11-08-2013 04:33 AM
Hi,
This may seem a very silly question but I've been stuck on this for the last 2 hours and I couldn't find an answer anywhere else on the forum, so here goes. The timestamp (dt) on my graph won't show. It only shows the integer numbers of samples and doesnt care about what my sampling frequency (dt) is. I've read somewhere that that could be because 'the time stamp is ignored' but that option is greyed out, so it's not applicable here.
I've added a picture of the code where the waveform is created (though it is pretty trivial). Any thoughts?
Solved! Go to Solution.
11-08-2013 04:41 AM - edited 11-08-2013 04:42 AM
Hi Adriaan,
dt isn't the "timestamp", it's the time interval between samples of your waveform. Timestamp is provided by "t0"...
Did you format the x axis to show "absolute timestamps"? Also there is a coercion dot at your graph terminal: where does it come from?
11-08-2013 04:49 AM - edited 11-08-2013 04:51 AM
I have no idea where the coercion dot comes from (is there a way to check this?). Even if i put the x axis in absolute/relative time, the samples still seem to come at a rate of 1 sample/s. (the problem lies in my dt which for some reason isn't being used by my graph).
11-08-2013 05:01 AM
11-08-2013 05:05 AM - edited 11-08-2013 05:08 AM
Hi,
I just noticed that the coercion dot comes from my data stream. Attaching a normal array to it makes the dot dissapear and then the graph does take my dt into account. So the problem must lie with my bitstream.
EDIT : For some reason, wiring the normal array to the graph, removing it and then rewiring my bitstream to the graph solved my problem... I have no idea what happend but now the sampling frequency is taken into account.... (the coercion dot moved to my convolution block, now indicating a problem with my filter coefficients, but it does seem to work now).