Norton everything he knows.

Faisal , CS dept. SMU, Dallas, TX