Clock Cycle Question

Share icon

Mon, 01/26/2015 - 07:51 pm

Quick question, the details I've found in the manuals give several different answers to the same question, so hopefully someone here knows the answer.
What is the equivalent of a clock cycle in units of time?
So you are asking about 'ticks'? I don't know the details, but I think it depends on your system (approx. 60 ticks ~ 1s). If you need a more precise measurement of time work with startMStimer and stopMStimer. But this also works with your systems timing. What do you need for your application?

January 26, 2015 at 09:44 pm - Permalink

I'm afraid your question is a bit ambiguous. Could you provide some details? What problem are you trying to solve?

As Chozo says, Ticks are very nearly 60 per second. StartMSTimer and StopMSTimer gives results in microseconds, but the underlying timers likely don't have microsecond resolution. StopMSTimer(-1) returns the clock frequency, so that will tell you the resolution.

But perhaps you are asking about CPU cycles? That will depend on your processor. It isn't usually something that you really need to know or can use usefully, except as a rough guide to the speed of your computer.

John Weeks
WaveMetrics, Inc.
support@wavemetrics.com

January 27, 2015 at 09:41 am - Permalink

johnweeks wrote:
I'm afraid your question is a bit ambiguous. Could you provide some details? What problem are you trying to solve?

As Chozo says, Ticks are very nearly 60 per second. StartMSTimer and StopMSTimer gives results in microseconds, but the underlying timers likely don't have microsecond resolution. StopMSTimer(-1) returns the clock frequency, so that will tell you the resolution.

But perhaps you are asking about CPU cycles? That will depend on your processor. It isn't usually something that you really need to know or can use usefully, except as a rough guide to the speed of your computer.

John Weeks
WaveMetrics, Inc.
support@wavemetrics.com



Actually I meant what is the equivalent of the clock cycle for Igor Pro.
I'm looking at pulse shape discrimination using PID value (example in attached figure), but I can't find anything definitive regarding what the software identifies as a clock cycle. The only two comments I've found even remotely close to what I'm looking for are in the following:

eMorphoGUI.pdf (page 10 & 11)
"Set mode=1 if you want the data to be stored in three 16-bit words per event. A 16-bit energy, a 16-bit PIT sum and a 16-bit time stamp.
The energy is scaled as before. The PID sum is a sum over the beginning part of the pulse and is used for rise time discrimination, pulse shape analysis and particle identification in phoswich detectors.
Use AdvancedControls() to set the PID sum parameter (PartialIntegrationTime) and the corresponding scaling value.
The 16-bit time stamp has a resolution of 32 timing clock cycles (1.333us) and a range of 87.4 ms"

qMorphoFPGAmapV4.pdf (page 22)
"RT: Real Time register. Measures the time since last being cleared in units of 65536 ADC_CLK (ADC sampling clock) cycles. On a 24 MHz system the LSB unit is 2.731 ms and the range is 135.7 days."
PSD_example.png

January 27, 2015 at 11:34 am - Permalink

I may have found what I'm looking for, but I just want to be sure.
In using Igor Pro for pulse shape discrimination applications, 1 ADC clock cycle is equivalent to 12.5 nanoseconds?

January 27, 2015 at 02:01 pm - Permalink

It sounds to me like you've confused data acquisition with something internal to Igor. Your graph shows the horizontal axis labelled, "Time, ADC Sampling Clock Cycles". That is a pretty unusual way to measure time, but I presume that it means "units of the sampling clock". To me that would simply mean "data points"- most data acquisition hardware takes a sample every time the sampling clock fires. To figure out what that means in seconds you need to know how the data were acquired, which might require knowing something about the hardware that acquired the data. Or it might mean finding out what the ADC hardware settings were when that particular data set was acquired.

John Weeks
WaveMetrics, Inc.
support@wavemetrics.com

January 27, 2015 at 02:51 pm - Permalink

johnweeks wrote:
It sounds to me like you've confused data acquisition with something internal to Igor. Your graph shows the horizontal axis labelled, "Time, ADC Sampling Clock Cycles". That is a pretty unusual way to measure time, but I presume that it means "units of the sampling clock". To me that would simply mean "data points"- most data acquisition hardware takes a sample every time the sampling clock fires. To figure out what that means in seconds you need to know how the data were acquired, which might require knowing something about the hardware that acquired the data. Or it might mean finding out what the ADC hardware settings were when that particular data set was acquired.

John Weeks
WaveMetrics, Inc.
support@wavemetrics.com


Ok, I'll continue looking into it then.

January 27, 2015 at 04:43 pm - Permalink