Products > Test Equipment
How does waveform updates on an oscilloscope work? Why do they work that way?
(1/14) > >>
ballsystemlord:
Although I tried to be an aware buying when I purchased my oscilloscope, I somehow didn't know about the implications of waveform updates per second. Dave mentioned it here: https://youtu.be/txPxo4TA0i4 offset 25:30 . And so I read up on the matter on rigol's blog where they have this picture:
Here's the rigol article https://int.rigol.com/news/blog/DS70000BLOG.html And the relevant quote:
"Since the speed of data sampling is much greater than the speed of data processing, sampling has to be stopped during data processing, which results in all waveforms during data processing to be lost because they are not collected."

Doesn't this cause a problem for protocol analysis?

Thanks!

EDIT: Never mind the second question. I read wrongly.
JPortici:

--- Quote from: ballsystemlord on April 17, 2023, 04:41:16 am ---Doesn't this cause a problem for protocol analysis?

--- End quote ---

No. I mean, yes, if you can't fit the frame of interest in the acquisition.
But you should really use an oscilloscope only to validate the analog part of the window. The moment the analog signal is correct you should switch to a protocol analyzer (which can usually also put data on the bus, which makes it a much more useful trigger. Imagine doing CANbus analysis of a network with only an oscilloscope).
And if you're looking for protocol error conditions you should use decode triggers (and an oscilloscope that use HARDWARE protocol decoders and not decodes from screen data) in which case you won't have dead time because the scope will process/display only buffers that contains the trigger conditions
Someone:

--- Quote from: ballsystemlord on April 17, 2023, 04:41:16 am ---Doesn't this cause a problem for protocol analysis?
--- End quote ---
Depends on the scope. Some scope + protocol decoders are run online in realtime hardware, they trigger the sampling/capture and would be blocked from triggering again until the "processing" is complete. But will not miss the first (or only) instance. Some scopes will only do the protocol analysis on captured data, slower than the data is coming in with those sorts of dead times between visibility (and chance to miss triggers). Obviously the scopes that aren't able to do realtime protocol decode are a little shy in saying that so it can be hard to tell, generally if the data sheet isn't making lots of noise about hardware protocol decode then its that slow type.
Fungus:

--- Quote from: ballsystemlord on April 17, 2023, 04:41:16 am ---Doesn't this cause a problem for protocol analysis?

--- End quote ---

Can you miss a packet? Yes.

But unless your eyes are fast enough to read all the data in real time as it decodes then it's not a problem. 

A modern 'scope will have enough memory to capture what came after the trigger point so you can stop the 'scope and inspect it.
Berni:
High waveform update rate only matters for 'analog' signals.

When looking for noise or glitches on a signal you want to see as much of the signal as possible. So a good modern high update rate oscilloscope might do 1000s of captures in the 1/60th of a second that it takes to update the LCD display. So to show all those waveforms the scope overlays them over each other in a transparent way. This gives them that smooth soft fuzzy look that you see on a classical CRT oscilloscope (since it is doing the same thing, drawing the waveform many many times on top of itself). In order to achieve this high update rate, the oscilloscope has to have a processing time as short as possible, so that it is capturing new signal data as much of the time as possible.

When looking at digital communication this is irrelevant. There you are only interested in making one big single shot capture. So the processing time is just the time before making the single shot capture and it appearing on screen. Since humans are really really slow in comparison that doesn't matter. Modern scopes have huge sample memories, so you can really capture a lot in single shot and then scroll trough all the data.
Navigation
Message Index
Next page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod