Siglent SDS2000X Plus

[Martin72]

Right, actually batronix offer them without a bundle for free- therefore the sig gen only license will cost 568€ incl. VAT, the all in bundle 1496€
568€ for a one channel AWG....
Apart from this I would like to have the discussion about sense or nonsense of inbuilt awgs separated from here, because it´s a general thing...

[tautech]

Still...... we must thank nctnico for continually bumping this thread about Siglents new offering that challenges all comers in this price class of upper entry level and semipro scopes.

[2N3055]

Right, actually batronix offer them without a bundle for free- therefore the sig gen only license will cost 568€ incl. VAT, the all in bundle 1496€
568€ for a one channel AWG....
Apart from this I would like to have the discussion about sense or nonsense of inbuilt awgs separated from here, because it´s a general thing...

You're right, I agree. I got carried away, sorry..

[kcbrown]

I'm obviously responding to something that is rather old, but I can't help myself. 

But it still needs the zoom crutch. Everyone can walk faster without a crutch.

What you call a "crutch", others clearly consider to be an advantage.

The scope decodes everything in the capture (subject to the limitations that were mentioned).

One capability that the Siglent approach has which other scopes might not have (and I can't say whether they have it or not.  The Rigol DS-1054Z certainly doesn't) is to make it possible to place the trigger point outside of the captured area.  From what I can tell, you can place the trigger point an arbitrary amount of time prior to the capture, and up to the right edge of the screen, which is the end of the capture buffer.

Also decoding only what is on screen gives you the problem that it is nearly impossible to correlate messages which may be in different parts of the acquisition memory. Get a Keysight or R&S scope on your bench and see how easy it is to be able to go through the entire record with timestamps relative to each packet.

In light of zoom mode, this deserves some explanation.  Exactly what are you referring to here?   What sort of correlation are you talking about?  In the Siglent, cursors can be placed anywhere within the capture -- they aren't limited to the zoomed portion of the screen.  This makes it possible to see the relative distance between any two portions of the capture.

Same for the automatic memory length. For the millionth time: limiting the memory size to the visible screen only is a serious productivity limitation. There is no arguing around that other than using yet another crutch (going back & forth between time/div settings). Everyone can walk faster without a crutch.

I'm sure it's a productivity limitation for some and a productivity enhancement for others.  With the approach Siglent is taking, you always know exactly what you're getting.  There is absolutely no ambiguity, no guesswork involved, no off the cuff computation required.  The time captured is the time represented by the screen, period.

There is one disadvantage to Siglent's approach, and it can be a significant one: the mask test operates only on the full screen, and will not operate in zoom mode.  This means that mask testing cannot be applied to a subset of the capture the way it is on other scopes, and this means you can't have it stop the scope upon failure and then examine some other portion of the waveform that is off-screen.   This would be easy enough for Siglent to fix: they'd merely need to make it possible to define and operate the mask against the zoomed portion of the screen.

Remember I used to own an SDS2000 and the automatic memory selection annoyed the hell out of me.

Yeah, but you're used to using scopes that don't operate the way the Siglent does, right?  How much of your annoyance is the result of what you're used to, versus the actual requirements of operating the scope?

Reasoning that none of the customers has asked for it is more indicative of the kind of customers Siglent has than automatic memory selection should be standard. Power users like me that need to have an oscilloscope which works efficiently will ask for it (or silently just don't buy the oscilloscopes from Siglent). Siglent has come a long way in the past few years and now they really have to start addressing the productivity issues (at least on the higher end models) in order to really compete with the A-brands.

I would argue that, at the very least, it should be an option -- make it possible for the user to choose whether to use the remaining memory for the automatic capture history or for the capture itself.  Nothing in principle prevents Siglent from making both approaches possible.

In any case, save for the mask test issue, the approach Siglent has taken here is just different.  It doesn't seem to be any worse or better as an objective matter (again, save for the mask test), only in terms of the specific use pattern that a given user brings to the table.  In your case, your use pattern is such that Siglent's approach isn't the right one.  But then, if zoom mode were your default use mode, that might change.  Can't say on that.

[nctnico]

I'm obviously responding to something that is rather old, but I can't help myself. 

But it still needs the zoom crutch. Everyone can walk faster without a crutch.

What you call a "crutch", others clearly consider to be an advantage.

The scope decodes everything in the capture (subject to the limitations that were mentioned).

One capability that the Siglent approach has which other scopes might not have (and I can't say whether they have it or not.  The Rigol DS-1054Z certainly doesn't) is to make it possible to place the trigger point outside of the captured area.  From what I can tell, you can place the trigger point an arbitrary amount of time prior to the capture, and up to the right edge of the screen, which is the end of the capture buffer.

If your reference point is a DS-1054Z then you have a boat load of catching up to do. None of the features on Siglent scopes are special in any way compared to their mainstream competitors. The rest has already been debated so I urge you to just read some other threads and look at Dave's 'zoom out' video.

[Martin72]

Hi,

and look at Dave's 'zoom out' video.

And see what ?

Won´t we let this thing in the suitable thread ?

As I wrote months ago, siglent has this memory thing on the list, with low priority, which I really appreciate.
Because it IS very low priority.

[kcbrown]

If your reference point is a DS-1054Z then you have a boat load of catching up to do.

It's not my "reference point", it's just a scope that I happen to have access to.  It's why I worded things the way I did.

None of the features on Siglent scopes are special in any way compared to their mainstream competitors.

Not even the "always on" history?

The rest has already been debated so I urge you to just read some other threads and look at Dave's 'zoom out' video.

Dave has quite a few videos.  Can you provide a link to the video you're referring to?

I'm fully aware that you can't "zoom out" with the Siglent, due to the way it operates.  But "zoom out" in the main capture window immediately after capture just means that you're not seeing everything in the capture in the first place.  Maybe that matters and maybe it doesn't.  Seems like that would depend entirely on the use case, no?

As I said, I do think it would be useful for Siglent to let the user choose whether to use the remainder of memory for the capture history or to use it for additional captured time before and/or after the visible window.  But save for the effect on the mask test feature, I've not seen any argument that Siglent's approach is a functional showstopper.  That said, I've certainly not read everything that has been said on the subject (not even in this thread -- I'm working on that), so that I've not seen such an argument doesn't mean it hasn't been made.  Feel free to make such an argument here, now, if you're so inclined -- I'd be very interested in hearing it.

[nctnico]

If your reference point is a DS-1054Z then you have a boat load of catching up to do.

It's not my "reference point", it's just a scope that I happen to have access to.  It's why I worded things the way I did.

None of the features on Siglent scopes are special in any way compared to their mainstream competitors.

Not even the "always on" history?

Yup. Standard on Yokogawa oscilloscopes for ages.

The rest has already been debated so I urge you to just read some other threads and look at Dave's 'zoom out' video.

Dave has quite a few videos.  Can you provide a link to the video you're referring to?

[kcbrown]

Not even the "always on" history?

Yup. Standard on Yokogawa oscilloscopes for ages.

Interesting.  Any other brands that have it?  I don't know how common Yokogawa scopes are, but they're certainly not a brand most would generally think of immediately when looking for scopes.  Pricing information seems to be a little difficult to come by, but it appears that their entry level scope (the DLM3022) commands nearly $4800.   

So while Siglent's "always on" history isn't unique, if Yokogawa is the only other manufacturer that has it, then Siglent is in a league of its own with respect to offering this feature in the market segment it plays in.

Dave has quite a few videos.  Can you provide a link to the video you're referring to?

Awesome.  Thanks!

[kcbrown]

I haven't seen anyone else address this, so ...

One example: a while ago I had to debug a SoC design and wanted to check the CPU's core voltage because every now and then it would crash. For that I wanted to see what happened to the core voltage when it is switched. One channel on the signal to the regulator and one on the core voltage. Trigger set to the control signal and go. No idea on what timescale the event is taking place. Turned out that I choose the time/div too short so I couldn't see the entire event where the supply voltage dropped for a while. Since the oscilloscope I used (RTM3004) was on maximum memory depth I could simply turn the time/div  / horizontal position knobs to get more signal. From that single capture I had all the information needed to see what the exact problem was even though I had no idea about what I was going to look for. Recapturing the event on a different time/div setting would have meant waiting for another crash to happen.

Then you got lucky.  You got lucky because it just happened that the supply voltage drop happened within the time window that your scope was able to record with its memory.

With the Siglent, you would know whether or not you've got the scope configured in such a way as to capture both the core voltage event and the control signal trigger event, assuming that the scope would be capable of doing it at all, because as long as you know the characteristics of the core voltage event, you could set the memory depth to maximum, set the timebase such that the samples per second is still high enough to capture the essence of the core voltage signal (the trigger mechanism apparently always runs at maximum sample rate, so triggering on the control signal would never be a problem), shift the capture view to move the trigger point to the far right, hit "single", and wait.  No guessing.

Another example: every now and then there is a problem with an I2C message. In order to see whether a message is correct or not you need to see the part which goes wrong. Once the message is captured turning the time/div / horizontal position knobs can be used to check what happened leading up to the event. Again, it is unknown on what timescale the event is taking place so more memory is better because recapturing the event isn't easy.

Then do something similar to what I stated above.  The decoding happens on the entirety of the capture (though it seems to be limited in length).  Once you've got a capture, you can do whatever you want with it (though with serial decoding, there seem to be limits in terms of how many things it will decode, so that's a clear limitation that I think needs to be addressed.  But that's not a limitation of the capture mechanism, but rather of the decoding mechanism).

It's really when the scope is running and updating the display that having the capture extend beyond the boundaries of the display is a real advantage, at least in any scenario I can think of.  Again, the exception is mask testing, where having capture extend beyond the boundaries of the display is a major advantage.  Siglent really needs to address that shortcoming, because it's quite significant if you use mask testing to detect faults and need to be able to see what led up to the fault.

Admittedly, the screen can get really busy if you're tracking multiple signals with it.  Zoom mode should be configurable in terms of how much real estate the full capture portion occupies, for exactly that reason.  But that's not an indictment of the capture approach Siglent is taking here, but rather of the lack of flexibility in their zoom mode.

[tautech]

Admittedly, the screen can get really busy if you're tracking multiple signals with it.  Zoom mode should be configurable in terms of how much real estate the full capture portion occupies, for exactly that reason.  But that's not an indictment of the capture approach Siglent is taking here, but rather of the lack of flexibility in their zoom mode.

This can be an issue on a 7" display with a 50/50 split yet far less so with a 10" display and a 25/75 split like SDS2000X Plus and SDS5000X models have......or we can port the display to an even larger one.

Some think this is an issue but there are solutions.

[rf-loop]

So while Siglent's "always on" history isn't unique, if Yokogawa is the only other manufacturer that has it, then Siglent is in a league of its own with respect to offering this feature in the market segment it plays in.

Of course one A² level brand -  Rohde@Schwarz  - have "perhaps" been the main source of the idea for Siglent.

https://scdn.rohde-schwarz.com/ur/pws/dl_downloads/dl_application/application_notes/1td02/1TD02_0e_RTO_History_Mode.pdf

There is normal acquisition mode history "stop and rewind" and then Ultra... something like fast sequence in Siglent

Of course they are not similar and not in same ballpark but idea is same.

The Oscilloscope. (pdf)

History function: looking back in time

Where does the interference pulse in the signal come
from? What caused the loss of a data bit? Finding the real
cause of a problem is often only possible by looking at the
history of a signal sequence.

The R&S®RTO history function provides access to previously
acquired waveforms at the press of a button. This allows
users to analyze the measurement data stored in memory.
They can scroll through the individual acquisitions with the
history player or use the persistence mode to display them
superimposed. This powerful function facilitates searches
for signal faults over all acquisitions. One timestamp per
waveform clearly identifies when events took place. Various
analysis tools such as automatic measurements, FFT,
mask tests and a search function are available for analyzing
past acquisitions.

[nctnico]

I haven't seen anyone else address this, so ...

One example: a while ago I had to debug a SoC design and wanted to check the CPU's core voltage because every now and then it would crash. For that I wanted to see what happened to the core voltage when it is switched. One channel on the signal to the regulator and one on the core voltage. Trigger set to the control signal and go. No idea on what timescale the event is taking place. Turned out that I choose the time/div too short so I couldn't see the entire event where the supply voltage dropped for a while. Since the oscilloscope I used (RTM3004) was on maximum memory depth I could simply turn the time/div  / horizontal position knobs to get more signal. From that single capture I had all the information needed to see what the exact problem was even though I had no idea about what I was going to look for. Recapturing the event on a different time/div setting would have meant waiting for another crash to happen.

Then you got lucky.  You got lucky because it just happened that the supply voltage drop happened within the time window that your scope was able to record with its memory.

With the Siglent, you would know whether or not you've got the scope configured in such a way as to capture both the core voltage event and the control signal trigger event, assuming that the scope would be capable of doing it at all, because as long as you know the characteristics of the core

There is no guessing involved! At some point you have enough experience to just know if the memory depth is enough or not. Just like you know how far your car will go when the fuel indicator says the tank is half full. And at some point you also acquire the experience to know which tools work with you and which tools work against you.

[rf-loop]

I haven't seen anyone else address this, so ...

One example: a while ago I had to debug a SoC design and wanted to check the CPU's core voltage because every now and then it would crash. For that I wanted to see what happened to the core voltage when it is switched. One channel on the signal to the regulator and one on the core voltage. Trigger set to the control signal and go. No idea on what timescale the event is taking place. Turned out that I choose the time/div too short so I couldn't see the entire event where the supply voltage dropped for a while. Since the oscilloscope I used (RTM3004) was on maximum memory depth I could simply turn the time/div  / horizontal position knobs to get more signal. From that single capture I had all the information needed to see what the exact problem was even though I had no idea about what I was going to look for. Recapturing the event on a different time/div setting would have meant waiting for another crash to happen.

Then you got lucky.  You got lucky because it just happened that the supply voltage drop happened within the time window that your scope was able to record with its memory.

With the Siglent, you would know whether or not you've got the scope configured in such a way as to capture both the core voltage event and the control signal trigger event, assuming that the scope would be capable of doing it at all, because as long as you know the characteristics of the core

There is no guessing involved! At some point you have enough experience to just know if the memory depth is enough or not. Just like you know how far your car will go when the fuel indicator says the tank is half full. And at some point you also acquire the experience to know which tools work with you and which tools work against you.

How and why it helps when part of acquisition is hidden outside of screen. 

[nctnico]

How and why it helps when part of acquisition is hidden outside of screen. 

I have explained that many time before how and why that speeds up measurements. For my work an oscilloscope which records only enough to fill the screen makes life much harder because making measurements becomes way more tedious. But if you still don't understand it then it is what it is.

[thinkfat]

Has anyone tried to plot an I/Q constellation with this scope, in XY mode?

[kcbrown]

There is no guessing involved! At some point you have enough experience to just know if the memory depth is enough or not. Just like you know how far your car will go when the fuel indicator says the tank is half full. And at some point you also acquire the experience to know which tools work with you and which tools work against you.

That may be.  But it's still a difference between being able to directly see what you're getting and having to infer, whether through experience or other indicators, what you're going to get.   And pardon me for being a little blunt, but your own wording suggests that you didn't know whether you were going to get what you wanted, because you said "Turned out that I choose the time/div too short so I couldn't see the entire event where the supply voltage dropped for a while".  What else are we to infer but that you used the fact that the capture includes more than what's on the screen as a fallback to save what might otherwise have required you to change the setup and wait for another instance of the problem?

My point clearly isn't that it's impossible to set up a scope with a more traditional acquisition mechanism to capture what you want, only that setting it up for the situation you described is straightforward and clear with Siglent's approach, and thus that the Siglent does not have any notable disadvantages in the situation you were using to illustrate why a more traditional acquisition mechanism is superior.  If anything, the Siglent has the advantage there, precisely because it makes it absolutely clear what you're going to get.

How would you perform the equivalent setup with a Keysight Megazoom scope, when it doesn't even tell you anything about the amount of memory that is being used for the capture, in the absence of an enormous amount of experience with it?

[kde]

Hm.. I am really surprised that Siglent does not have it. Even my old Rigol DS1102E has the "zoom-out" feature. And I use it quite often - I just set up the horizontal system so I can clearly see what I want, but I know that I have a couple of "screens" on the left and on the right if I need them. And I often go there. At least scroll left/right for about one screen. It's obvious that I can capture at more us/div, stop and then zoom in, but it's not that convenient for me.

[Szybkijanek]

Hi, fast question, it's possible view %THD factor from FFT screen? Tek MDO4104C have this option.
" alt="" class="bbc_img" />

[kcbrown]

Hm.. I am really surprised that Siglent does not have it. Even my old Rigol DS1102E has the "zoom-out" feature. And I use it quite often - I just set up the horizontal system so I can clearly see what I want, but I know that I have a couple of "screens" on the left and on the right if I need them. And I often go there. At least scroll left/right for about one screen. It's obvious that I can capture at more us/div, stop and then zoom in, but it's not that convenient for me.

The Siglent approach can certainly be more cumbersome in some cases, to be sure.

I have to wonder how or whether viewpoints on this subject would differ if the earliest DSOs (and, thus, most later ones) had been implemented in the way the current Siglent line is implemented.   If people had gotten used to the notion of "what you see is all you get", would they find the alternative "there's more than what you see" approach to be better, or worse?

It's one thing to prefer one thing over another because the first is clearly superior.  It's quite another to prefer one thing over another just because of familiarity.  As regards this particular characteristic, I don't think either approach is clearly inherently superior -- which one is better just depends on the circumstances.   The debate over it probably wouldn't be happening at all otherwise (well, that said, one shouldn't underestimate the ability or willingness of people to debate anything at all!  ).

[Martin72]

Hi, fast question, it's possible view %THD factor from FFT screen? Tek MDO4104C have this option.

Afaik no, but it´s a nice idea to have this - will put it on the wishlist-thread...

[Elasia]

You dont zoom out on this model by default, you start by zooming in THEN zoom out.

Say you are at 50us but want to see more, if you have zoom kicked on and at 50us with a timebase of 100M Points... you just turn the horizontal dial and... zoom out

cost? screen space of about 1/5th to 1/4th of the viewing area

 Is it the best UI? Is it the best to even explain? No.. but it works

[Szybkijanek]

And that's it, thanks Martin  maybe Tautech will speed it up.

[kcbrown]

I previously posted a message here about a bug in the SDS-1204X-E serial trigger.  Turns out that it wasn't a bug in the trigger at all.   I feel rather ashamed of that.  Anyway, I've deleted the message, but thought I should at least explain why I did so.

You might be wondering why I posted something about it here.  I did so because I thought perhaps the 2000X Plus had the same bug, and was curious if that was the case.

In any case, it was user error (or, rather, failure of the user to manually check the waveform contents to ensure they matched what the decoder was claiming about them).  Problem solved.  Sorry for wasting people's time on that, however little that might have been ...

[kcbrown]

With respect to the use of zoom, on the 1000X-E series, it does not work with the digital channels.   Given how the scope works, this is, of course, a massive problem, because it forces one to stop the scope in order to zoom in and look around.

Does the 2000X Plus have this limitation with its digital channels?   If not, then that would make it a very tempting upgrade.