PicoScope 7 Software
Available on Windows, Mac and Linux
random 0xFED1 (-303) spikes in streaming data and other fun things...
random 0xFED1 (-303) spikes in streaming data and other fun things...
The saga continues! It might be something being done wrong on my end once more, but I get weird spikes, typically offsets by 0xFED1 (-0x12F or 303 decimal) more or less often in my streaming data through the SDK. I cannot replicate this in the Picoscope software, but I am also unsure when it is streaming or not. Any tips on what to do about this?
It seems like it's not a simple issue, as using aggregation, a large amount of noise is overlayed, too, presumably because of these spikes. Now this only happens when streaming 1 or 2 channels, streaming 3 or 4 channels seems to make it go away.
Any ideas what could be going on? For now, I can just stream 4 channels always, but it seems like that'd preclude me from getting the maximum samples/sec on a single channel.
That being said, it also seems like there is quite a significant delay (several 10s of seconds, as time goes on) in the data stream, which I cannot account for, either. (Sampling at 1MS/s on all 4 channels over USB3 on a 3000D series scope)
Has anyone seem similar issues? It could be mistakes on my application's end, but if they are, I can't see them, for now.
It seems like it's not a simple issue, as using aggregation, a large amount of noise is overlayed, too, presumably because of these spikes. Now this only happens when streaming 1 or 2 channels, streaming 3 or 4 channels seems to make it go away.
Any ideas what could be going on? For now, I can just stream 4 channels always, but it seems like that'd preclude me from getting the maximum samples/sec on a single channel.
That being said, it also seems like there is quite a significant delay (several 10s of seconds, as time goes on) in the data stream, which I cannot account for, either. (Sampling at 1MS/s on all 4 channels over USB3 on a 3000D series scope)
Has anyone seem similar issues? It could be mistakes on my application's end, but if they are, I can't see them, for now.
Re: random 0xFED1 (-303) spikes in streaming data and other fun things...
A little further investigation shows that going to from 1Msps to 100ksps significantly improves the data streaming. Of course, this isn't great if we want high temporal resolution data, but at least it's a clue. Moving the scope API calls off the main thread might help, in case there is some contention there.