PicoScope 7 Software
Available on Windows, Mac and Linux
Trace Disappears When Zoomed
Trace Disappears When Zoomed
I frequently use my 5244B to capture large data sets and then zoom in on the area I want to explore. I also save the data as well. I am currently running version 6.14.10 software, but I believe this bug has been around for a while. From running multiple tests and closing and restarting the program I think I have narrowed down the issue, but I'm not entirely sure.
The problem is that when zooming into a stopped waveform, or even a saved waveform, the trace on the display disappears partially or completely. For instance with a capture that I have saved at 1s/div, 992.1kS/s with the horizontal zoom set at x500 I can see about 1/3 of the trace, set at x1k I can see none of the trace.
As far as I can tell, this only happens if I have "Lowpass Filtering" activated when the waveform is captured. If I save this file then it will show this problem even if I deactivate Lowpass Filtering, however if the file is the active sample on the scope then deactivating Lowpass Filtering allows me to see the waveform again. It might make sense if activating Lowpass Filtering caused the filtered waveform to be displayed improperly, but it causes the unfiltered waveform on a 2Ch setup to have the same issue.
There seems to be some interaction with the selected number of samples, as well as the time/div and the zoom factor, but the real culprit seems to be capturing the waveform with the "Lowpass Filtering" active.
If need be I can provide a saved waveform that produces this undesired behavior. The problem is that the files tend to be quite large.
The problem is that when zooming into a stopped waveform, or even a saved waveform, the trace on the display disappears partially or completely. For instance with a capture that I have saved at 1s/div, 992.1kS/s with the horizontal zoom set at x500 I can see about 1/3 of the trace, set at x1k I can see none of the trace.
As far as I can tell, this only happens if I have "Lowpass Filtering" activated when the waveform is captured. If I save this file then it will show this problem even if I deactivate Lowpass Filtering, however if the file is the active sample on the scope then deactivating Lowpass Filtering allows me to see the waveform again. It might make sense if activating Lowpass Filtering caused the filtered waveform to be displayed improperly, but it causes the unfiltered waveform on a 2Ch setup to have the same issue.
There seems to be some interaction with the selected number of samples, as well as the time/div and the zoom factor, but the real culprit seems to be capturing the waveform with the "Lowpass Filtering" active.
If need be I can provide a saved waveform that produces this undesired behavior. The problem is that the files tend to be quite large.
Re: Trace Disappears When Zoomed
Hi BCG,
The zoom function can be limited by the amount of data available to perform the plotting. If you use the fastest possible sample rate of 500MS/s (for 2 active channels), on a Timebase of 2ns/div, you will have only 10 samples captured (see below data file). For this set up the displayed zoom disappears between a zoom level of x1K and x2k, which is less than 100th of the sample interval, and obviously way more than you will ever need). So the zoom function is directly related to the number of samples used for the capture.
The zoom function also has a hard maximum limit, where a large number of samples are being displayed. If you use a 1s/div Timebase (giving you a 1us sample rate for 2 active channels) with 10MS requested you can zoom in to x100M (the absolute maximum) giving you 1/10th of the sample interval (see below data file).
The Lowpass Filtering in PicoScope 6 is an algorithm that results in a significant reduction in the number of samples displayed. So, what can happen, is that if channels are filtered, resulting in not enough samples to display for the zoom level (as in the first example I gave you), the data will disappear from the display. However, even if there will be enough data to display (as you correctly pointed out) the data may not be displayed due to what appears to be an error in the algorithm (e.g. perhaps because the filtering is applied after capture but the zoom request, for some reason, can't check how much data will be returned, so it just applies a predefined limit of data for any filtering). (Note that all of the data disappears because there is no point in just displaying some of it). However, the workaround for this is as you mentioned, i.e. capture the data first and then apply the filter after it has been captured. That way you can, get the all of the data in your example displayed, and you can save the file and reload it to still get the correct behaviour.
So thanks for pointing out the error, and I will log it. However, just so that you are aware, we have announced the up and coming new major release of our software, and our focus is on getting this out to our customers, so we are only going to devote resources to fixing critical errors in our current PicoScope 6 software. As this error has an easy workaround, it is not likely to be a high enough priority for fixing.
Regards,
Gerry
The zoom function can be limited by the amount of data available to perform the plotting. If you use the fastest possible sample rate of 500MS/s (for 2 active channels), on a Timebase of 2ns/div, you will have only 10 samples captured (see below data file). For this set up the displayed zoom disappears between a zoom level of x1K and x2k, which is less than 100th of the sample interval, and obviously way more than you will ever need). So the zoom function is directly related to the number of samples used for the capture.
The zoom function also has a hard maximum limit, where a large number of samples are being displayed. If you use a 1s/div Timebase (giving you a 1us sample rate for 2 active channels) with 10MS requested you can zoom in to x100M (the absolute maximum) giving you 1/10th of the sample interval (see below data file).
The Lowpass Filtering in PicoScope 6 is an algorithm that results in a significant reduction in the number of samples displayed. So, what can happen, is that if channels are filtered, resulting in not enough samples to display for the zoom level (as in the first example I gave you), the data will disappear from the display. However, even if there will be enough data to display (as you correctly pointed out) the data may not be displayed due to what appears to be an error in the algorithm (e.g. perhaps because the filtering is applied after capture but the zoom request, for some reason, can't check how much data will be returned, so it just applies a predefined limit of data for any filtering). (Note that all of the data disappears because there is no point in just displaying some of it). However, the workaround for this is as you mentioned, i.e. capture the data first and then apply the filter after it has been captured. That way you can, get the all of the data in your example displayed, and you can save the file and reload it to still get the correct behaviour.
So thanks for pointing out the error, and I will log it. However, just so that you are aware, we have announced the up and coming new major release of our software, and our focus is on getting this out to our customers, so we are only going to devote resources to fixing critical errors in our current PicoScope 6 software. As this error has an easy workaround, it is not likely to be a high enough priority for fixing.
Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist