Duration and resolution

Duration and resolution, a forum discussion on Cleverscope Mixed Signal USB Oscilloscopes. Join us for more discussions on Duration and resolution on our Questions forum.

Back to Forum Index : Back to Questions   RSS

14 Mar 2007

I believe I understand how the duration, resolution and N display works.
This is automatically calculated based on the scope's window time to optimize number of points displayed.

The question I have is how to get the full buffer of 2M points at 10ns?
Will Get buffer do it?
Will it get all 2M points instead of just the ones displayed?

Also, how do I change the capture resolution to get for example 2M points at 10uS resolution?

And a comment:
I think there should be a slider control in Scope and Tracking graph windows to show the buffer length and the portion currently displayed and allow adjustment of the viewing area by sliding or clicking on the area of interest.

There may be easy answers to all my questions but they are not clear to me from the manual.

14 Mar 2007
Posts: 413

Thanks for the question - they are answered in the guided tour document, but here are some quick answers:
1. The duration is total time captured by the scope, which is the larger of 10msec or the scope graph time width.
2. The resolution is the sample resolution in the acquisition unit. With 4M samples, and 2 frames, or 2M samples per frame, the resolution will be 10ns up to durations of 20ms, and then will be duration/2M. So with a 1 second duration, the resolution is 1/2M = 500ns.
3. The N Display number is the number of decimated samples that are present on the display.

To get a full buffer into the PC click the 'Get Frame' button on the Cleverscope Control Panel. All the samples in the frame will be transferred to the PC, and N disp number updated. Decimation now takes place out of the buffer in the PC. This is a slower process than decimation in the acquisition unit, even taking into account USB overheads (the acquisition unit decimates at 14ns/sample, while a fast PC decimates at about 60 ns/sample).

If you want 2M samples at 10us resolution, you will need a capture duration of 2M x 10u = 20 seconds. So set the scope graph 20secs wide. (Note that the scope graph is bounded to -22s to +22 s).

Usually we use the scope graph as the navigator - it shows the whole sample set, and use the tracking graph to zoom on a portion. As the tracers are mirrored from scope to tracking graph, and as the tracking graph tracks the scope graph tracer (if you have the tracking graph linked option on), you already have full context. There seems no need for a slider. We hope these answers are useful!

14 Mar 2007

That explains most of my issues.
I guess I was confused by the bar below the Get frame which is always very short.
I did not understand that the bar represents the decimated number if samples in a PC and not in the scope.
To make it full, I need to Get frame with all the points. Correct?

I am asking these questions because I do not have the scope by me.

Regarding the graphs, I am confused, because I am using your 'SPI Dac update.apc' file and the two graphs do not even look the same at the trigger point.
The real problem with tracking is that:
- the tracking works only if the window you are moving your mouse at is active. If it is not, the tracking line moves but the other graph does not track. You have to click the mouse to activate the window before tracking works.
- if your graph does not include the other graph, then the tracking line is outside the other graph and you do not know where you are (here you need a slider wich shows the position of one graph within the other)
- tracking only works properly if the scope graph displays complete graph and the tracking graph is used to zoom in and it must be selected. This is easily messed up when you click on any other Clever scope window.

My problem is that I always forget to click on the tracking graph to activate it and I see the cursor moving but the other graph does not respond.

I hope that this explains my problem and I think that the graph synchronization could be improved and I still see a need for a display slider.

14 Mar 2007
Posts: 413

Sort of! You are right - to get all the points you need to 'Get Frame' with the Full buffer transfer. ( You can also transfer 50, 10, 5, 2% of the buffer with correct decimation to make the file save smaller if you save the sample set). The progress bar shows progress as the frame is transferred. It is only active during a Get Frame, and shows progress for all transfer sizes.

I am confused by your statement 'the two graphs do not look the same at the trigger point'. To me they look identical, though off course the tracking graph is zoomed in 25x. Can you explain what you mean by that.

We have deliberately made it so that the graph must be focussed for the tracer to move the other graph. Otherwise, if you moved your mouse over the graph while moving to another window, the tracking graph would move, probably to your annoyance. How else could we do it?

If the two graphs are not looking at the same time point, and you are therefore looking at two zoomed views, our assumption is that you want to do this, and you know where you are in the graph. Otherwise, you would use the suggested model, and use the scope graph as the navigator, and the tracking graph as the zoomed view. Clicking on another Cleverscope window should not change how either the scope or tracking graphs look.

I do see that if you have not clicked on the tracking graph, the tracer will not move in the scope graph. We shall think about that (though again inadvertant change of anything is not something we really want to do).

We will think about a nice way of implementing a slider.

15 Mar 2007

I will email you a picture of the two graphs being different.
All I did is to open 'SPI Dac update.apc' and zoom in the scope graph to show exactly the same area as the tracking graph and the two graphs are different and I think they should be the same.

I think to fix the tracer problem you should not move the tracer on the inactive window when the mouse is moving over it.
Right now, the tracer moves but the tracer on the active window does not. In this case the tracers are not the same and is a major cause of confusion.
If you click any other non-graph window, then either tracer still moves with mouse over the graph and they should not.

The slider should show the area of the current view with respect to the full frame for each graph.
After zooming and moving some I lose track of where I am and the slider is necessary in both graphs.
It can also be used to quickly move the view in addition to current controls.

15 Mar 2007
Posts: 413

Ok, I agree that a slider could be useful as another way of moving the view. I'm think of a triangular pointer handle on the bottom edge of the graph, where the width of the width of the graph represents the width of the frame. Slide the handle to move the view.

Ok, it makes sense that we should not move the tracer when the graph is not active.

15 Mar 2007

I like your idea and I cannot wait for it to be done.
Back to Forum Index : Back to Questions   RSS
You must be logged in to post a reply

You need to Register or Log In before posting on these forums.


Your shopping cart is empty.