Siglent SDS2000X HD, 12bit, 2GSa/s high definition oscilloscope serie.

[kripton2035]

I´m still hoping there will be a discount offer for the probes only someday.
Yes, there will be DIY alternatives, but I like to have the "original".
But not for this price.

yep a digital probe clone on ebay is only $50 ...
... and you have to hack the LA licence...

[kripton2035]

Having some fun with writing apps for the scope:

Code: [Select]

/usr/bin/siglent/usr/mass_storage/U-disk1 # ./rusty_siglent_app
Hello, world!!!


would you like to publish the source code of this "hello" app ?

[TopQuark]

Having some fun with writing apps for the scope:

Code: [Select]

/usr/bin/siglent/usr/mass_storage/U-disk1 # ./rusty_siglent_app
Hello, world!!!


would you like to publish the source code of this "hello" app ?

Sure, you need the Nix (https://nixos.org/) build system with flakes enabled (https://nixos.wiki/wiki/Flakes), and only need to run

Code: [Select]

nix develop to reproduce my my cross compile environment and

Code: [Select]

nix build   to cross compile the correct binaries.

product_rcs_begin.sh contains the recommended contents of /usr/bin/siglent/product_rcs_begin.sh on your scope's filesystem, so that you can telnet and ftp to the scope after the scope boots up.

upload.sh is a helper script to upload the compiled binary file to the scope and run it.

[TopQuark]

Also, here's the scope's devicetree in case you feel like interfacing with the fpga and poke around the DMA busses...

[TopQuark]

Can anyone with the SDS2kHD and a digital probe set verify what I am seeing? It seems like the digital channel capture rate of the HD is very very slow compared to the SDS2k Plus.

I am only getting 16 wfm/sec on the HD (100ns/div, 2kpts, 8 channel), whereas the Plus is achieving 2.3k wfm/sec with the same settings.    I have reset the scope a few times and repeated the tests, and the results are the same.

Also on the HD, fast mode in the acq menu cannot be activated with digital channels on, whereas that's possible on the Plus. 

Edit: I am using the original Siglent digital probes, so it shouldn't be a hardware problem

[Martin72]

Sounds like a bug to me - Want you place it in the bug/feature thread ?

[TopQuark]

Just not sure if I'm doing something silly and accidentally killing the performance of the digital mode. Want some confirmation before I make a big fuss out of it.

As a sanity check, I just verified the analog waveform update rate, and I can get the advertised 100k wfm/s with the analog channels.

[Martin72]

I do have the MSO option, but unfortunately no logic probes...
Maybe tautech can test it when he got a HD there.

[tautech]

I do have the MSO option, but unfortunately no logic probes...
Maybe tautech can test it when he got a HD there.

Sorry, no SDS2000X HD here.

[TopQuark]

I do have the MSO option, but unfortunately no logic probes...
Maybe tautech can test it when he got a HD there.

Sorry, no SDS2000X HD here.

@tautech, is it possible for you to reach out to Siglent and ask them what we should expect out of the HD in terms of digital channel waveform update performance? Thanks

[Martin72]

Surely not 16wfms/s...

[2N3055]

Surely not 16wfms/s...

Yeah I get 16wfms/s too.
Well in that mode it handles double the data and data is routed slightly differently.. It has to be in Slow mode, because it has to capture buffer and process all of it before going to next one. It cannot just do bursts like in Fast mode.

Martin, you can check without probes, enable digital channels, and connect 100KHz (or more) signal to CH1 and trigger on that..

@TopQuark, sorry, what FW are you running..

[Martin72]

Hi,

OK, try it later on friday.

It has to be in Slow mode, because it has to capture buffer and process all of it before going to next one.

TopQuark also got the 2000X+, with 2300 wfms/s...

[tautech]

I do have the MSO option, but unfortunately no logic probes...
Maybe tautech can test it when he got a HD there.

Sorry, no SDS2000X HD here.

@tautech, is it possible for you to reach out to Siglent and ask them what we should expect out of the HD in terms of digital channel waveform update performance? Thanks

Sorry, no.

First consult datasheet and manuals then wait for feedback from the few beta testers watching this thread as they have more direct contact with product managers than anybody.

I do note max sampling is listed in the datasheet as 500 MSa/s and Memory Depth 50 Mpts/ch although analog memory settings might also impact on WFPS performance. Those with these instruments at their fingertips are best qualified to research this issue.

[TopQuark]

Surely not 16wfms/s...

Yeah I get 16wfms/s too.
Well in that mode it handles double the data and data is routed slightly differently.. It has to be in Slow mode, because it has to capture buffer and process all of it before going to next one. It cannot just do bursts like in Fast mode.

Martin, you can check without probes, enable digital channels, and connect 100KHz (or more) signal to CH1 and trigger on that..

@TopQuark, sorry, what FW are you running..

Well, good to know the problem is reproducible, thanks for checking. I am running the latest 1.2.1.1 software, with FPGA version 2022-07-08.

On the SDS2504X Plus it is significantly faster at 2300 wfm/s, digital phosphor works, where you can see the digital trace overlap, the responsiveness is much better, and fast mode can be toggled on in the acquire menu. It is almost like the digital channels are processed in software on the HD, whereas the Plus processes it in hardware.

I do have the MSO option, but unfortunately no logic probes...
Maybe tautech can test it when he got a HD there.

Sorry, no SDS2000X HD here.

@tautech, is it possible for you to reach out to Siglent and ask them what we should expect out of the HD in terms of digital channel waveform update performance? Thanks

Sorry, no.

First consult datasheet and manuals then wait for feedback from the few beta testers watching this thread as they have more direct contact with product managers than anybody.

I do note max sampling is listed in the datasheet as 500 MSa/s and Memory Depth 50 Mpts/ch although analog memory settings might also impact on WFPS performance. Those with these instruments at their fingertips are best qualified to research this issue.

No worries, it was a long shot anyways. I'll file the issue in the bugs thread and see if it gets picked up by anyone. 

[kripton2035]

before ... and after...
I see things ... better !

[Martin72]

On the SDS2504X Plus it is significantly faster at 2300 wfm/s, digital phosphor works, where you can see the digital trace overlap, the responsiveness is much better, and fast mode can be toggled on in the acquire menu. It is almost like the digital channels are processed in software on the HD, whereas the Plus processes it in hardware.

Hm-hm..Sounds "interesting"...

OK, try it later on friday.

Haha, can´t do this, I didn´t got a second scope here for wfm measure, totally forgotten.

[rf-loop]

Can anyone with the SDS2kHD and a digital probe set verify what I am seeing? It seems like the digital channel capture rate of the HD is very very slow compared to the SDS2k Plus.

I am only getting 16 wfm/sec on the HD (100ns/div, 2kpts, 8 channel), whereas the Plus is achieving 2.3k wfm/sec with the same settings.    I have reset the scope a few times and repeated the tests, and the results are the same.

Also on the HD, fast mode in the acq menu cannot be activated with digital channels on, whereas that's possible on the Plus. 

Yes it is DSO like in this. Acquistion, display, next acquistion, display.

If Bus display is off around 16wfm/s max

And with bus decode, it slows bit. here example 8bit bus
With Auto memory (Analog channel length 400pts) around 12wfm/s
With Fixed memory (Analog channel length 20Mpts (400pts displayed and rest outside of screen) around 11wfm/s
Analog channel 40MHz and digital 8 channels all get parallel 20MHz
wfms measured using other scope.

Btw, if wfm/s is lot of faster and signal is changing, how you think what bus decode is displaying.. only way is that it capture just single and display its decoded result and after then next capture... no DPO stacking overlaid (for what).

How it can read if it works like DPO where is severe acquistions overlaid in every display refresh. Think twice.

[Martin72]

And what´s the explanation for that the 2000X+ wfms/s is about 140x faster in the same situation ?

[TopQuark]

Hi, shot a quick video showing the differences between the Plus and the HD in digital mode. It demonstrates the speed difference, and that the Plus does have digital phosphor working.

https://youtu.be/tsYzZ_vCalo

[baldurn]

I am considering buying but can I hack it? How? I am not finding anything...

[TopQuark]

I am considering buying but can I hack it? How? I am not finding anything...

Yes, you can hack it, there's a few owner of the HD who have fully unlocked their scopes. As to how, I am not the one to be able to answer, look through the thread to see who's the mastermind. 

[elimenohpee]

I am considering buying but can I hack it? How? I am not finding anything...

I'm with you, I'm in the market for a scope of this caliber. I'm up for buying one but if it's not exactly hackable, I'm leaning to the SD2000x+

[Martin72]

Actually you get a bandwithupgrade and the decoders for free when buying a HD, just saying...
IMHO a easy to hack guide won´t be published because it is not easy to hack.
But it could be done, that´s assured.
The rest TopQuark have already "said" in his last post.

[baldurn]

Actually you get a bandwithupgrade and the decoders for free when buying a HD, just saying...
IMHO a easy to hack guide won´t be published because it is not easy to hack.
But it could be done, that´s assured.
The rest TopQuark have already "said" in his last post.

Yes it is a good offer but I would still need to go for one the more expensive options for this not be a downgrade. And that would put it out of my budget range :-(

I did manage to hack my SSA3032X-R and that was not too easy either. I can do Linux all day long. May even be willing to open it up to connect to the serial. But I do not want to try to hack it all on my own. Then it becomes a project with no guarantee that I would succeed and I will rather spend my time on something else.