Re: Clock rate on SGI's

New Message	Reply	Date view	Thread view	Subject view	Author view

John W. Barrus (barrus++at++merl.com)
Wed, 3 Jul 96 14:39:01 EDT

Next message: Todd R Pravata: "Performer + Purify + DSO problem"
Previous message: Ray Twiddy: "textured polygon in HUD overlay"

When writing code to test the actual clock rate, we had a funny experience.

The output at the end of this mail message contains a series of runs of the
"clocktest" program at different times and at different accuracies (which
decides how long the program should run). The number after 'clocktest' tells
the program how accurate the answer needs to be.

We found the actual calculated clock rate to be 40,000 sometimes and 41,600
at other times. This was on a 250Mhz R4400 Extreme. Our Elan changes between
39999 and 39995 for minutes at a time also. Same program, just run at a
different time. Of course, I am less concerned about a 0.01 percent change
on the Elan than the huge 4 percent variation on the Extreme.

What could be the problem here?

John B.

(Note: at one point I changed and recompiled the program, but it didn't make
a difference in the output)
==========

barrus++at++extreme>clocktest
Given: 32000 Calculated: 39999
setenv SGI_CYCLE_COUNT 39999

barrus++at++extreme>clocktest <--- instantaneous dramatic change
Given: 32000 Calculated: 41620
setenv SGI_CYCLE_COUNT 41620

barrus++at++extreme>clocktest 0.0001
Given: 32000 Calculated: 41600
setenv SGI_CYCLE_COUNT 41600

barrus++at++extreme>clocktest
Given: 32000 Calculated: 41613
setenv SGI_CYCLE_COUNT 41613

barrus++at++extreme>clocktest 0.00004
Given: 32000 Calculated: 41599
setenv SGI_CYCLE_COUNT 41599

barrus++at++extreme>clocktest 0.00004
Given: 32000 Calculated: 41600
setenv SGI_CYCLE_COUNT 41600

barrus++at++extreme>clocktest
Given: 32000 Calculated: 41613
setenv SGI_CYCLE_COUNT 41613

barrus++at++extreme>clocktest 1.0 <-- asked for 1% accuracy
Given: 32000 Calculated: 35410
setenv SGI_CYCLE_COUNT 35410

barrus++at++extreme>clocktest 0.1
Given: 32000 Calculated: 43155
setenv SGI_CYCLE_COUNT 43155

barrus++at++extreme>clocktest 0.01
Given: 32000 Calculated: 41663
setenv SGI_CYCLE_COUNT 41663

barrus++at++extreme>clocktest 0.001
Given: 32000 Calculated: 41606
setenv SGI_CYCLE_COUNT 41606

barrus++at++extreme>clocktest 0.0001
Given: 32000 Calculated: 41600
setenv SGI_CYCLE_COUNT 41600

barrus++at++extreme>clocktest 0.00004
Given: 32000 Calculated: 41600
setenv SGI_CYCLE_COUNT 41600

barrus++at++extreme>clocktest 0.001 <-- instantaneous dramatic change
Given: 32000 Calculated: 39999
setenv SGI_CYCLE_COUNT 39999

barrus++at++extreme>clocktest 0.00004
Given: 32000 Calculated: 39999
setenv SGI_CYCLE_COUNT 39999

barrus++at++extreme>clocktest 0.001
Given: 32000 Calculated: 39999
setenv SGI_CYCLE_COUNT 39999

barrus++at++extreme>clocktest 0.001
Picosecs/tick given: 32000 Calculated: 40032
setenv SGI_CYCLE_COUNT 40032

barrus++at++extreme>clocktest 0.00004
Picosecs/tick given: 32000 Calculated: 40025
setenv SGI_CYCLE_COUNT 40025

barrus++at++extreme>clocktest 0.00004
Picosecs/tick given: 32000 Calculated: 40000
setenv SGI_CYCLE_COUNT 40000

=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/ <--new!
Submissions: info-performer++at++sgi.com
Admin. requests: info-performer-request++at++sgi.com

Next message: Todd R Pravata: "Performer + Purify + DSO problem"
Previous message: Ray Twiddy: "textured polygon in HUD overlay"

New Message	Reply	Date view	Thread view	Subject view	Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:53:09 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.