REVIEW - Siglent SDS2000 series - A Comparison of Features with Rigol DS2000

[marmad]

The fact that I paid for the SDS2202 and i hope that soon he will come to me ...
You claim that the firmware update will not? And Siglent will not fix bugs?

Well, Mark_O answered your question perfectly - I couldn't have said it any better myself.

One question for you: can I ask why you bought a Siglent SDS 2002?

While I certainly think the SDS2004 is the only contender in it's price class/feature set for 4-channels at the moment - as I stated in the video at the beginning this thread, for just 2-channels I think the Rigol DS2000 beats the Siglent SDS2002 in both features and overall value - and even more so if you consider the hacks available for the DS2000.

[pickle9000]

Excellent review. 

[gms]

Mark_O, Thank you very much for such a detailed response!

None of the products are perfect, and all have at least some minor issues that owners have to get used to.

You are of course absolutely right!
For me very important only the one bug - doesn't work the peak detector. With other disadvantages i can probably be able to reconciled.

One question for you: can I ask why you bought a Siglent SDS 2002?

marmad, the thing is that when I paid for this oscilloscope, I have not read this wonderful forum! My choice was based on the advertising promises. Obviously, I did a very foolish and short-sighted. But, as my mother said - clever people learn to others' mistakes and stupid from their mistakes  O0. I'm upset  .

I very-very hope that Siglent decent manufacturer of oscilloscopes and that he wants eventually making a competitive product. But I'm very-very concerned about their long silence... 

Friends, I very much hope that Google Translate translated my answer correctly to you and that you have properly understood the essence of my response  .

[rf-loop]

For me very important only the one bug - doesn't work the peak detector. With other disadvantages i can probably be able to reconciled.

Peak detect is working.
(exeptions: Bug1 and bug2)

Also previous claims that it use after ADC decimated data is totally wrong. It works just normally using full speed undecimated ADC data.   

I hope Siglent repair some Peak Detect mode bug(s) asap in some future FW.
Peak detect bug1: peak detect do not work at all if user have selected highest possible memory in menu but with other memory selections it works starting from 50ms/div and more fast horizontal speeds.
Peak detect bug2: Do not work with low horizontal speeds. It need work also in 100ms/div to 50s/div speed range.

It works 50ms/div and faster speeds. But of course if go to horizontal speed (dependent of selected memory depth)  where sampling speed is full, it do not work and also there it do not need work because it can there full peak detection naturally.  Example in image 2 there is 10kSa/s decimated sampling speed. One sample in every 100000ns. Pulse width 12ns. Of course this  12ns wide pulse is easy piece of cake to detect. If it works  after decimation this test fails totally.

Of course it can detect more short pulses what is used in these imaqes.  Here I just use 12ns pulses from SDG5082 because pulse generator what can produce more short pulses was not available for this quick run for these images.



Here some examples.
In last two images pulse period is 10us.
In other pictures pulse period is same as horizontal time/div
In SDS00009 there can see that some pulses produce "alias" and are semirandomly displayed.. but traces are quite dim... due to rare hits. (btw, if look image there can see used pulses shape alias...  )

EDIT:
bug 2 need study more.
= It works with all speeds down to 50s/div.
https://www.eevblog.com/forum/testgear/siglent's-new-product-msosds2000-series/msg609695/#msg609695

[marmad]

Peak detect is working.[/b]
(exeptions: Bug1 and bug2)

   Funny - is that similar to: my car is working - I just can't push the accelerator all the way to the floor or drive slower than 50km per hour?

Also previous claims that it use after ADC decimated data is totally wrong. It works just normally using full speed undecimated ADC data. 

Your images certainly don't disprove anything I wrote (notice the word "longer"):

Instead, it seems to me that the problem is that (at some/all longer memory lengths and/or slower timebases) the Siglent doesn't do Peak Detect on the sample memory before down-sampling, thus making it possible to miss peaks that existed. And it seems that this might be a general problem with the Siglent in a few areas; i.e. in service of speed (fast throughput) they are doing certain (or all?) things post-decimation.

1) I would hope that the Siglent could do Peak Detect correctly with only a 7k sample size to display.
2) I never "claimed" anything - I was only speculating - but based on evidence that it misses peaks with long memory setting(s).

[gms]

Thank you very much, rf-loop! Your comment could probably serve as for me a consolation  , but I need a fully functional peak detector over the entire range.

Therefore, it remains for me now only wait and hope that Siglent finalize this feature .

Tell me please, can I ask you a tactless question (I apologize in advance ) - You bought Siglent sds2000 or just use them (eg at work)?

[tautech]

Thank you very much, rf-loop! Your comment could probably serve as for me a consolation  , but I need a fully functional peak detector over the entire range.

Therefore, it remains for me now only wait and hope that Siglent finalize this feature .

Tell me please, can I ask you a tactless question (I apologize in advance ) - You bought Siglent sds2000 or just use them (eg at work)?

Please see post:
https://www.eevblog.com/forum/testgear/siglent's-new-product-msosds2000-series/msg562665/#msg562665
If you examine rf-loop's profile you will find the answer you seek. 

[gms]

Please see post:
https://www.eevblog.com/forum/testgear/siglent's-new-product-msosds2000-series/msg562665/#msg562665

Thank you, tautech! This information is reassuring  .

If you examine rf-loop's profile you will find the answer you seek. 

Yes, thank you, I have guessed  .

[nctnico]

Bugs:
- MSO signals on wrong timescale with decode enabled. IMHO digital signals should be visible with decode enabled.
- 'Peak detect' is not 'peak detect'. but envelope mode
- Signals display wrong (out of phase) when turning timescale or horizontal position knob quickly in 'stop' mode.
- When averaging is enabled the averaged signals are no longer visible in 'stop' mode (*).
- High resolution mode does nothing
- Finite persistance time setting does not seem to change persistance period.
- MSO signals are out of sync with analog channels when zooming in. Sometimes too early sometimes behind.
- Cursors (X / vertical) not working in MSO mode without analog channels active
- Decode doesn't work in single trigger mode
- Serial decoding doesn't always decode (shows empty red balloon) if the timing of the serial data is slightly off.

Improvements:
- Use partial FFT with higher frequency resolution when FFT is zoomed in
- Use averaging on FFT result (on my TDS744A I usually use 10 to 20 averages)
- Allow digital signals for decode
- Make 'intensity adjust' button less sensitive to rotation when selecting something from a menu list
- Make clear that decode and measurement statistics cannot be enabled simultaneously
- Allow decode and measurements to be used simultaneously.

(*) Yesterday I was doing some measurements with the cursors on a very noisy unstable signal. To clean the signal up I enabled averaging. But when I pressed the 'run/stop' button to stop the acquisition the noisy signal is displayed instead of the averaged signal.

edit: added finite persistance bug.
edit2: removed envelope mode
edit3: Added MSO signals out of sync with analog channels
edit4: Cursors not working in MSO mode
edit5: Decode doesn't work in single trigger mode, serial decode bug

It seems the new firmware became available. It is version 1.1.37.1. I would have expected something like 1.2.x which indicates a major version jump. So far very little seems to be fixed:

Fixed:
- I have not been able to reproduce the digital + analog out of sync bug
- Cursors work in MSO mode
- Variable persistence settings work

Improved:
- The select knob seems less sensitive
- Peak detect works in slow time/div settings but not in roll mode
- In slow time/div settings a red bar is shown to tell what the scope is doing

New bugs:
- FFT is broken. I can't get a spectrum of the 1kHz probe adjust signal. But then again the FFT wasn't useful to begin with.
- Zoom doesn't work with analog + digital channels

Some cosmetic things have been changed but the fundamental problems remain. The new firmware version doesn't show where more than 4 months of work went into. At this rate we'll never see firmware on the SDS2000 which can live up to the promisses Siglent made.

All in all I think I can stick with my earlier conclusion: DON'T BUY! My new year's resolution is to get an Agilent/Keysight oscilloscope in 2015!

[marmad]

It seems the new firmware became available. It is version 1.1.37.1. I would have expected something like 1.2.x which indicates a major version jump. So far very little seems to be fixed:

Fixed:
- I have not been able to reproduce the digital + analog out of sync bug
- Cursors work in MSO mode
- Variable persistence settings work

Improved:
- The select knob seems less sensitive
- Peak detect works in slow time/div settings but not in roll mode
- In slow time/div settings a red bar is shown to tell what the scope is doing

New bugs:
- FFT is broken. I can't get a spectrum of the 1kHz probe adjust signal. But then again the FFT wasn't useful to begin with.
- Zoom doesn't work with analog + digital channels

Some cosmetic things have been changed but the fundamental problems remain. The new firmware version doesn't show where more than 4 months of work went into. At this rate we'll never see firmware on the SDS2000 which can live up to the promisses Siglent made.

All in all I think I can stick with my earlier conclusion: DON'T BUY! My new year's resolution is to get an Agilent/Keysight oscilloscope in 2015!

Thanks for the update. Unfortunately, I don't have an SDS2000 anymore to test with. Did they manage to get High Res mode working? And any change with the way that intensity-grading is handled (i.e. previously it was strictly based on wfrm/s and didn't include sample overlap results pre-decimation)?

[nctnico]

Selecting high res mode doesn't alter anything so I guess they didn't manage to implement that either.

[Mark_O]

Some cosmetic things have been changed but the fundamental problems remain. The new firmware version doesn't show where more than 4 months of work went into. At this rate we'll never see firmware on the SDS2000 which can live up to the promises Siglent made.

That's very disappointing to hear.     It must be even more disappointing to you, after the long wait.  I can see how your patience would be exhausted.  Thanks a lot for the update though.

All in all I think I can stick with my earlier conclusion: DON'T BUY! My new year's resolution is to get an Agilent/Keysight oscilloscope in 2015!

That seems like sound advice.  Once again, an excellent hardware design with lots of potential, let down by an inadequate firmware implementation.

Vendors need to learn that simply shipping a "sort of working" product, and marketing it aggressively, is just not good enough.  The promise that "some day" the firmware and working features will match the hardware potential (maybe) is not good enough.  Being in a hurry to ship an unfinished product, just because you see market-share eroding to sales of competitor's products, is not good enough.

Frankly, this is unprofessional performance, and should be an embarrassment to Siglent.  I feel bad for the hardware team there, who worked hard to get their share of the efforts done well, and provide the foundation for a significant competitive advantage.  Only to be let down by the firmware team.

[tautech]

It seems the new firmware became available. It is version 1.1.37.1. I would have expected something like 1.2.x which indicates a major version jump.

Did you not get 1.1.1.37.2 linked in Siglents post:
https://www.eevblog.com/forum/testgear/siglent-technical-support-join-in-eevblog/msg573377/#msg573377

[nctnico]

I'm using version 1.1.37.2. So many numbers....

[nctnico]

Oh and the cursors got screwed up too. Before (version 35) pushing the select button iterated between cursor1, cursor2 and both but the cursors stayed in horizontal or vertical mode. Now it iterates between vertical cursor1, vertical cursor 2, vertical both, horizontal cursor1, horizontal cursor2, horizontal both.  The old behaviour was already a dissaster; IMHO pushing the button should toggle between cursor 1 and cursors 2 and not include both.

[Siglent]

New bugs:
- FFT is broken. I can't get a spectrum of the 1kHz probe adjust signal. But then again the FFT wasn't useful to begin with.
- Zoom doesn't work with analog + digital channels

Dear nctnico,
First the FFT function is not broken, you  need to  reduce the memory depth and sampling rate while testing the 1KHz probe adjust signal.
The second problem is about the Zoom function, it is true,  this version do not support zoom with analog + digital channels, we hope to add it  in next firmware version, thanks!

[rf-loop]

Using FW version 1.1.1.37.2

FFT is not broken.

Here some examples. Image names tell something.

First images are levels using 1MHz (sine) signal. Highest level is +3dBm what is over TFT visible area but still not clipped and not so much distorted in front end (and inside ADC range) Images are: +3, -7. - 17, -27, -37, -47, -57dBm.  (generator: one crap HP8657B,  displayed level - level not checked ! )

Then some other pictures, just for fun.

Then last image with 1kHz square, just for:

New bugs:
- FFT is broken. I can't get a spectrum of the 1kHz probe adjust signal.

FFT  is not broken after FW upgrade.

[nctnico]

It seems you are using the zoom function as a window to select the part of the signal you want to see with FFT. I have not tried that yet.

[rf-loop]

It seems you are using the zoom function as a window to select the part of the signal you want to see with FFT. I have not tried that yet.

Yes, mostly 2x zoom for better visibility.

For other readers who do not know this model: YT display zoom do NOT affect FFT. But zoomed YT is displayed under FFT window.

FFT have 1k sample points what give 0.5k resolution. (pity)
FFT full scale is always 0Hz to  samplefrequency/2

So if want look example range 0 - 25kHz then user need set oscilloscope for 50kSa/s for best available resolution (~50Hz in this case) for this frequency range.  FFT zoom 1x - 10x do not give more FFT resolution, is is only image horizontal stretch (zoom) horizontally.

 

[Wuerstchenhund]

I've been following this thread with some interest as I considered getting a SDS2000 as a tertiary scope, but quite frankly I think the state of the firmware at this stage is pathetic. They have shown that they can produce good hardware but Siglent is a prime example why the Chinese have a reputation for their software being crap.

At this point in time the only Siglent scope I would consider buying is the SDS3000, and this only because the software is made by someone who actually knows what they're doing (LeCroy). I guess the ugly state of the SDS2000 firmware is the reason why LeCroy has again choosen some Iwatsu scope over the SDS2000 as their replacement for the WaveJet 300A, despite using Siglent for other products.

This one-step-forward-and-two-steps back with the SDS2000 firmware just shows me that sticking with second hand scopes isn't such a bad idea after all.

[rf-loop]

This one-step-forward-and-two-steps back with the SDS2000 firmware

Steps to forward are too short and slow, but what are these back steps?
Back step is something that it have been ok previously and after update bad or broken.
I have not seen any documents about back steps. (exept terrible situation wirh SDG1000 series LOG sweep total hassle after new FW but this was nearly immediately fixed)

[nctnico]

The way the cursors work and no zoom in MSO mode in 'version 37' are definitely a step back. Besides that there are still many functions which are mutually exclusive like measurements and digital mode.

[Wuerstchenhund]

Steps to forward are too short and slow, but what are these back steps?
Back step is something that it have been ok previously and after update bad or broken.
I have not seen any documents about back steps. (exept terrible situation wirh SDG1000 series LOG sweep total hassle after new FW but this was nearly immediately fixed)

What about this:

It seems the new firmware became available. It is version 1.1.37.1. I would have expected something like 1.2.x which indicates a major version jump. So far very little seems to be fixed:

[...]

New bugs:
- FFT is broken. I can't get a spectrum of the 1kHz probe adjust signal. But then again the FFT wasn't useful to begin with.
- Zoom doesn't work with analog + digital channels

Take the zoom as an example.

And I know you said that FFT isn't broken but the way it's described in this thread doesn't look like it's working as it should.

[nctnico]

Regarding FFT I'm really missing my old Tektronix TDS744A. Albeit being very slow at least it allowed to choose the FFT length (and thus resolution) and clean up the FFT result using averaging.

[Wuerstchenhund]

Regarding FFT I'm really missing my old Tektronix TDS744A. Albeit being very slow at least it allowed to choose the FFT length (and thus resolution) and clean up the FFT result using averaging.

The TDS744A's FFT wasn't that great but in my opinion the way FFT is implemented in modern low-end scopes (not just Siglent, Rigol isn't really much better) is absolutely atrocious.