TimerInit & TimerDiff limits / max value

[Anne 0]

Hi everyone.

I have a question and wasn't able to find any answer for this.

What is the maximum value/limit for TimerInit & TimerDiff.

I mean, If I use TimerInit() in my script, how long can it 'last'? How long will I be able to read it with TimerDiff()?

I realize that the result of TimerDiff isn't an Integer. It's a floating number and I am not sure what the maximum value is.

I would like my program to use $Time = TimerInit() today, and then I'll be able to use TimerDiff($Time) tomorrow or even the day after. Is this possible? and what happens if TimerDiff exceeds the maximum value (if there is any)?

[water 2,117]

TimerDiff returns the elapsed time in milliseconds calculated from a previous call to TimerInit. I can't test at the moment but I'm sure the result is an integer.
Use IsInt or IsFloat to check if it is an integer or a floating number.

[BrewManNH 1,266]

If TimerInit uses the Kernel32.dll function QueryPerformanceCounter to calculate what values it gives, then it's probably about 68 years all the way up to 80,000 years according to this post I found. But a developer would need to answer this question as to how they obtain the values it uses.

[water 2,117]

 

I would like my program to use $Time = TimerInit() today, and then I'll be able to use TimerDiff($Time) tomorrow or even the day after. Is this possible? and what happens if TimerDiff exceeds the maximum value (if there is any)?

Just give it a try and you know if it works by tomorrow.

[Anne 0]

TimerDiff returns the elapsed time in milliseconds calculated from a previous call to TimerInit. I can't test at the moment but I'm sure the result is an integer.

Use IsInt or IsFloat to check if it is an integer or a floating number.

 

Well I know that it isn't an Integer. Because after 1 second has elapsed, the result of TimerDiff would be something like 1000.48243

Which is why I asked what the maximum value of a floating number would be.

 

If TimerInit uses the Kernel32.dll function QueryPerformanceCounter to calculate what values it gives, then it's probably about 68 years all the way up to 80,000 years according to this post I found. But a developer would need to answer this question as to how they obtain the values it uses.

 

Very good reply! Now, is it possible for a floating number to hold that value in milliseconds? I mean, that would be "2144448000000" milliseconds.

Just give it a try and you know if it works by tomorrow.

 

Yea I am already testing this, but I asked because what happens if I let it wait 2 days, 3 days, 1 week etc. It would take forever to get the answer by testing it.

[Melba23 3,063]

Anne,

TimerInit/Diff should generally only be used for measuring relatively small time periods. If you are looking to determine delays of "2 days, 3 days, 1 week etc" I would recommend using either DateDiff on suitable DTG strings or (if accuracy is not that necessary) the various date/time macros that you can find in this list.

M23

[Anne 0]

Anne,

TimerInit/Diff should generally only be used for measuring relatively small time periods. If you are looking to determine delays of "2 days, 3 days, 1 week etc" I would recommend using either DateDiff on suitable DTG strings or (if accuracy is not that necessary) the various date/time macros that you can find in this list.

M23

 

I had actually considered this as a solution but after giving it some work I realized that I couldn't make it accurate enough.

Say I want the script to perform a certain task 75½ hours after it has been executed.

Or exactly 24 hours (+/- 1 min) after execution.

In practise, I will need it to measure anything from 5 mins to a few days, which is why I'd prefer TimerInit/TimerDiff. However, it could be possible to use both solutions along each other.

I guess I will just have to try working with the Timer functions untill I (maybe) hit a natural wall and then move on to DateDiff for some of the timing instead.

[Melba23 3,063]

Anne,

 

perform a certain task 75½ hours after it has been executed. Or exactly 24 hours (+/- 1 min) after execution

The macros are perfectly suitable for that. Just poll them at a suitable interval - say 1 hr intervals for 23 hours, then 1 min intervals until 23:59, and finally 1 sec intervals until you reach the correct moment.

M23

[Anne 0]

Anne,

 

The macros are perfectly suitable for that. Just poll them at a suitable interval - say 1 hr intervals for 23 hours, then 1 min intervals until 23:59, and finally 1 sec intervals until you reach the correct moment.

M23

 

I must have been drunk since I did not think of this.

That's perfect thanks M23!

[Melba23 3,063]

Anne,

Glad I could help.

M23