Suggested by KedasProbe on the EEVblog forum. Thank you!
The video is NovaOS's screen on its built-in test signals, recorded from the same code that runs on the scope.
| Linear only | Between two samples, the straight line from one to the next |
| Dots | Only the acquired samples, each at its own time |
| Sample marks | The line, and a mark on every sample it joins |
NovaOS joins two samples with a straight line, never with sin(x)/x, so nothing on the screen is added between them that was not measured. When the screen holds fewer samples than the trace has points (a fast time base, or two or four channels sharing the sample rate), the acquisition engine sends the measured samples with each trace, so the scope UI knows exactly where every one of them is.
Display, Waveform, Dots draws only the samples the scope acquired, each at its own time and value, with nothing between them, at every time base. At 10 ns/div with two channels on a DHO, that is about 62 samples across the screen, each a dot. The XY plot pairs the measured samples the same way, and with persistence every acquisition's samples are drawn as dots. Over SCPI: :DISPlay:TYPE DOTS.
With Vectors, Sample marks on the Display sheet's Waveform row marks each measured sample on the line with a small square, once the samples are at least 6 pixels apart on the screen; closer than that the line alone is drawn, and the marks come back as you zoom in. Kept across power cycles, and set over SCPI with :NOVAos:DISPlay:MARKs ON or OFF.
The tests check that the trace is exactly the straight-line interpolation of the samples sent with it, with one and two channels on, and that the GPU and CPU drawing of dots and sample marks agree pixel for pixel.
Dots and sample marks at fast time bases with two and four channels on our DHO924S and MHO934, against the samples read back over SCPI.
Your Bode ask is TimFox's request too: each point settled, checked for the right frequency and averaged before it is plotted, however long the sweep takes. It has its own page.
Try Dots and Sample marks on your DHO with two or four channels on at a fast time base, and tell us whether you can tell every measured sample.
Linear interpolation between samples, and a dots-only view that shows just the measured samples, so it is clear what was measured, also when several channels share a DHO's sample rate.
I would like the option to have linear interpolation or dots only selection, so I know for sure what I measure. (I have an DHO so sample rate isn't that high when you use more channels.)
Try it and tell us what to change: reply on NovaOS's thread on the EEVblog forum.