Re^4: Time::HiRes sleep does not always work

If I run your script on my machine (bad) I get:

Counter changes 3579545 times/second
QPC: 35752003971
QPC: 35752004170
QPC: 35752004356
QPC: 35752004527
QPC: 35752004695
QPC: 35752004854
QPC: 35752005017
QPC: 35752005169
QPC: 35752005333
QPC: 35752005487
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
GToD: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
Time: 1219149309.90625000
[download]

And here is the output from a "good" machine:

Counter changes 3579545 times/second
QPC: 76031246430
QPC: 76031246657
QPC: 76031246790
QPC: 76031247127
QPC: 76031247273
QPC: 76031247479
QPC: 76031247686
QPC: 76031247838
QPC: 76031247972
QPC: 76031248165
GToD: 1219150762.25000000
GToD: 1219150762.25004410
GToD: 1219150762.25007610
GToD: 1219150762.25010900
GToD: 1219150762.25013900
GToD: 1219150762.25017500
GToD: 1219150762.25020500
GToD: 1219150762.25024100
GToD: 1219150762.25027490
GToD: 1219150762.25030590
Time: 1219150762.25033810
Time: 1219150762.25036690
Time: 1219150762.25039600
Time: 1219150762.25042800
Time: 1219150762.25047900
Time: 1219150762.25053190
Time: 1219150762.25058200
Time: 1219150762.25063110
Time: 1219150762.25066710
Time: 1219150762.25069900
[download]

Comment on Re^4: Time::HiRes sleep does not always work Select or Download Code

Replies are listed 'Best First'.
Re^5: Time::HiRes sleep does not always work by BrowserUk (Patriarch) on Aug 19, 2008 at 13:28 UTC
It's not your hardware! Your machine is equally capable of high resolution as the "good" machine. That's identified by the first line: `Counter changes 3579545 times/second` which displays the granularity of the timer, which is the same on both machines. That means if they are both using the same version of Time::HiRes, then they should be giving the same results. The first thing I would try is re-installing the module. From CPAN if you are compiler-capable. From a PPM otherwise. And, if your paranoid like me, you'll blow away the existing files first: `<yourperl>/lib/Time::HiRes.pm <yourperl>/lib/auto/Time/HiRes/*` [download] If the problem persists after that, something very hookey is going on with your install. Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice. "Too many [] have been sedated by an oppressive environment of political correctness and risk aversion."	[reply] [d/l] [select]
Re^6: Time::HiRes sleep does not always work by tone (Novice) on Aug 20, 2008 at 08:13 UTC
The "Good" machine that I tried above wasn't the original one that my testing was done on because that colleague wasn't at work yesterday but I got to run the test today and the results from that test are below Counter changes 14318180 times/second QPC: 4512760987142 QPC: 4512760987382 QPC: 4512760987508 QPC: 4512760987617 QPC: 4512760987726 QPC: 4512760987835 QPC: 4512760987939 QPC: 4512760988043 QPC: 4512760988148 QPC: 4512760988253 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 GToD: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 Time: 1219217113.05580000 [download] I'm going to try and reinstall my Time::HiRes anyway to to make sure the problem isn't there	[reply] [d/l]
Re^7: Time::HiRes sleep does not always work by BrowserUk (Patriarch) on Aug 20, 2008 at 08:53 UTC
Very strange. The hardware is capable of high resolution as evidenced by the `QPC:` output. And Time::HiRes has the code to utilise that capability, but from the evidence of the `gettimeofday()` & `time()` output, that code is not being exercised. It's like something is caching the value, or some set of build options was used when the code was compiled to cause it to bypass the code? Did you say this was an AS built Perl? Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice. "Too many [] have been sedated by an oppressive environment of political correctness and risk aversion."	[reply] [d/l] [select]
Re^8: Time::HiRes sleep does not always work by tone (Novice) on Aug 20, 2008 at 13:29 UTC