Legacy Hardware and Apps
High sample rates

Hi Roger,

I was trying to do some measurements at 20ms Sample Rate using one VNC channel and the FAST modifier. In theory the Datataker should be able to cope with that, but what I find is that about each 20 measurements or so it skips 100ms of measurements.

I think this is probably because the dataTaker is busy with the communications since I am using the /R switch to send the results back to the computer. The full switch configuration is: /a/B/C/d/e/f/H/i/J/k/l/M/n/o/p/Q/R/S/t/u/V/w/x/y/Z

The JOB is basically:

RB1S,FAST 
1VNC
2VNC
3VNC 
RA20T:9~W,FAST 
2*VNC

Maybe the problem is the (:9~W statement?) Is it possible to make that the datataker does not skip any scans?

Sergi Udina
Universitat de Barcelona

Hi Roger, I was trying to do some measurements at 20ms Sample Rate using one VNC channel and the FAST modifier. In theory the Datataker should be able to cope with that, but what I find is that about each 20 measurements or so it skips 100ms of measurements. I think this is probably because the dataTaker is busy with the communications since I am using the /R switch to send the results back to the computer. The full switch configuration is: /a/B/C/d/e/f/H/i/J/k/l/M/n/o/p/Q/R/S/t/u/V/w/x/y/Z The JOB is basically: ```` RB1S,FAST 1VNC 2VNC 3VNC RA20T:9~W,FAST 2*VNC ```` Maybe the problem is the (:9~W statement?) Is it possible to make that the datataker does not skip any scans? Sergi Udina Universitat de Barcelona

Good morning Sergi,

Is it the dataTaker missing data point or windows not able to keep up with the data flow and dropping the data??

To check try logging the data to memory then unloading the data and comparing it with the real time data. If the same data is missing from both sets then it's the logger. If it is missing from the real time data then it's most likely to be Windows.

Our testing here show that using our USB to serial converter gives faster more reliable data transfer. This is because the converter is doing the flow control not Windows. Also close down any unused programs and processes giving as much CPU time to the data handling.

Cheers,
Roger

Good morning Sergi, Is it the dataTaker missing data point or windows not able to keep up with the data flow and dropping the data?? To check try logging the data to memory then unloading the data and comparing it with the real time data. If the same data is missing from both sets then it's the logger. If it is missing from the real time data then it's most likely to be Windows. Our testing here show that using our USB to serial converter gives faster more reliable data transfer. This is because the converter is doing the flow control not Windows. Also close down any unused programs and processes giving as much CPU time to the data handling. Cheers, Roger

The data is being logged internally into the logger flash memory card as well as being returned to the computer. Data returned in real time to the computer is only for preview purposes.

The missing scans I talk about are missing in the datalogger internal file, so I see small chance that it has something to do with windows unless the real-time data return using Ethernet is influencing directly (slowing?) the internal datalogging, but since i have the /y switch I understand that this shouldn't be happening, is that so?

Regards,
Sergi Udina
Universitat de Barcelona

The data is being logged internally into the logger flash memory card as well as being returned to the computer. Data returned in real time to the computer is only for preview purposes. The missing scans I talk about are missing in the datalogger internal file, so I see small chance that it has something to do with windows unless the real-time data return using Ethernet is influencing directly (slowing?) the internal datalogging, but since i have the /y switch I understand that this shouldn't be happening, is that so? Regards, Sergi Udina Universitat de Barcelona

Good afternoon Sergi,

When you say data is missing from the logged data, how much data is missing? How long is the test running for? and what else is the dataTaker doing?

There could be any number of reasons for small gaps in the data. For example, every so often the dataTaker will do some house keeping. While it is doing this house keeping you might miss small sections of data.
Also if the sampling very fast it might skip a scan because the time for that schedule has passed but it will pick up the next.

If you could send me a copy of your dataTaker code and a sample of the data I can look at it for you.

Cheers,
Roger

Good afternoon Sergi, When you say data is missing from the logged data, how much data is missing? How long is the test running for? and what else is the dataTaker doing? There could be any number of reasons for small gaps in the data. For example, every so often the dataTaker will do some house keeping. While it is doing this house keeping you might miss small sections of data. Also if the sampling very fast it might skip a scan because the time for that schedule has passed but it will pick up the next. If you could send me a copy of your dataTaker code and a sample of the data I can look at it for you. Cheers, Roger

Hi Roger,

Can you please provide an email address where I can send you the files you ask for?

Answering your questions, about 30% of the data is missing, housekeeping is deactivated as shown in the switch list I posted in the first email.

It might be also interesting for you to know that I use an algorithm to check if the datataker is able to cope with the sample rate which is based on the sample rates you give in page 49 of the DT800 manual, but much more conservative (I allow about 50% of the speed you report there) because the datataker has to attend heavy communications due to the /R. And normally this calculation works, but strangely not in this case (though the usual sample rate is 100ms and this one is 20ms).

Thanks for your help and regards,

Sergi Udina
Universitat de Barcelona

Hi Roger, Can you please provide an email address where I can send you the files you ask for? Answering your questions, about 30% of the data is missing, housekeeping is deactivated as shown in the switch list I posted in the first email. It might be also interesting for you to know that I use an algorithm to check if the datataker is able to cope with the sample rate which is based on the sample rates you give in page 49 of the DT800 manual, but much more conservative (I allow about 50% of the speed you report there) because the datataker has to attend heavy communications due to the /R. And normally this calculation works, but strangely not in this case (though the usual sample rate is 100ms and this one is 20ms). Thanks for your help and regards, Sergi Udina Universitat de Barcelona

Good morning Sergi,

My email address is support@datataker.com.au

I'm surprised that your speeds are 50% less than quoted. The real time data return only takes a few Hz of the top end speeds. What firmware version is you logger running?

Cheers,
Roger

Good morning Sergi, My email address is support@datataker.com.au I'm surprised that your speeds are 50% less than quoted. The real time data return only takes a few Hz of the top end speeds. What firmware version is you logger running? Cheers, Roger
22
5
2
live preview
enter atleast 10 characters
WARNING: You mentioned %MENTIONS%, but they cannot see this message and will not be notified
Saving...
Saved
With selected deselect posts show selected posts
All posts under this topic will be deleted ?
Pending draft ... Click to resume editing
Discard draft