Products > Test Equipment
Keysight DSOS204A waveform update rate
<< < (5/8) > >>
bdunham7:

--- Quote from: free_electron on June 11, 2022, 01:04:56 am ---Sample fast , collect everything.

--- End quote ---

There are going to be limitations.  In my example with a 10MHz signal and a 10mHz anomaly, at a bare minimum you'd need 100MSa/s which for 100s is 10 billion samples--and typically you'd want 10X that many samples for a 10Mhz square wave.  So 100 billion samples is a bit more memory than is commonly available, so your alternative is to acquire and process (for display) all that data in real time--and that just takes a lot more FPGA horsepower than your typical budget scope has.  So yes, more faster is better, but you aren't going to eliminate retrigger and blind time over the scale of my test in anything I can afford.
KrzysztofB:
I'm starting to have very mixed feelings about this scope.
I setup a test: 10kHz sin waveform with a small glitch every second. Made a mask test on both RTB2004 and DSOS. Similar settings, 5MSa/s, 10kPoints.
During 60 sec test:
RTB did 21k tests and failed 26
DSOS did 2.4 ktests and failed 3

Pretty disappointing to be honest.
bdunham7:

--- Quote from: KrzysztofB on June 11, 2022, 07:33:12 pm ---RTB did 21k tests and failed 26
DSOS did 2.4 ktests and failed 3

Pretty disappointing to be honest.

--- End quote ---

That's an interesting slower-speed test.  Just for comparison I did the same-ish thing with a Siglent SDS2354X+.  I had 200µs/div and 20 cycles per screen--does that match?

@ 20kpts/10MSa/s (the closest I could get to yours with one channel) on I got 23.9K tests with 50 fails.
@ 10kpts/5MSA/s (switching on CH2 to halve the sample rate) it was 20.3k tests and 41 fails.

Yet in segmented mode, depending on the exact settings, I get 330k plus wfms/s similar to yours.  So the DSOS204A seems to be powerful enough in the acquisition department and very sophisticated in its analysis software, but just not quick about doing certain things.  I don't usually get to play with such nice things, but I suspect that this sort of tradeoff is common.  It's why a $100K scope may not make a good daily driver.
KrzysztofB:
Totally can accept that in real-time, waveform update rate is not great, but ok, something to accept.
In segmented mode it has absolutely no problem grabbing so much frames that would be enough to find all of those anomalies.
But one thing I absolutely cannot understand, is WHY you can't run mask test on the data that are already sitting in segments previously captured?
I dont mind if that test would run for a while, but it's absolutely not possible???? WHAT?
Also after capturing segmented data, I can't make it to work to Search for specific events, which ie. in RTB is easy, you just go to Search.

If you set triggers right it has absolutely no problem capturing every single glitch, but to set the trigger you have to have at least little idea what are you dealing with.

I'm very confused, especially that we are talking kHz signals. Why it even has that mask test if it works so poorely?
Someone:

--- Quote from: KrzysztofB on June 11, 2022, 09:43:28 pm ---Totally can accept that in real-time, waveform update rate is not great, but ok, something to accept.
In segmented mode it has absolutely no problem grabbing so much frames that would be enough to find all of those anomalies.
--- End quote ---
You come at the problem with fixated mind on the "one" way to solve it. The deep memory offline analysis scopes work very differently and use different approaches for the "same" problem.


--- Quote from: KrzysztofB on June 11, 2022, 09:43:28 pm ---Also after capturing segmented data, I can't make it to work to Search for specific events, which ie. in RTB is easy, you just go to Search.
--- End quote ---
Not sure if that scope has hunting tools for segmented mode, but you didn't use that scopes headline/flagship tool for that sort of problem: InfiniiScan (requires license?) which takes a deep memory capture and then searches back through it with complex or multiple different triggers.

If that is faster or not, you can measure!
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod