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

[wpwrak]

rf-loop, I was thinking of something like this: show N segments overlaid (with intensity grading, hardware allowing), one knob moves by +N/-N, other knob doubles/halves N. (Or pick a slightly larger factor if you like.) That way, one could very quickly search those thousands of segments for rare events. You don't even need an "overlay all" button, just spin the "zoom" knob some 16-17 stops in the "zoom out direction" and you're at the maximum N.

The history could be a lot more flexible if one could select what gets incremented/decremented. E.g., default to the least significant digit of the segment number, but have a knob/button(s) to select a different digit or a digit of the capture time instead. Then you could move by 10/100/etc. segments, by microseconds, whatever. It wouldn't necessarily show what happens between the jumps but one could at least navigate quickly to a specific segment or time and also see trends in the waveform.

Disclaimer: all I know about the SDS2000 is from eevblog, so I can't really tell whether those ideas fit with the user interface design, whether the scope would be responsive enough, etc.

- Werner

[marmad]

have you tried waveform update rate with 1&3 vs 1&2?

@Fsck:  I've enlarged the table in the first post to include vector waveform update rates for channels 1 & 3 plus all channels simultaneously (1 & 2 & 3 & 4) .

[marmad]

Also there need be one selection for overlay all frames. (with intensity grading of course), just one button, "overlay all segments".

Yes! This alone would increase the usefulness greatly - making checking for glitches or hunting anomalies much faster. I also thought a bit about how, if a glitch is found, you might locate that frame. One way that might work (and would be a fairly easy UI implementation), is to make the Universal knob move a selection box across all the divisions, top to bottom, left to right. When you stop over a division and click, the DSO jumps to the first frame (forwards or backwards depending on settings) that contains a waveform that enters that division - as in the following image with overlaid glitch frames:





For History mode, it's pretty easy to make it much more useful (assuming the hardware supports it): better acceleration parameters when using the Universal knob for manual selection of frame. And for Automatic play, replace the Reverse play, Pause, and Forward play buttons with Interval (from fastest play possible to 10 seconds), Operate, and Mode - as in the following image:

[marmad]

marmad, thanks for the great review !

You're welcome, wprak - it was enjoyable to do.

One thing you didn't mention was overall responsiveness of basic operations. Especially with deep memory some scopes can get painfully slow when doing simple things like changing the horizontal/vertical position of a waveform. How do the two compare in that regard ?

They felt quite similar to me - I think they're both decently responsive most of the time. There are some times when the Siglent is a bit more responsive (perhaps due to the fact that the underlying hardware might be faster) - and some times when the Rigol is more responsive (perhaps due to more mature firmware).

For example, when the DSOs are running in Normal mode while using deep memory (e.g. 14M), it's a bit quicker moving the waveform around on the Siglent - but this might be due to the fact that the Siglent is doing less (or no) intensity grading compared to the Rigol. Hard to know for sure until Siglent improves the intensity grading at the slower time bases (which apparently they're working on). When the DSOs are stopped with a 14M waveform, the Rigol is slightly quicker moving around within the waveform.

[rf-loop]

@Fsck:  I've enlarged the table in the first post to include vector waveform update rates for channels 1 & 3 plus all channels simultaneously (1 & 2 & 3 & 4) .
[/quote]

There is Siglent sin(x)x and Sin(x) off. But then Rigol is not both.  Is it so that Rigol have same speed with both settings. 
Then I ask agen why comparison is made so that Siglent is set for display connected dots.
Is it so that Siglent is too good vs Rigol if also it is adjusted for its best waveform update speed.

Is Rigol  speed same with Sin(x)/x on and off? If not, then for fair compare it is good also show it.
Is  Rigol speed same if display mode is select dots or line connected dots?

[marmad]

There is Siglent sin(x)x and Sin(x) off. But then Rigol is not both.  Is it so that Rigol have same speed with both settings.

As mentioned in the video, there is no manual switch on the Rigol DS2000 - it is always sin(x)/x with sample speeds >= 1GSa/s and linear with sample speeds <= 500MSa/s.
 

Then I ask agen why comparison is made so that Siglent is set for display connected dots.

As mentioned in the video - and the last time I answered you - I, personally, use my DSO with vectors turned on most of the time. That is my real world usage, so that's what I'm interested in: i.e. performance, intensity grading, etc, with vectors ON. Perhaps it's different for other people - I couldn't say (you could perhaps start a poll thread to ask what percentage of time people use their DSOs with vectors off and on).

Also, you had already posted in other threads your very detailed table of the waveform update rates using dots on the Siglent - but no one had measured and posted one using vectors (sin(x)/x).
 

Is  Rigol speed same if display mode is select dots or line connected dots?

Again, I posted the Rigol's waveform update rate with dots in the last post answering you. If you compare tables, you can see that the Rigol, unlike the Siglent, has fairly similar speeds regardless of whether it's in dots or vectors mode (although, obviously, a bit faster with dots) - except for 20ns/div with 1 channel.

[rf-loop]

There is Siglent sin(x)x and Sin(x) off. But then Rigol is not both.  Is it so that Rigol have same speed with both settings.

As mentioned in the video, there is no manual switch on the Rigol DS2000 - it is always sin(x)/x with sample speeds >= 1GSa/s and linear with sample speeds <= 500MSa/s.
 

Tnx, this limit in Rigol I did not know. Perhaps they repair it in later FW:s.

[tautech]

I hope that all the bugs in SDS2000 firmware will be fixed in next years. But remember that this scope is not so much new, it was released in November 2013 according to their website. http://www.siglent.com/en/news/detail.aspx?id=100000065214624&nodecode=119002004

The release date is often wrongly quoted and/or misinterpreted.
Yes, selected recipients received the SDS2000 in Nov and Dec 2013. HW 3.3
In no way was it ready for market release at that time as was indicated to us by Siglent.
The first firmware updates were sent to us early in the new year.
I have received 4 to date.
The latest, 35.1 finally enabled all HW options and signaled the full product ready for release. https://www.eevblog.com/forum/testgear/siglent's-new-product-msosds2000-series/msg482629/#msg482629

I think that there is no fine horizontal setting at SDS2000. That is not good.

The Horizontal position adjustment resolution is scaled to suit the timebase selected.
Eg. 1 nS/div timebase horizontal adjustment is in 20pS increments, that is 50 steps/div.
@ 1uS/div the resolution is 20nS.
Adjusted quickly, a 1 div change is less than 1 turn, and slowly several turns.
The response of adjustments have dramatically improved from the early firmware.

Marmad used this adjustment frequently in the video and I don't remember him mentioning any deficiencies.

Is a finer Horizontal position resolution required?

[marmad]

The Horizontal position adjustment resolution is scaled to suit the timebase selected.
Eg. 1 nS/div timebase horizontal adjustment is in 20pS increments, that is 50 steps/div.
@ 1uS/div the resolution is 20nS.
Adjusted quickly, a 1 div change is less than 1 turn, and slowly several turns.

I believe he's referring to a vernier (fine) adjust for the Horizontal scale (timebase), which is missing on the Siglent. But to be honest, it's something I rarely use on the Rigol, so I don't think it matters all that much.

Fine adjustment for the Vertical scale (which the Siglent DOES have) is more important, IMO, since the Vertical scale is normally limited to ~12 or so coarse settings. The Horizontal scale, OTOH, normally has at least 24 coarse settings - and has a total of 32 on the Siglent.

[Hydrawerk]

I believe he's referring to a vernier (fine) adjust for the Horizontal scale

Yes, I am.
-------
Is it possible to use cursors in XY mode? It is not clear from the manual.
-------
What do you think about the AWG? According to the user manual it looks nice.
-------
What is the Trigger Output peak-peak voltage at SDS2000? At DSOX2000 it is 5V, at Rigol DS2000 it is OK (I do not remember...), at GW Instek GDS-2000A it was strangely low, about 150mV. 
http://youtu.be/-yS5YLWqf_Y?t=5m2s

[marmad]

Is it possible to use cursors in XY mode? It is not clear from the manual.

No.

What do you think about the AWG? According to the user manual it looks nice.

It seems good for a built-in one - although it's the first I've ever used inside a DSO.

What is the Trigger Output peak-peak voltage at SDS2000?

Visible here: https://www.youtube.com/watch?v=zOjLNyzxkTQ&feature=youtu.be&t=1h2m13s

[Hydrawerk]

They will probably add the cursors to XY mode later in the firmware. Even the Tek TDS2000 series do not have this feature.
So the Trig Out level is about 2.5Vpp. That's OK.

[Hydrawerk]

Does this digital trigger work as promised?

[tautech]

I used a borrowed DSO from my dealer (i.e. not a brand new model straight from the factory)

How much longer will you have the borrowed SDS2304?

I haven't looked at the PC interface, Teneyes; perhaps I'll do it over this weekend. It's a little early to know about the possibility of modifying RUU to work with it - we're still waiting for Siglent to publish the SCPI documentation.

The SCPI manual and command set is soon to be released. 
Siglent sent me a pre-release copy today.

[marmad]

How much longer will you have the borrowed SDS2304?

I can keep it for another couple of weeks if necessary. I'm just making another short video today about the beautiful intensity grading of the Siglent - which I didn't due justice to in the original comparison. I'm trying different exposures to try to record the screen as accurately as possible.

[nctnico]

Marmard: I'm wondering what the -3dB bandwidth is on the scope you have.

[marmad]

Marmard: I'm wondering what the -3dB bandwidth is on the scope you have.

I wish I could check but I don't have an RF generator, avalanche pulse generator, etc. Perhaps rf-loop has done it (although maybe his hardware is an older version).

[tautech]

Marmard: I'm wondering what the -3dB bandwidth is on the scope you have.

I wish I could check but I don't have an RF generator, avalanche pulse generator, etc. Perhaps rf-loop has done it (although maybe his hardware is an older version).

It is buried deep in the "Siglents New Product" thread and IIRC was ~320 MHz.
My client checked mine as very similar, however as you correctly state, ours are older HW.

[nctnico]

I check the other channels on my 200MHz oscilloscope and the bandwidth of the channels varies from 310MHz to 340MHz.

[rf-loop]

Perhaps rf-loop has done it (although maybe his hardware is an older version).

it is very early version manufactured  November 2013.

. Attached image about response.
Signal generator HP8644B and checked that level accuracy is much better than its specified limits.
I have used 10MHz steps. First freg is 1MHz, next 11, and then 21 etc. (10MHz steps)
-3dB points is roughly 350MHz and flatness is not only just ok, but good.
In picture most left wide yellow
is 1MHz, then next around half div wide is 11MHz, next is 21, then 31, 41,51 etc.
Middle is 251MHz. (around 50MHz/div)

[marmad]

@tautech and rf-loop:  Doing some more testing with the Siglent, and I can't seem to see any evidence that High Res mode is working. Can either of you confirm with Siglent if it's supposed to be a functional feature yet?

I can't find any info in the datasheet about which timebases it's active on, but if you look at the attached chart from the Rigol you'll see that 200us is about in the middle of it's High Res effective range, with a bandwidth at that timebase of between ~138kHz - 172 kHz. So at that setting, a 10MHz sine wave is very much low-pass filtered, as seen in the attached image. But I don't see any effect on the Siglent (at that timebase - or any other).

It's not a big deal - but I'm curious if it's functional or some setting I have turned on that's interfering with it's implementation.

[Teneyes]

[ So at that setting, a 10MHz sine wave is very much low-pass filtered, as seen in the attached image. But I don't see any effect on the Siglent (at that timebase - or any other).

Does the Siglent detect the 10 MHz from counter and auto shutoff the High Res??

Maybe need different test input , sinewave with a glitch pulse?

[marmad]

Does the Siglent detect the 10 MHz from counter and auto shutoff the High Res??

I'm not sure what the purpose would be to have the DSO decide what should be averaged or not - I just used the 10MHz waveform as an obvious example. But I can't see any effect on any kind of waveform I use.

[nctnico]

Same here. I expect the high-res mode to produce a much thinner line but that doesn't happen.

[tautech]

@tautech and rf-loop:  Doing some more testing with the Siglent, and I can't seem to see any evidence that High Res mode is working. Can either of you confirm with Siglent if it's supposed to be a functional feature yet?
It's not a big deal - but I'm curious if it's functional or some setting I have turned on that's interfering with it's implementation.

From the manual:

High Resolution
This mode uses a kind of ultra-sample technique to average the neighboring points of the
sample waveform to reduce the random noise on the input signal and generate much
smoother waveforms on the screen. This is generally used when the sample rate of the
digital converter is higher than the storage rate of the acquisition memory.
High Resolution mode can be used on both single- shot and repetitive signals and it does
not slow waveform update. This mode limits the oscilloscope's real- time bandwidth
because it effectively acts like a low- pass filter.
Note: “Average” and “High Res” modes use different averaging methods. The former uses
“Waveform Average” and the latter uses “Dot Average”.

I too, do not have the expected result.
We will ask Siglent if High Res is fully functional.