07-21-2015 10:07 AM
my channels sampling rate I want to be 1000Hz but with the dt of 1/100 and new data to be displayed every second on the graph
07-21-2015 11:07 AM
So you want to take your 1kHz sample, average it down to 100Hz (dt=1/100) and then update the graph every 1 second with the 100Hz data? The buffer should be fine with that for 5 minutes of data since that's 20x smaller and therefore about 1MB of data.
Cheers
--------, Unofficial Forum Rules and Guidelines ,--------
'--- >The shortest distance between two nodes is a straight wire> ---'
07-21-2015 12:23 PM
Yea but with your modification of the dt when I change my loop timing to 1000, it doesnt work.
Should I go back to how mine was or is there something else to do with yours since I didnt quite get what you did
07-21-2015 01:39 PM
What do you mean by "doesn't work"? I changed it to 1000 samples, 1000 ms timer, and 1000 decimation input and it seems to be working fine to me.
Cheers
--------, Unofficial Forum Rules and Guidelines ,--------
'--- >The shortest distance between two nodes is a straight wire> ---'
07-21-2015 02:06 PM
when you look at the timestamp on your graph is there a difference of 5 min from your t0 to tf. What Im saying now I get alot of data but on the timestamp on the graph is wrong. To get a 5 min buffer, it increases very slow and to would start 2-3 min less than the actual time clock
07-21-2015 02:07 PM
when you look at the timestamp on your graph is there a difference of 5 min from your t0 to tf. What Im saying now I get alot of data but on the timestamp on the graph is wrong. To get a 5 min buffer, it increases very slow and to would start 2-3 min less than the actual time clock
07-21-2015 02:12 PM
It looks like it's a problem with your consumer loop, doing something funky with the data before graphing. I assume it has something to do with you chopping down the sample rate, but writing the dt through as the same value. This would result in the same t0, but a different length of data (in the time domain).
Here's a screenshot of what I'm seeing. The graph I put in there on the left is correct, but your graph has the weird timestamp issue.
Cheers
--------, Unofficial Forum Rules and Guidelines ,--------
'--- >The shortest distance between two nodes is a straight wire> ---'
07-21-2015 02:39 PM
I was looking at it wrong. I was hoping the whole time to get 5 min buffer display with the decimated data plot in the consumer loop. To get that Id have to aim for a higher buffer size I see. thanks alot
07-21-2015 04:12 PM
Let me know if you need any further help. In the mean time, please mark one or a few of the posts here as the solution, so that people in the future can find a solution if they have a similar application. Most likely the solution is from 3 weeks ago as a solution to your original question.
Cheers
--------, Unofficial Forum Rules and Guidelines ,--------
'--- >The shortest distance between two nodes is a straight wire> ---'