Choosing between entry-level 12-bit DSOs

[tautech]

Performa, thank you for responding. I am afraid that I am still too new to all of this fully to understand. Are you saying that the overshoot is happening in the scope (front end), not actually part of the signal? Again, I know my questions are revealing my ignorance ... but hopefully by asking I will get a little less ignorant!

There are several levels to what Performa01 is trying to convey.

Each scope series is available in a # of BW's so requires a response filter to the max BW in the series then applies some rolloff.
Today this is less done in HW but by SW controlling the PGA that sets the BW for each model.

Filters are never perfect so if overshoot is contained to small amounts it is acceptable.

[Fungus]

Performa, thank you for responding. I am afraid that I am still too new to all of this fully to understand. Are you saying that the overshoot is happening in the scope (front end), not actually part of the signal? Again, I know my questions are revealing my ignorance ... but hopefully by asking I will get a little less ignorant!

It's a complex subject which I'm not going to pretend I fully understand, but... both of those 'scopes are telling you the exact same thing about the circuit:

a) The voltage is reaching the required level, no problem.
b) It's taking at most about 1.4ns to get there (it may be faster but you can't know that).

That's it.

Anybody who tells you that you can infer more than that about the circuit by looking at "ringing" is delusional.

[Performa01]

Are you saying that the overshoot is happening in the scope (front end), not actually part of the signal?

Of course overshoot cannot be ruled out to be present in the original signal already, but in practice the contribution of the DSO frontend is dominating by far. Let's have a look at some possible signal sources:

1. AWGs can generate squarewaves and pulses, but transitions will only be fast if a high bandwidth model with appropriate sample rate is used, i.e. at least a Siglent SDG6000X. At extreme (mostly unspecified) settings we might get some form of pre-, over- and undershoot - actually "Gibbs Ears" - because of a steep FIR-reconstruction filter at the generator output. See this example:

https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg5538565/#msg5538565

2. Digital CMOS circuitry (presumably line drivers) which should not be able to overshoot as they simply switch between Vss and Vdd. On the other hand, the signal will still overshoot if the construction is sloppy with significant parasitic inductances and/or poor probing technique is applied.

3. Analog highspeed OpAmp output with improper HF-consturction techniques and poor output impedance matching can show overshoot because of cable reflections and high supply impedances.

Long story short: you should not see any significant overshoot with serious signal sources like pulse generators or AWGs, but you can never be sure when probing a circuit or using the signal outputs of sloppy DIY-projects.

Just to be sure, you can either have a reference source with known overshoot - or a reference oscilloscope with much higher bandwidth. The following link demonstrates the difference in overshoot of the very same signal with two DSOs that have a bandwidth ratio of 8:1. Simple conclusion here: the better/faster DSO will always show less pulse distortion, i.e. overshoot.

https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg5537667/#msg5537667

EDIT: missing link added.

[awakephd]

Thank you! This is very helpful, and has helped to confirm what I suspected - but didn't know enough to describe or have any clarity about.

[vcor]

Thanks to all the contributors in this (and other posts)! 

Way too many years ago I assembled my first HealthKit Oscilloscope when I was 14.  After my EE degree, I used Tektronix and HP CRT-based scopes but haven’t used one for a long time.  I have a new need for a couple of projects and was impressed by the value available today and narrowed it down to these Rigol and Siglent 12-bit scopes after discarding other bottom-end scopes from Frirsi, Owon, and Hantek from consideration.  I’ve never used products from either company before.

In this post and other review videos, it appears there are quite a few misrepresentations and errors with the Rigol HDO series.  Let me know if I’m wrong.

1)   Rigol never fixes bugs or does updates - Rigol has made 9 releases since August 2023, correcting over 40 issues.

2)   Rigol never adds features after release – So far, they’ve added 4 improvements: Quick on/off, SMB network storage function, improved self-calibration and DHO800 added 5M storage depth.

3)   You can’t save your settings after power down – There is an option to remember your last settings or use the defaults when powering up.

4)   Slow to boot – Now offers a sleep mode with instant power-up so long as the scope remains connected to power.

5)   Math functions are always in a separate window, not the waveform window. There is an option in the Math options to show results in a separate window or along with the waveforms. 

Bode plots on the DHO914S/924S appear to still have issues, although a few tests/users/videos show zero problems. One problem appears to be if the auto gain is too sensitive you may get problems if clipping occurs at higher frequencies. Seems there is a workaround by manually setting the gain of each channel before starting. I’m not sure I’ll ever use Bode plots, but it is a ‘nice to have’.

Silgent SDS800X has had fewer issues but has been on the market for a shorter time.  So far, they have made 3 releases and fixed 16 bugs. They have not added any features since the first release.

[Fungus]

In this post and other review videos, it appears there are quite a few misrepresentations and errors with the Rigol HDO series.

Yep. A lot of people seem to go into Rigol threads just to hate on them.

The Rigol has some advantages, and one of them is that it's extremely hackable. There's improved UIs and an improved fullscreen FFT.

Bode plots on the DHO914S/924S appear to still have issues, although a few tests/users/videos show zero problems. One problem appears to be if the auto gain is too sensitive you may get problems if clipping occurs at higher frequencies. Seems there is a workaround by manually setting the gain of each channel before starting. I’m not sure I’ll ever use Bode plots, but it is a ‘nice to have’.

Similar to the FFT, there's people working in another thread to add Bode plot functions to the DHO804 by controlling an external Signal generator.

[KungFuJosh]

Yup, nothing says "high-quality scope" like the fact that users need to hack it to give it a usable UI and bode functionality. 😉

[shapirus]

There is an option in the Math options to show results in a separate window or along with the waveforms. 

Is it? Was it added in the latest firmware?

[Fungus]

Yup, nothing says "high-quality scope" like the fact that users need to hack it to give it a usable UI and bode functionality. 😉

The general opinion is that even the "standard" UI of the Rigol is better than a Siglent, so... 

[ebastler]

The general opinion is that even the "standard" UI of the Rigol is better than a Siglent, so... 

No, I don't think the consensus is that Rigol's UI is "better". It is easier to get acquainted with than the Siglent, especially for new DSO users. But is rather wasteful on screen space, due to its many graphical trimmings.

I think the Rigol UI works well on the higher-resolution 10" screen it was originally designed for. Rigol did not do a great job scaling it back to the DHO800/900 screen.

[KungFuJosh]

The general opinion is that even the "standard" UI of the Rigol is better than a Siglent, so... 

Primarily among Rigol fanboys. 😉

[ebourg]

5)   Math functions are always in a separate window, not the waveform window. There is an option in the Math options to show results in a separate window or along with the waveforms. 

The display area option for the math functions was added to the first firmware update of the DHO1000/4000 series in April 2023 (the October 2022 user manual hasn't been updated and doesn't mention it),  but the DHO800/900 don't have it unfortunately.

[Fungus]

Are you saying there's literally nothing you'd change about the Siglent UI?

If you could.

(which you can't)

[KungFuJosh]

Are you saying there's literally nothing you'd change about the Siglent UI?

If you could.

(which you can't)

I'd change the logo (to one of mine), and add some custom icons for fun. But in general, the UI is fine with me. There's certainly some functional changes I'd like to see that I posted in the SDS2000X HD feature request thread. Nothing's perfect, but IME, Siglent is a lot closer than its closest competitor. 😉

[Fungus]

I'd change the logo (to one of mine), and add some custom icons for fun. But in general, the UI is fine with me.

With the Rigol: I'd change the logo (to one of mine), and add some custom icons for fun. But in general, the UI is fine with me.

[vcor]

The display area option for the math functions was added to the first firmware update of the DHO1000/4000 series in April 2023 (the October 2022 user manual hasn't been updated and doesn't mention it),  but the DHO800/900 don't have it unfortunately.

I see that now. Thanks!  Perhaps they will add it in a future update, as they have been adding features since the first DHO release.

[2N3055]

The display area option for the math functions was added to the first firmware update of the DHO1000/4000 series in April 2023 (the October 2022 user manual hasn't been updated and doesn't mention it),  but the DHO800/900 don't have it unfortunately.

I see that now. Thanks!  Perhaps they will add it in a future update, as they have been adding features since the first DHO release.

And pray tell, what specifically did they add since they released it?

[mawyatt]

Here's a simple FFT of a 80MHz -120dBv (1mv) signal on the Rigol DHO814 and Siglent SDS814X HD.

Rigol's FFT looks good but doesn't offer an FFT Average Mode (at least we couldn't find this) like the Siglent which shows the advantage of such with 4 and 64 spectral averages. Hopefully Rigol will offer an FFT Average mode soon

We are not interested in getting into the Rigol vs Siglent debate, nor doing any specific test, folks interested in this can get both and do their own comparisons

Best,

[awakephd]

Thanks to all the contributors in this (and other posts)! 

Way too many years ago I assembled my first HealthKit Oscilloscope when I was 14.  After my EE degree, I used Tektronix and HP CRT-based scopes but haven’t used one for a long time.  I have a new need for a couple of projects and was impressed by the value available today and narrowed it down to these Rigol and Siglent 12-bit scopes after discarding other bottom-end scopes from Frirsi, Owon, and Hantek from consideration.  I’ve never used products from either company before.

In this post and other review videos, it appears there are quite a few misrepresentations and errors with the Rigol HDO series.  Let me know if I’m wrong.

1)   Rigol never fixes bugs or does updates - Rigol has made 9 releases since August 2023, correcting over 40 issues.

2)   Rigol never adds features after release – So far, they’ve added 4 improvements: Quick on/off, SMB network storage function, improved self-calibration and DHO800 added 5M storage depth.

3)   You can’t save your settings after power down – There is an option to remember your last settings or use the defaults when powering up.

4)   Slow to boot – Now offers a sleep mode with instant power-up so long as the scope remains connected to power.

5)   Math functions are always in a separate window, not the waveform window. There is an option in the Math options to show results in a separate window or along with the waveforms. 

Bode plots on the DHO914S/924S appear to still have issues, although a few tests/users/videos show zero problems. One problem appears to be if the auto gain is too sensitive you may get problems if clipping occurs at higher frequencies. Seems there is a workaround by manually setting the gain of each channel before starting. I’m not sure I’ll ever use Bode plots, but it is a ‘nice to have’.

Silgent SDS800X has had fewer issues but has been on the market for a shorter time.  So far, they have made 3 releases and fixed 16 bugs. They have not added any features since the first release.

Interestingly, none of the numbered items above factored into my decision at all. At the time I made my decision, it was not clear how the scopes would differ in terms of support and updates. The other numbered items weren't even on my radar. I was aware of the Bode plot difference, but wasn't sure whether it was something I would need or want - didn't understand what a Bode plot was well enough fully to understand the discussions about it.

After buying and using the Siglent, I have learned what a Bode plot is (in a class I have been able to sit in on), and have LOVED being able to produce excellent Bode plots at home. I don't know, even now, that I will ever truly need that capability, but then again, I don't know that I truly need a DSO. (Please don't tell my spouse I said that!)

Meanwhile, the thread on the Rigol DHO804 bugs / firmware has made me far more aware of the issue of support and updates; the faux-pas that occurred with their 1.03 firmware and the length of time it took to fix it has made me happy I went with the Siglent. That said, based on what some of the more balanced voices have said in various threads, it is not clear that the bugs (in either scope) would actually have made much if any difference to me as a newbie.

From what I can see, the various arguments for and against fall into four categories:

• There are various "complaints" made against either scope which should probably be discounted, or at least treated as "YMMV" (e.g., the noise of the fan - bothers some, not others).
• Features which one scope has and the other doesn't, which can be overcome relatively easily but may "offend" one's aesthetic sensibilities - e.g., can't use a simple wifi USB adapter on the Siglent but can on the Rigol ... easily overcome with a small, USB-powerd LAN-to-wifi adapter; VESA mount on the Rigol ... easily resolved with a little 3d printing; only one USB port on the Rigol but two on the Siglent ... easily overcome with a USB hub; and so on.
• Features which one scope which the other doesn't, which seem like prime candidates for software upgrades, e.g., Bode plot and FFT averaging on the Rigol, wifi USB adapter on the Siglent.
• Features which one scope has and the other doesn't which cannot be overcome easily or perhaps at all, e.g., ability to power the Rigol with a USB-C power bank, HDMI output on the Rigol, higher sampling rate on the Siglent.

Naturally, each person will weigh these various factors differently. I would say the main reason I went with the Siglent was the sampling rate - that's a "hard number" that, as far as I know, cannot be changed by any software update or hack or so on. I am very happy with my choice. But I strongly suspect that I would have been happy with either, as either of these is such a huge leap beyond my old boat-anchor Tektronix.

[ttelectronic]

For me it was the Sample Rate, Memory and better use of screen space of the Siglent

[vcor]

Ok, some may flame me, but after looking at all the posts in this thread, many videos, and a lot more on these 12-bit scopes, I went with the Rigol DHO914S. 

First here is my reasoning for choosing the DHO914S over a DHO800 (later I’ll talk about the Siglent):

• The 914 has the logic analyzer option.  While some might want to cut up the case and solder in the connector, hack the software, and void the warranty, I wasn’t willing to do that.  Still, my need for LA isn’t immediate, more of a nice to have for the future.  I like the $85 option for a 16-probe 5V/3.3V, as I never work with logic outside that voltage range.
  
  
• The primary reason was getting the AWG.  There appear to be good alternatives in the $180 and up for a stand-alone AWG. Having it built in eliminates one more box and power cord and allows for Bode Plots that I might like to use in the future (still some controversy on Bode!). My primary use case is for the signal generator, not Bode.
  
  
• The third rather silly reason is I like the black faceplate better than the white one.  The lighted indicators and buttons stand out a lot better than on the white DHO800 series.

Now before I went through this entire post, I was thinking of the DHO802/SDS802X, both are super bargains.  Others here convinced me that 4-channels are worth the extra cost.  I can see a few cases where more than 2 channels might be useful.  Most of my work has been with 1 or 2 channels. You often adapt to what you have available!

If there was a DHO804S, I would have likely gone for it.  Yes, I’m aware of the hacks for higher bandwidth, which is cool in its own right, The bandwidth was not an important consideration for me as I’m hard-pressed to think of a need beyond 10 MHz for my use cases.

After a lot of handwringing, I decided to splurge and get the DHO914S.  Next up, why Siglent didn’t quite match up for me, but a close second.

[KungFuJosh]

Having it built in eliminates one more box and power cord and allows for Bode Plots that I might like to use in the future (still some controversy on Bode!).

It's convenient to have it internally, but there might come a day when you buy an external AWG anyway. I primarily used internal AWGs for a long time, but external AWGs give you more power and more options if/when you need them. For now, enjoy the internal one. 😉

[mwb1100]

• The third rather silly reason is I like the black faceplate better than the white one...

It still surprises me how much weight people give to white vs. black on test equipment.  Note that I'm not saying there's anything wrong with giving color weight (even a lot of weight) - I'm just surprised how far up the list it is for so many people.

Thanks for explaining your thought process.  I haven't bought either of the DHO800/DHO900 or SDS800X HD scopes (and that surprises me - I sold my SDS1104X-E in anticipation),  But every time I think I might pull the trigger I have a hard time deciding, though I'll admit I lean a bit toward the SDS804X-HD.

[vcor]

My reasoning for choosing the Rigol DHO914S over the similar Siglent SDS814X with AWG:

Right off, we have some specification differences.  The Rigol has better single-channel bandwidth, while the Siglent is better at 4 channels.  The Rigol has half the sampling rate.  The memory depth is the same.  I’m ignoring the hacks that can offer more bandwidth.  None of these differences matter to me, but may to others depending on the use cases.

For these specific two configurations, the Siglent is also about $90 more expensive.

Going back to the 20 pages of posts earlier on this thread shows that when each is pushed 2-3 times beyond their design limit to 500 MHz, both scopes produce poor results – These are the wrong scopes to be using and a higher-end scope should be considered.

For me the information display is critical – you must be comfortable understanding the settings and accessing features.  IMHO the Siglent looks like something like a 1990s PC app whereas the Rigol has a fresh smartphone design style. The fonts are cleaner and easier to read on Rigol. When enlarged on a PC or external display, the Rigol’s display looks even better.  Not so with the Siglent. Still, this appears to be a major personal choice issue.

Then there is how the UI functions.  I expect if you are used to Siglent’s UI, most don’t want to switch to something more modern. The Auto Setup is a cool feature on both scopes, but why Siglent forces you to confirm the operation each time seems clunky.  I hope they don’t do this for other operations.

The ability to further customize the UI is also so cool on Rigol.  The free add-in to provide a toggle to full-screen waveform display looks nice (I haven’t tried it yet).  Android allows so much more flexibility that doesn’t exist on Siglent.  I’ve written commercial Android apps, so being able to add more features adds a lot of value to me.  One current limitation is the Math functions are always displayed in a separate Window.  Hopefully, Rigol will add the option to combine them in one window as some of their higher-end scopes. On the other hand, I think Siglent doesn’t offer separate Windows as an option.

Next is the AWG option.  On the Siglent, this is an expensive $285 dongle that requires more cables and seems like such a clunky solution.  On the Rigol, it’s fully integrated and only adds about $100 to the cost.  I was on the fence and looked at other standalone AWG equipment.  Seems most of the good 14-bit solutions are in the $180-400 category.  I really didn’t want more boxes to deal with and my needs are well covered by the AWG in either Rigol or Siglent.

A small negative is the external power supply unit on the Rigol.  The Rigol supply and the scope together consume quite a bit less volume than the Siglent, but I doubt anyone cares.  With any equipment over its life the power supply is the most likely to fail, so having it easily replaceable is a plus. The external power also allows powering via a 12v battery or a car (I do some CAN/LIN bus work in EVs).

Small pluses for the Rigol include the ability to install tiny $10 USB WiFi, whereas the Siglent has no native support for WiFi.  The Rigol also offers a HDMI output that can display up to 1920x1080 resolution.  Others have also said the included Rigol probes are better, but I have no way to evaluate that.  The Rigol is also significantly thinner and lighter than the Siglent.

A small plus to me for the Siglent is the black faceplate and the Logic analyzer connection is standard on all models. Sadly, the Siglent LA option is another $425 with no lower-cost third-party option that I’m aware of.

I’m unlikely to win any Siglent converts, and that is fine.  Both do so much more than the scopes I’ve used in the past, at a price point that is truly incredible – one should be very happy with buying either scope.  I’ll try and add some more notes when I do more testing.

[2N3055]

I see some misinformation is being repeated here that need correction.

Rigol DHO900 does not have less bandwidth at 4 channels. That IS the problem.
And if you use 4 analog and also digital together, it has sample rate that is almost 4x less than SDS800xHD.

Those two limitations of DHO900 combined, will result in almost 300MHz BW sampled at 156 MS/s.
That is sample rate sufficient only for 60-70 MHz realtime oscilloscope.

Rigol DHO800/900 are having adequate sample rate only when 1 or 2 channels are used. All other combinations will make scope show plainly wrong data.

If those facts are known and understood by customers, and they are happy with these limitations, then Ok.
But saying that they have "less BW at more than 2 channels" is dangerous misinformation.

As for colors, I have 3 scopes that are beige/light gray and 3 that are dark gray/black. They are both exactly the same to me, usability wise.

Also comments about how something looks like 1990s design and something is better to read because it looks like a smartphone I find funny. Obviously it has been forgotten how Windows 3.1 looked. And I personally find smartphone GUI design ugly, limited and nonfunctional. But hey, to each it's own.

Also, "are you sure" for Auto button is there for same reason as " This command will format your disk and delete all data on it. Are you sure you want to proceed ??". Sometimes it takes few minutes to setup the scope with all the measurements and such.
And it is really easy. You just press Auto twice, in a succession. When that became so hard to do? You already have finger on it.
I personally would not mind if I could change Auto button to something else useful. That is my complaint.

What am I saying here?

If person wants to nitpick, any of us, any, will find a mile long list things they would do differently.
Also to some 4 math channels with fully featured FFT is a must.
For some, more important is the colour. Or they never use math but they need HDMI.

That is why I think best is to try to speak about facts. Price, capabilities, bugs, support, color, whatever...
Each person maps it to their need and preferences.