Siglent's new product- MSO/SDS2000 Series

[tinhead]

@tinhead
 What is  your waveform?I just test a sine waveform

actually square, but that didn't matter at all, from your pictures it is clearly visible that Siglent signal looks great, nothing capped

[Carrington]

This signal seems to be still within the dynamic range of the input stage.
And regard to noise it looks good, but I prefer the first images, @500ns and 14Kpts.


@ Herman: Thanks for satisfying our curiosity.

[Herman]

@Carrington
I didn't reset the measurement data when I tested noise at 500ns,14Kpts,so it maybe the first data.

[Herman]

@tinhead
I found a strange thing?input a 10MHz sine wave,when I turn on the Zoom function,and set the zoom time base to 20ns,the result is different. anybody kowns why?

Rigol DS2000,shakes left and right

Owan SDS shakes a little

Siglent SDS2000 

[marmad]

@tinhead
I found a strange thing?input a 10MHz sine wave,when I turn on the Zoom function,and set the zoom time base to 20ns,the result is different. anybody kowns why?

Sorry if I'm dense, but why are you (someone who is clearly either an engineer or salesman from Siglent) asking Siglent comparison "questions" of forum members here?

It seems rather strange to say the least, but likely disingenuous - as some sort of underhanded way to point out some strength of the device.

Instead of the continual "leaking" of information about the DSO in this crude, obvious way, just send one to Dave already and have it reviewed.

[tinhead]

@tinhead
I found a strange thing?input a 10MHz sine wave,when I turn on the Zoom function,and set the zoom time base to 20ns,the result is different. anybody kowns why?
...
Rigol DS2000,shakes left and right
Owan SDS shakes a little

According to the datasheet, Siglent is using digital trigger which reduce trigger jitter, therefore no "shaking". However, the Siglent pictures shows small bug, the trigger need to be adjusted. The position where the two blue lines crossing should be there where red line is. So in principe the waveform need to be moved to the right by 30ns.

[marmad]

According to the datasheet, Siglent is using digital trigger...

According to Rigol (via Drieg), the DS2000 uses a digital trigger as well - it's just not advertised in the datasheet as a feature.

[BravoV]

Sorry if I'm dense, but why are you (someone who is clearly either an engineer or salesman from Siglent) asking Siglent comparison "questions" of forum members here?

It seems rather strange to say the least, but likely disingenuous - as some sort of underhanded way to point out some strength of the device.

Instead of the continual "leaking" of information about the DSO in this crude, obvious way, just send one to Dave already and have it reviewed.

C'mon Mark, this is starting to cross the fine line here.

Personally, I don't mind even there are truly Siglent's armies in disguise posting Siglent's stuff here, why ?

Easy and simple, and you don't need to be a genius to recognize forum's etiquette, this thread is "specifically" made from the beginning for discussing "Siglent" product.



Pstt... to Herman, keep updating us on this new scope, I'm pretty sure many of us here in the forum would like to hear and know more, and also maybe you should start posting these infos in those Rigol's thread to tick off someone there.   

[marmad]

C'mon Mark, this is starting to cross the fine line here.

Sorry - which line is that? An ethical one?

Easy and simple, and you don't need to be a genius to recognize forum's etiquette, this thread is "specifically" made from the beginning for discussing "Siglent" product.

I've got no problem with the posting of info about Siglent's new product - and I'm curious to hear/see more details about it myself. What I find strange - as I think I made clear in my comment - are attempts to pretend the poster does NOT work for Siglent - and is asking "questions" to illicit serious replies; you have only to read some of his posts to understand what I mean - to me, this seems to be contrary to 'forum etiquette'. Why not just state clearly his affiliation - and then just post the information/documentation of the DSO?

I find it weird - YMMV.

[BravoV]

I've got no problem with the posting of info about Siglent's new product - and I'm curious to hear/see more details about it myself. What I find strange - as I think I made clear in my comment - are attempts to pretend the poster does NOT work for Siglent - and is asking "questions" to illicit serious replies; you have only to read some of his posts to understand what I mean. Why not just state clearly his affiliation - and then just post the information/documentation of the DSO?

I find it weird - YMMV.

Weird, maybe, but again, as long all of this "drama" is confined within this thread, I don't see what is wrong here.

Even there are tons of "artificially orchestrated" discussions, again, as long its confined in a specific thread without leaking or contaminating "other non related thread" like we've seen many examples in here , then ... again ... why not ?

[marmad]

Even there are tons of "artificially orchestrated" discussions, again, as long its confined in a specific thread without leaking or contaminating "other non related thread" like we've seen many examples in here , then ... again ... why not ?

I suppose... and honestly, I didn't really care much about it until the OP's "question":

"Dear forum members - can you please explain back to me why the DSO that I'm promoting/selling seems better than the competing brands in the following scenario?" 

Just post the data and say "the trigger jitter (in Delayed Sweep mode) appears better on Siglent than Rigol or Owon"... sheesh!

But anyway - I guess it's just me - I will bow out of this thread now...

[BravoV]

Mark, you are a very well respected member, at least to my eyes, especially with those contributions you have made in this forum, no doubt about those values. Its just it will be a waste if this kind of silly thing starting to affect your reputation which personally I respect it very much, of course assuming you are not taking this reputation thing lightly too, just my 2 carbon resistors worth of "suggestion".

I'm done here.

[tinhead]

According to the datasheet, Siglent is using digital trigger...

According to Rigol, the DS2000 uses a digital trigger as well - it's just not advertised in the datasheet as a feature.

R&S did spend lot of money to advertise their digital trigger, but Rigol forgot to incude SUCH feature into datasheet?
Not even in DS4000 or DS6000 datasheet?
Seriously, they must have the worst marketing dep. in our and all parallel universes.

As i don't give a shit about marketing papers, let's check facts:
- there is 16bit DAC on the Rigol PCB
- there are buffers and ref mux
- FPGA LVDS input can be easily used as fast speed comparator, so no need to use an extra parts.
- LMH6518 aux output is routed somewhere (usually this signal will be used for triggering), i guess to FPGA

What i don't know:
- is the DAC / mux out routed as well to FPGA?
- what the FPGA design is really doing

So i can't say for sure how the triggering has been implementd. But the fact that Rigol is using two signal paths (and this is one of the reason why there is tigger jitter visible) is reason enought to NOT call, whatever they really implemented, this an digital trigger because one of the major points and benefits of digital triggering is single signal path.

Anyway, i'm sure, in case Siglent did used two paths as well, Rigol will then update their datasheets

I don't mind, for me is sufficient to have protocol trigger and decoders (and yes, i know how great digital trigger is), but that's me, other will have other requirements.

[marmad]

R&S did spend lot of money to advertise their digital trigger, but Rigol forgot to incude SUCH feature into datasheet?
Not even in DS4000 or DS6000 datasheet?

Just when I thought I was out...   

As I mentioned in PM to you, I don't know if it's true or not - I only have the following snippet from a Rigol employee email:

"I asked R&D about this situation, it is not a bug, because DS4 and DS6 use analog trigger, but DS2000 use digital trigger, the advance of digital trigger is the signal displayed on screen stably and no offset compared with analog trigger."

Whether this is really true - or why they wouldn't advertise it if it was - I couldn't say.

The posted GIFs are of Delayed Sweep mode - there is no trigger jitter visible on the DS2000 in normal mode - so clearly, the jitter is related to the Delayed Sweep. So I ran some tests on the DS2000 using a 10MHz sine wave like Herman.

Edit: Final results below.

Using 14MB memory depth and entering Delayed Sweep mode:

@ any time base setting from 2ns/div to 500us/div (SR: 2GSa/s) = no conspicuous jitter and the zero-crossing of the waveform is located at screen center.
@ 1ms/div (SR: 1GSa/s) = no conspicuous jitter but the zero-crossing of the waveform is offset ~5ns to the left of screen center (just like, but less than, the Siglent image).
@ 2ms/div (SR: 500MSa/s) = no conspicuous jitter but the zero-crossing of the waveform is offset ~5ns to the right of screen center.
@ 5ms/div (SR: 200MSa/s) = ~±2ns jitter around screen center.
@ 10ms/div (SR: 100MSa/s) = ~±8ns jitter around screen center.

Running the tests again while using the 56MB memory depth:

@ any time base setting from 2ns/div to 2ms/div (SR: 2GSa/s) = no conspicuous jitter and the zero-crossing of the waveform is located at screen center.
@ 5ms/div (SR: 500MSa/s) = no conspicuous jitter but the zero-crossing of the waveform is offset ~5ns to the right of screen center.
@ 10ms/div (SR: 200MSa/s) = ~±2ns jitter around screen center.

...shows that any jitter (and/or any waveform offset to trigger point) in Delayed Sweep is directly related to the sample rate on the DS2000.

[Hydrawerk]

Is there a digital trigger in Agilend DSOX2000 / 3000 series? But I do not find it much important.

[Herman]

@tinhead

According to the datasheet, Siglent is using digital trigger which reduce trigger jitter, therefore no "shaking". However, the Siglent pictures shows small bug, the trigger need to be adjusted. The position where the two blue lines crossing should be there where red line is. So in principe the waveform need to be moved to the right by 30ns.

digital trigger reduce trigger jitter, but I didn't mean no jitter. maybe it is just a little better than analog trigger.

[Herman]

test set up
input FM Modulation waveform to the oscilloscope(sine 50KHz,4Vpp,FM frequency 50Hz and FM Dev 50KHz), at the same time base, like 1us,2us,and 5us, and measure the frequency of the trigger out signal.
20864wfm/s


2715wfm/s


15495wfm/s


1430wfm/s


9434wfm/s


1081wfm/s

It is pity they did not give detailed information what was exactly test set up (signal, measuring method - whole detailed test description  so that it can repeat exatly in other lab than siglent's own)

[grego]

I've got no problem with the posting of info about Siglent's new product - and I'm curious to hear/see more details about it myself. What I find strange - as I think I made clear in my comment - are attempts to pretend the poster does NOT work for Siglent - and is asking "questions" to illicit serious replies; you have only to read some of his posts to understand what I mean. Why not just state clearly his affiliation - and then just post the information/documentation of the DSO?

I find it weird - YMMV.

Weird, maybe, but again, as long all of this "drama" is confined within this thread, I don't see what is wrong here.

Even there are tons of "artificially orchestrated" discussions, again, as long its confined in a specific thread without leaking or contaminating "other non related thread" like we've seen many examples in here , then ... again ... why not ?

I'm with Mark on this one to some extent.  If someone is posting here and has a commercial interest in what they are posting about it should be disclosed. 

[Mark_O]

test set up
input FM Modulation waveform to the oscilloscope(sine 50KHz,4Vpp,FM frequency 50Hz and FM Dev 50KHz), at the same time base, like 1us,2us,and 5us, and measure the frequency of the trigger out signal.

Thanks for the demos.  The new Siglents appear to have a very fast and fluid update rate, with analog-like gradients.  Congrats.  You may as well post the wfms/sec numbers from the file names: 20,864 vs. 2,715; 15,495 vs. 1,430; and 9,434 vs. 1,081.  (Ah, I see you did.  I just didn't notice them at the corners of each video.   )

It is odd though that the d@mned Rigol scope in pic2 won't sit still on the desk.  It keeps jumping around like a mexican jumping bean.    Makes it a bit hard to watch.

It's obvious that the frequency counter on the Siglent is getting updated faster, but the test isn't completely fair to the Rigol's, since they have a Stats box with 4 numbers being updated on every cycle as well.  Which the Siglent does not.  Why not switch that off?  (Or turn the equivalent item on on the Siglent too.)

NB:  Also, in #2: 500M in 14k on the Siglent and 2G in 56k on the Rigol isn't comparable either.  It's also curious that the Rigol FC is fairly stably locked at 109kHz there, instead of the varying 50k on all the rest of the tests.  I'm not sure why?

[marmad]

After further testing of the DS2000, the trigger jitter/offset of the Zoomed trace while in Delayed Sweep mode appears to follow this pattern:

When in Delayed Sweep mode:

Sample rate >= 1GSa/s: no conspicuous jitter or offset
Sample rate <= 500Ms/s: jitter and/or offset equal to a maximum of plus or minus one sample period

So for example, @ 200MSa/s it's equal to ±5n, while @ 10MSa/s it's equal to ±100ns.

Since the new Siglent obviously has some kind of similar issue (i.e. the offset pointed out by tinhead above), I'd be curious what it's specs were for other sample rates. I'm not sure about the sample rate in the posted GIFs (hard to read), but it looks to me like 100MSa/s on the Siglent GIF - which, on the Rigol, equals ±10ns of jitter/offset in Delayed Sweep. Since the Siglent is displaying ~30ns of offset, it seems in the same ballpark.


P.S. Further in-depth discussion of this vis-à-vis the Rigol DS2000 series should probably go to the appropriate thread.

[tinhead]

digital trigger reduce trigger jitter, but I didn't mean no jitter. maybe it is just a little better than analog trigger.

well there will be no extra jitter caused by e.g. separate (trigger) signal path, but of course clocked circuits are never jitter-free.

There is good R&S article about digital triggering (and it's benefits):

http://www.ecnmag.com/articles/2011/07/oscilloscope-performance-digital-triggering

[Carrington]

I have a SDS8102V and DS2072, and I can say that the SDS8102V's trigger sucks.


About video trigger in DS2000 and DS4000:

https://www.eevblog.com/forum/testgear/first-impressions-and-review-of-the-rigol-ds2072-ds2000-series-dso/msg266933/#msg266933

[Herman]

input a rapid rise edge signal(about 500ps), and then test the trigger jitter of SDS2000 series. the result is below, it also can be seen from the datasheet.

After further testing of the DS2000, the trigger jitter/offset of the Zoomed trace while in Delayed Sweep mode appears to follow this pattern:

When in Delayed Sweep mode:

Sample rate >= 1GSa/s: no conspicuous jitter or offset
Sample rate <= 500Ms/s: jitter and/or offset equal to a maximum of plus or minus one sample period

So for example, @ 200MSa/s it's equal to ±5n, while @ 10MSa/s it's equal to ±100ns.

Since the new Siglent obviously has some kind of similar issue (i.e. the offset pointed out by tinhead above), I'd be curious what it's specs were for other sample rates. I'm not sure about the sample rate in the posted GIFs (hard to read), but it looks to me like 100MSa/s on the Siglent GIF - which, on the Rigol, equals ±10ns of jitter/offset in Delayed Sweep. Since the Siglent is displaying ~30ns of offset, it seems in the same ballpark.


P.S. Further in-depth discussion of this vis-à-vis the Rigol DS2000 series should probably go to the appropriate thread.

[marmad]

input a rapid rise edge signal(about 500ps), and then test the trigger jitter of SDS2000 series. the result is below, it also can be seen from the datasheet.

Thanks for that, but I was talking about (because you posted GIFs from) Delayed Sweep (Zoom) mode. Any idea where the offset error (which tinhead pointed out) came from?

[Fagear]

Seems like LCD screen has very poor vertical angle - dim traces at the top of the screen disappear and black color already inverting. Feels like Rigol's screen has better viewing angle.