Author Topic: the eternal question about rigol vs siglent  (Read 29399 times)

0 Members and 3 Guests are viewing this topic.

Offline gf

  • Frequent Contributor
  • **
  • Posts: 393
  • Country: de
Re: the eternal question about rigol vs siglent
« Reply #100 on: March 25, 2019, 07:17:14 pm »
When I trigger a noisy signal, and the trigger is basically stable, but the displayed waveform still jitters a little bit horizontally around the nominal trigger point, then I'm inclined to say that I can see this jittering.
But it's not, it's the trigger re-triggering on a non-repetitive waveform.
Perfectly normal.

To address this we either change the Holdoff and/or engage other trigger conditions to get stable triggering.
A not perfectly stable waveform doesn't matter to much for general scope work but for accurate consistent measurements it does.

I mean something different. Assume a sine wave signal. If it is noise-free then its (say) rising edge always crosses the trigger level at (almost) exactly the same phase angle of the sine wave. If I add a significant amount of noise, then the edge of the noisy sine wave signal is no longer supposed to cross the trigger level at exactly the same phase angle of the sine wave at each acquisition. The comparision of the noisy sine wave edge with a given trigger level eventually "converts" amplitude noise into trigger-jitter. Steep edges are certainly less affected than edges with only a modest slope.
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 3920
  • Country: de
Re: the eternal question about rigol vs siglent
« Reply #101 on: March 25, 2019, 07:26:53 pm »
I mean something different. Assume a sine wave signal. If it is noise-free then its (say) rising edge always crosses the trigger level at (almost) exactly the same phase angle of the sine wave. If I add a significant amount of noise, then the edge of the noisy sine wave signal is no longer supposed to cross the trigger level at exactly the same phase angle of the sine wave at each acquisition. The comparision of the noisy sine wave edge with a given trigger level eventually "converts" amplitude noise into trigger-jitter. Steep edges are certainly less affected than edges with only a modest slope.

If I understand your description correctly, in the situation you describe the sine signal should show some jitter on the screen, right? You have added noise to your signal, so the point in time when the signal crosses the trigger threshold will vary; this will not always happen at the same phase angle of the sine wave.

I thought we were discussing imperfections of real-world scopes here, but that does not seem to be what you are talking about here. Or did I misunderstand? If you had an "ideal", perfect scope, what would you expect to observe if you feed it a signal as described?
 

Offline gf

  • Frequent Contributor
  • **
  • Posts: 393
  • Country: de
Re: the eternal question about rigol vs siglent
« Reply #102 on: March 25, 2019, 07:33:05 pm »
Here you go, Have fun, At shorter Tdiv the samples start decreasing, but feel it covers what your asking for.

I'd like to ask, how is the dB scale of these FFT plots supposed to be interpreted?
Or in other words, what is the 0 dB reference level?
 

Offline gf

  • Frequent Contributor
  • **
  • Posts: 393
  • Country: de
Re: the eternal question about rigol vs siglent
« Reply #103 on: March 25, 2019, 07:37:43 pm »
If I understand your description correctly, in the situation you describe the sine signal should show some jitter on the screen, right? You have added noise to your signal, so the point in time when the signal crosses the trigger threshold will vary; this will not always happen at the same phase angle of the sine wave.

I thought we were discussing imperfections of real-world scopes here, but that does not seem to be what you are talking about here. Or did I misunderstand? If you had an "ideal", perfect scope, what would you expect to observe if you feed it a signal as described?

Let me very briefly summarize the whole story I wanted to tell:

1) the "imperfection" is noise (not I, but the scope adds it)
2) noise leads to trigger-jitter
3) averaging time-shifted acquisitions (due to trigger-jitter) turns averaging into a convolution (-> low-pass effect)

EDIT:

The origin was

...Averaging just serves as a low pass filter...

where I answered that an averaging operation is not a low-pass per se, but indirectly it can be...

« Last Edit: March 25, 2019, 07:55:43 pm by gf »
 
The following users thanked this post: ebastler

Offline gf

  • Frequent Contributor
  • **
  • Posts: 393
  • Country: de
Re: the eternal question about rigol vs siglent
« Reply #104 on: March 25, 2019, 08:40:50 pm »
The issue with the spectrum @gf posted was the variance of the spectral estimate was too large to show the problem.

Here is a better one, calculated offline with Octave from 64k points @ 2.5MSPS, 2 mV/div, normal acquisition mode, no BW limit.
0 dB = highest peak in the spectrum, which is ~125 uV DC (offset error)
Frequency scale is in MHz.

I have no idea where the 500kHz spur is coming from.
I guess (but I'm not sure) that the spur at ~1.9 kHz might come from the (PWM-driven) offset control feeding into the frontent.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3560
  • Country: cn
  • Born with DLL21 in hand
Re: the eternal question about rigol vs siglent
« Reply #105 on: March 25, 2019, 09:48:57 pm »
Here you go, Have fun, At shorter Tdiv the samples start decreasing, but feel it covers what your asking for.

I'd like to ask, how is the dB scale of these FFT plots supposed to be interpreted?
Or in other words, what is the 0 dB reference level?

In @Rerouter images it looks like he have selected  dBVrms.
So "0dB" is 1Vrms

If select external load impedance (it can adjust between 1ohm to 1Mohm in SDS1000X-E)  then there can use dBm  "0dB" is 1mW (In SDS1000X-E impedance (external load) can set from 1ohm to 1Mohm.

I hope Siglent add these units to display so that they are visible on screen independent of what bottom screen menu is selected.

Images from 2017 manufactured SDS1104X-E mod1204X-E.
If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #106 on: March 25, 2019, 10:42:43 pm »
Let me very briefly summarize the whole story I wanted to tell:

1) the "imperfection" is noise (not I, but the scope adds it)
2) noise leads to trigger-jitter
3) averaging time-shifted acquisitions (due to trigger-jitter) turns averaging into a convolution (-> low-pass effect)

Nobody's denying that but in practice it works quite well.  :)
 

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #107 on: March 26, 2019, 12:43:29 am »

And you did not even note what they did not tell... and it was point (of course)

I think it is also important to observe what has not been said.

Is it meaningless that Rigol do not give any info that there is no true 1mV or 2mV/div but they are magnification from 5mV/div range.

But then you take example how Decent manufacturers like Keysight are good because they tell.

Perhaps you forget this your own or do you tell now that all this is also meaningless.
While I love how you're desperately attempting to shift attention away from how crappy Siglent is, it ain't working. I said both Rigol and Siglent are crap at making datasheets. If I was talking to a Rigol distributor I'd mention the flaws in their datasheet, though I must say they at least seem to be capable of performing an error calculation, didn't add a ridiculous 500 µV/div mode to look better and stuck to a realistic limit for the frontend. This is once again a demonstration why you won't find any Rigol or Siglent gear in professional environments, a company that realised what it was building wouldn't try to put such a mode in their scopes in the first place. Want to know why? Calculate the thermal noise a regular X10 probe throws off at room temperature and you might understand why microvolt level measurements are a bit futile...

That doesn't stop it from being trotted out in every single "Rigol vs. Siglent" thread.

As we've seen today though, the Rigol doesn't have as much noise as the Siglent boys like to pretend. Their famous screenshot turned out to be just a pathological comination of the Rigol decimation combined with measurements based on screen date. Twist the timebase knob to a more "natural" value and it vanishes

(which most people will do, as soon as you see noise on screen the natural temptation is to zoom in a bit for a closer look).
Of course they keep highlighting that specific aspect, because they can't actually beat R&S, GW, HPAK, Tek, Lecroy, etc. at performance. So instead it's far easier to just come up with a crappy test like this which will convince their main non-educational customer base (Arduino nutcases). But I'm fairly certain the timebase stability of the Rigols I've had the "pleasure" of using at a local hackerspace doesn't actually meet the +/- 25ppm spec, or at least it was way worse than any HPAKs or Teks I've ever used.

Ok, I'll pay there, so 5mV input scale,  default VGAC is x17.8, so 89mV scale, the ADC also has a x50 internal amplifier, but we only need to use it with a gain of about x20 to get a scale of 1.78V, seems this may be how they have done it.
Most likely it's really just a digital multiplication. If you actually add it in properly it's going to make for one heck of a scope frontend design challenge. There's a reason the mainstream manufacturers don't bother with it...

Averaging makes only sense, IMO, if the signal is periodic and if you have a stable trigger. Then it decreases the random noise and retains the non-random components of the signal.
Leave the IMHO out. Averaging will only work well on a purely periodic signal.
I find it also works reasonably well to catch certain types of periodic glitches in a high-speed bus, but then it needs to be able to trigger on the bus signals correctly.

That and the Bode plots of crystal resonance will make it easy to hold several Chinese OEM's feet in the fire.
Not really, if you get to deal with someone who knows their field they'll just point out the potential measurement error on the scope and state that it's not proof of component quality.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3560
  • Country: cn
  • Born with DLL21 in hand
Re: the eternal question about rigol vs siglent
« Reply #108 on: March 26, 2019, 07:46:11 am »
I said both Rigol and Siglent are crap at making datasheets. If I was talking to a Rigol distributor I'd mention the flaws in their datasheet, though I must say they at least seem to be capable of performing an error calculation, didn't add a ridiculous 500 µV/div mode to look better and stuck to a realistic limit for the frontend. This is once again a demonstration why you won't find any Rigol or Siglent gear in professional environments, a company that realised what it was building wouldn't try to put such a mode in their scopes in the first place. Want to know why? Calculate the thermal noise a regular X10 probe throws off at room temperature and you might understand why microvolt level measurements are a bit futile...

Futile and futile...  of course I have knocked my head to thermal noise when I was young and play with some radio's in 60's.

Of course with these entry level scopes, example Siglent, really can not accurate analyze signals in microvolt range - naturally and because basic level fundamentals. All know this and no one try talk that impossible is possible.

But, still with it and some others can do something more than nothing.

Take lowest price range Keysight and try this. I can tell to you, result is nada. But still with this cheap entry level scope can see this. Try with Keysight, result, nothing.
Also previously shown FRA example with some xtal, try with double priced Keysight. Result: nothing.

And if you think A brand did not implement 500uV/div to 1Mohm scope front end...  bullshit.
Take example just one poor example, old TEK 2225
(more things can find in this older thread about 500uV/div  scopes)



futile and futile...try it with same price class Keysight, not even if you double or triple this price..


 but this can do in real life and there is many oscilloscopes in this price class or even higher what can not do even this. Of course accuracy, or is it better say inaccuracy,  is what it is but still this give lot of more than nothing. Watching 3.3V supply rail level (this is why DC coupling and fixed vertical offset mode) and some details in it. 10x probe just because you said it.


« Last Edit: March 26, 2019, 03:27:55 pm by rf-loop »
If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 21419
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: the eternal question about rigol vs siglent
« Reply #109 on: March 26, 2019, 08:21:20 am »
............ a ridiculous 500 µV/div mode to look better and stuck to a realistic limit for the frontend. This is once again a demonstration why you won't find any Rigol or Siglent gear in professional environments, a company that realised what it was building wouldn't try to put such a mode in their scopes in the first place. Want to know why? Calculate the thermal noise a regular X10 probe throws off at room temperature and you might understand why microvolt level measurements are a bit futile...
Oh, you are a one eyed SOB aren't you ?
You remind me of many ppls in the 70's and 80's that would never buy a Jap car 'cause they're shit, right ?
I was one of them back then but later my eyes opened, maybe there is still hope for you.


From an 'A' brand 50 MHz DSO @ 2x price of SDS1104X-E:

Bandwidth (–3 dB) 1, 2 .......................50-200 MHz models
Input sensitivity range 3 500 μV/div to 10 V/div  :-+
Maximum input voltage 150 Vrms, 200 Vpk  :--

 :blah:
 :blah:

In small print:
1. Denotes warranted specifications........
2. For 1 mV/div to 10 V/div settings, bandwidth is 20 MHz at the 500 μV/div setting.
3. 500 μV/div is a magnification of 1 mV/div setting.



The 100-200 MHz X-E's offer the full BW @ 500uV/div and it's not a magnification of some lower sensitivity range hence it shows more noise and therefore allows the operator to make decisions on best how to reduce displayed noise, not some designer hidden automatic BW limit that thinks it knows better than the competent user.

Like farmers who know how to sift the wheat from the chaff, a good EE should know how to get the info they need from a dirty waveform.

And what's with the Yanks, only 200Vpk inputs.  :wtf:
Whereas X-E's in fact all Siglent DSO's are 400Vpk rated.
Right, Siglent have no idea what they're doing........Right ?

Avid Rabid Hobbyist
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #110 on: March 26, 2019, 08:30:29 am »
Like farmers who know how to sift the wheat from the chaff, a good EE should know how to get the info they need from a dirty waveform.

Weird that you think 100% of Siglent owners are capable of this.

(...and that 0% of Rigol owners are)
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 21419
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: the eternal question about rigol vs siglent
« Reply #111 on: March 26, 2019, 08:40:27 am »
Like farmers who know how to sift the wheat from the chaff, a good EE should know how to get the info they need from a dirty waveform.

Weird that you think 100% of Siglent owners are capable of this.

(...and that 0% of Rigol owners are)
Refresher courses for you:

https://youtu.be/Znwp0pK8Tzk

Then Pt 2:

https://youtu.be/l5Cts5nPpcA
Avid Rabid Hobbyist
 

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #112 on: March 26, 2019, 10:09:41 am »
You know, this would have been easier on both of us if you'd have shut up and known when to stop your sales pitch.

Futile and futile...  of course I have knocked my head to thermal noise when I was young and play with some radio's in 60's.

Of course with these entry level scopes, example Siglent, really can not accurate analyze signals in microvolt range - naturally and because basic level fundamentals. All know this and no one try talk that impossible is possible.
That 9 MOhm resistance in the probe tip is suddenly non-existent for Siglent and only for Siglent? And it doesn't have a voltage noise in the microvolt range attached to it? Tell me more about these magical flux capacitor super conducting scope probe tips. And then I haven't started about the realistic lower limits for a 1 MOhm scope frontend.

But, still with it and some others can do something more than nothing.

Take lowest price range Keysight and try this. I can tell to you, result is nada. But still with this cheap entry level scope can see this. Try with Keysight, result, nothing.
Also previously shown FRA example with some xtal, try with double priced Keysight. Result: nothing.
Keysight will not advertise things they know they can't deliver on.

And if you think A brand did not implement 500uV/div to 1Mohm scope front end...  bullshit.
Take example just one poor example, old TEK 2225
(more things can find in this older thread about 500uV/div  scopes)

futile and futile...try it with same price class Keysight, not even if you double or triple this price..
You're going to use the Tektronix 2225 as example?  :-DD  So, before we continue this ridiculous use of a very specific scope as example, you are aware that the 2225 has a bandwidth of only 50 MHz which reduced to 5 MHz when you turned on the x10 amplifier? At least mention a semi-recent Keysight scope like your colleague did. Mind you, Keysight actually was smart enough to enable the bandwidth limiter for this "magnification".

And it's not because the manufacturer puts "it can do <something>"  on the box that it is either meaningful or true. Stop the damned sales pitch or at the very least admit you're associated with Siglent at this point, because it's quite obvious. All you have to do to sell scopes is say "look, we built something with a usable interface at this price point and it can do this", stop trying to add on features that don't actually work properly.


 but this can do in real life and there is many oscilloscopes in this price class or even higher what can not do even this. Of course accuracy, or is it better say inaccuracy,  is what it is but still this give lot of more than nothing. Watching 3.3V supply rail level (this is why DC coupling and fixed vertical offset mode) and some details in it. 10x probe just because you said it.
[/quote]
You do know this is quite meaningless right? That count at the side of the table betrays you, the scope is averaging measurements over multiple waveforms. If you had read the previous pages you'd have seen I posted screenshots of taking a DSO-X 2012A down to tens of microvolt of RMS voltage noise using exactly the same method. Possible, yes, practical, not really. So please, just stop it, this is getting ridiculous.

Oh, you are a one eyed SOB aren't you ?
You remind me of many ppls in the 70's and 80's that would never buy a Jap car 'cause they're shit, right ?
I was one of them back then but later my eyes opened, maybe there is still hope for you.
My eyes are fine, but clearly you think that our analytical thinking skills are not. Rigol and Siglent can build decent scopes when they're manufacturing them for another manufacturer. Sadly they still have to realise that the market for engineering lab equipment tends to be aimed at engineers. So unlike consumer products, you're not going to sell them by lying out of your arse and hoping they only notice after purchase.

From an 'A' brand 50 MHz DSO @ 2x price of SDS1104X-E:

Bandwidth (–3 dB) 1, 2 .......................50-200 MHz models
Input sensitivity range 3 500 μV/div to 10 V/div  :-+
Maximum input voltage 150 Vrms, 200 Vpk  :--

 :blah:
 :blah:

In small print:
1. Denotes warranted specifications........
2. For 1 mV/div to 10 V/div settings, bandwidth is 20 MHz at the 500 μV/div setting.
3. 500 μV/div is a magnification of 1 mV/div setting.



The 100-200 MHz X-E's offer the full BW @ 500uV/div and it's not a magnification of some lower sensitivity range hence it shows more noise and therefore allows the operator to make decisions on best how to reduce displayed noise, not some designer hidden automatic BW limit that thinks it knows better than the competent user.
Thank you for demonstrating my point for me, note the footnotes on the datasheet: Not only do they admit it's simply a multiplication of the 1 mV range, they also enable the bandwidth limiter. Because something magical happens when you calculate sqrt(4 k T R dF), do you see that "dF" factor in there? Thermal noise is a function of bandwidth... And guess what, if you calculate the thermal noise it seems to magically coincide with the noise voltage you get on a scope input, I wonder why that is?  |O  In other words, you'll find that the 500 µV/div mode on a keysight is slightly more useful than on the Siglent because they enable the bandwidth limiter for you, but you'll still need averaging or high-resolution mode to get anything useful out of it.

So basically what you call "hiding from the user" is what competent engineers will call "understanding the physical limitations of our universe".

Like farmers who know how to sift the wheat from the chaff, a good EE should know how to get the info they need from a dirty waveform.
Not an excuse for lying on datasheets.

And what's with the Yanks, only 200Vpk inputs.  :wtf:
Whereas X-E's in fact all Siglent DSO's are 400Vpk rated.
Right, Siglent have no idea what they're doing........Right ?
Aaaaand, again with the lies. Most modern scopes are built identically for all markets through the magic of switch mode power supplies. So in practice that means most are rated for at least 240V RMS on the input (339Vpk). And all the semi-recent HPAKs sitting on my desk do 300Vrms (425Vpk). So what were you saying?

All you're doing with this type of response is proving my initial point, just please be honest for once and actually say what the scopes can do instead of cherry picking results.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #113 on: March 26, 2019, 10:18:08 am »
Stop the damned sales pitch or at the very least admit you're associated with Siglent at this point, because it's quite obvious.

At least he's not hiding that part...


« Last Edit: March 26, 2019, 10:26:38 am by Fungus »
 

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #114 on: March 26, 2019, 10:23:48 am »
And it's not because the manufacturer puts "it can do <something>"  on the box that it is either meaningful or true. Stop the damned sales pitch or at the very least admit you're associated with Siglent at this point, because it's quite obvious.

He's not hiding that part...



Yeah, I know. I was talking about rf-loops in that part though. :P
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #115 on: March 26, 2019, 10:28:04 am »
Yeah, I know. I was talking about rf-loops in that part though. :P

Oh, so you were.  :-X

Yeah, I don't get rf-loop's angle. He seems to have enough knowledge to know that he's posting rubbish but he still does it anyway.
« Last Edit: March 26, 2019, 10:30:13 am by Fungus »
 

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #116 on: March 26, 2019, 10:33:42 am »
Yeah, I know. I was talking about rf-loops in that part though. :P

Oh, so you were.  :-X

Yeah, I don't get rf-loop's angle. He seems to have enough knowledge to know that he's posting rubbish but he still does it anyway.
I suspect Siglent pays them a fee per post. I'd understand in that case, heck if they pay enough I'll do it as well :P 
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3560
  • Country: cn
  • Born with DLL21 in hand
Re: the eternal question about rigol vs siglent
« Reply #117 on: March 26, 2019, 11:37:22 am »
Yeah, I know. I was talking about rf-loops in that part though. :P

Oh, so you were.  :-X

Yeah, I don't get rf-loop's angle. He seems to have enough knowledge to know that he's posting rubbish but he still does it anyway.
I suspect Siglent pays them a fee per post. I'd understand in that case, heck if they pay enough I'll do it as well :P

It is typical that after technical arguments or examples in in real life and so on ends then these  AdHoc style starts.

Booth of you, Rigol expert Fungus and Keysight's HalFET  can show in practice with they scopes how they perform example with roughly same signal and setup in my last image   Except that I understand why not... specially @Fungus because his oscilloscope is also in this thread topic. How it performs.

@HalFET

Quote
If you had read the previous pages you'd have seen I posted screenshots of taking a DSO-X 2012A down to tens of microvolt of RMS voltage noise using exactly the same method.
What is this doing under my image as some kind of "comment". I have seen your nonsense DSO-X images and they really have nothing to do with this. Here is captured signal. (Including tiny detail:  measured rough DC level using offset DAC)
If you do not really understand what all you see in image - well it is not my head ache. Btw, it looks that only what you have seen in image is this "count" and just this measurement table is perhaps most futile thing in the whole image.
If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #118 on: March 26, 2019, 12:14:17 pm »
Booth of you, Rigol expert Fungus and Keysight's HalFET  can show in practice with they scopes how they perform example with roughly same signal and setup in my last image 

Why should I even care?

 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 3920
  • Country: de
Re: the eternal question about rigol vs siglent
« Reply #119 on: March 26, 2019, 12:27:23 pm »
Booth of you, Rigol expert Fungus and Keysight's HalFET  can show in practice with they scopes how they perform example with roughly same signal and setup in my last image 

Why should I even care?

Because you are happy to provide a service to the community here? If the only thing it does is to put an end to these "Rigol-vs-Siglent entry level scopes" debates once and for all, it would be a big service to the community...  ;)
 
The following users thanked this post: nctnico, Aidanator7000

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #120 on: March 26, 2019, 12:58:52 pm »
eblaster, this will never end, because they cherry pick their demonstration cases. For example, here he went for a 25 Ohm signal generator output with settings beyond what most signal generators can do. Furthermore he selected an offset voltage above 2V, because that's the limit that most manufacturers will allow as offset voltage for the mV ranges (though there are easy tricks around that one). Then he selected a 10 mV pulse, well knowing that many signal generators can't generate clean 100 ns pulses to begin with, so what you get is noisy by default. The images he showed is mostly a demonstration of how good his signal generator is, it doesn't actually show the performance of the scope. Of course, most electronics engineers are aware of these things, but that's not their intended market sadly.

It is typical that after technical arguments or examples in in real life and so on ends then these  AdHoc style starts.
Which technical arguments? Highly selective screenshots? Do I really have to make a screenshot of a DSOX-2012A measuring a 500 MHz sinewave and then ask you to do the same with a 100 MHz bandwidth Siglent?  :popcorn:

Booth of you, Rigol expert Fungus and Keysight's HalFET  can show in practice with they scopes how they perform example with roughly same signal and setup in my last image   Except that I understand why not... specially @Fungus because his oscilloscope is also in this thread topic. How it performs.
I work for Keysight now? Does this mean I got the keysight job I wanted anyway? That actually brings up a sad memory, I was contacted by a Keysight recruiter a few months ago, but due to the messed up way LinkedIn works she didn't see my response. So by the time she saw the position had already been filled.  :(  But that sad story aside, interesting how you don't deny working for or being associated with Siglent.  ::)

And I love how you selected a challenge that is actually hard on the signal generator instead of the scope.  Which signal generator did you use? Most common signal generators (i.e. a Rigol DG1022) can't do what you asked for, so by default it's hard to "disprove". I especially like the touch of lowering it to 25 Ohm impedance.  As a result I did have to lower the offset voltage due to the limits of my desktop signal generator. But as you can see, a low-end DSO-X 2012A can trigger on such a signal and measure it. Used for this: an old Wavetek signal generator, DSO-X2012A with a N2862A 10:1 probe. Note that the noise is actually due to the inadequate nature of the signal generator for this job. If it was the noise from the scope averaging 8000 waveforms would remove it entirely. Are we now done with the silly mock measurements or is this going to continue?

@HalFET
What is this doing under my image as some kind of "comment". I have seen your nonsense DSO-X images and they really have nothing to do with this. Here is captured signal. (Including tiny detail:  measured rough DC level using offset DAC)
If you do not really understand what all you see in image - well it is not my head ache. Btw, it looks that only what you have seen in image is this "count" and just this measurement table is perhaps most futile thing in the whole image.
So you entirely deny that it says?
Code: [Select]
Sa 1.00 GSa/s
Curr 65536 pts
delta f = 15.26 kHz
AVG = 4
I'm fairly certain that "AVG" stands for averaging... Furthermore, claiming measurement accuracy is all fun and games, but if you don't understand why averaging over 567 periods is significant, do the math please. It significantly lowers the measurement error.

And I am well aware the images I posted were ridiculous, that was very much the point of them. I essentially did the same you folks did, the only difference is that I showed what I was actually doing.
« Last Edit: March 26, 2019, 01:19:47 pm by HalFET »
 
The following users thanked this post: ebastler

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #121 on: March 26, 2019, 01:05:16 pm »
And just to add insult to injury, since the R&S SMH was running anyway, I hooked it up to the DSO-X 2012A to a 535 MHz sinewave. Please note that the oscilloscope's internal frequency reference is stable enough to match that of a signal generator driven of a GPS disciplined rubidium clock. Furthermore, it can measure a 5.8 mV signal at this frequency! Even the scope's internal DVM works at this frequency!  |O
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3560
  • Country: cn
  • Born with DLL21 in hand
Re: the eternal question about rigol vs siglent
« Reply #122 on: March 26, 2019, 03:22:40 pm »
eblaster, this will never end, because they cherry pick their demonstration cases. For example, here he went for a 25 Ohm signal generator output with settings beyond what most signal generators can do. Furthermore he selected an offset voltage above 2V, because that's the limit that most manufacturers will allow as offset voltage for the mV ranges (though there are easy tricks around that one). Then he selected a 10 mV pulse, well knowing that many signal generators can't generate clean 100 ns pulses to begin with, so what you get is noisy by default. The images he showed is mostly a demonstration of how good his signal generator is, it doesn't actually show the performance of the scope. Of course, most electronics engineers are aware of these things, but that's not their intended market sadly.

It is typical that after technical arguments or examples in in real life and so on ends then these  AdHoc style starts.
Which technical arguments? Highly selective screenshots? Do I really have to make a screenshot of a DSOX-2012A measuring a 500 MHz sinewave and then ask you to do the same with a 100 MHz bandwidth Siglent?  :popcorn:

Booth of you, Rigol expert Fungus and Keysight's HalFET  can show in practice with they scopes how they perform example with roughly same signal and setup in my last image   Except that I understand why not... specially @Fungus because his oscilloscope is also in this thread topic. How it performs.
I work for Keysight now? Does this mean I got the keysight job I wanted anyway? That actually brings up a sad memory, I was contacted by a Keysight recruiter a few months ago, but due to the messed up way LinkedIn works she didn't see my response. So by the time she saw the position had already been filled.  :(  But that sad story aside, interesting how you don't deny working for or being associated with Siglent.  ::)

And I love how you selected a challenge that is actually hard on the signal generator instead of the scope.  Which signal generator did you use? Most common signal generators (i.e. a Rigol DG1022) can't do what you asked for, so by default it's hard to "disprove". I especially like the touch of lowering it to 25 Ohm impedance.  As a result I did have to lower the offset voltage due to the limits of my desktop signal generator. But as you can see, a low-end DSO-X 2012A can trigger on such a signal and measure it. Used for this: an old Wavetek signal generator, DSO-X2012A with a N2862A 10:1 probe. Note that the noise is actually due to the inadequate nature of the signal generator for this job. If it was the noise from the scope averaging 8000 waveforms would remove it entirely. Are we now done with the silly mock measurements or is this going to continue?

@HalFET
What is this doing under my image as some kind of "comment". I have seen your nonsense DSO-X images and they really have nothing to do with this. Here is captured signal. (Including tiny detail:  measured rough DC level using offset DAC)
If you do not really understand what all you see in image - well it is not my head ache. Btw, it looks that only what you have seen in image is this "count" and just this measurement table is perhaps most futile thing in the whole image.
So you entirely deny that it says?
Code: [Select]
Sa 1.00 GSa/s
Curr 65536 pts
delta f = 15.26 kHz
AVG = 4
I'm fairly certain that "AVG" stands for averaging... Furthermore, claiming measurement accuracy is all fun and games, but if you don't understand why averaging over 567 periods is significant, do the math please. It significantly lowers the measurement error.

And I am well aware the images I posted were ridiculous, that was very much the point of them. I essentially did the same you folks did, the only difference is that I showed what I was actually doing.

There is so many things in your post what need comment but I will pick-up some with limited time.

25 ohm. Are you really serious you wonder this?  It is extremely common way example when we test probes..  for me it is very normal. Note that Signal was connected to scope using normal probe, not 50ohm transmission line.

"AVG #4" meaning in image..
Time domain trace is NOT averaged at all. It is just normal realtime acquisition.
FFT is averaged (#4)

Why I select 3.3V offset in this previous image. Of course I select 3.3V offset because I want look (just as simulation for teaching purposes) 3.3V power rail and its details because there is "riding" something... what is not fully "unknown" anymore, least we have some trumpth (alternative truth) about it. There is around 10mV 100ns pulses.
But then this visible noise, this is really undefined and I do not know but I believe least biggest part of it is from scope front end and rest. (of course I have looked it with same settings and probe without connection to anything but just same termination without connected to signal source)

Offset can here be up to 20V. (perhaps you forget 1:10 probe when you wonder this)
For 1:1 it is in Siglent + 2V (500uV/div - 1180mV/div (most low voltage band)).

If I look  your  image with over 500MHz sinewave. Naturally this Siglent model can not do it because max 1GSa/s.






Some difference with Keysight. My eyes are very old and poor but even I can see difference...and also here Siglent time domain trace is without any averaging.../
With Rigol, do not even try.

For avoid next misunderstandings... attached again this same 3.3V DC with around 10mV 100ns pulses  as previous but now without any extra on display and what can with Keysight detect in class... "there is something" if I look your images.
 

-as previously, splitted window (time zoom) and without meas and FFT
-no average, full screen
-average #4, full screen
-average 1024# (siglent maximum) full screen

« Last Edit: March 26, 2019, 03:33:58 pm by rf-loop »
If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #123 on: March 26, 2019, 03:24:17 pm »
Because you are happy to provide a service to the community here? If the only thing it does is to put an end to these "Rigol-vs-Siglent entry level scopes" debates once and for all, it would be a big service to the community...  ;)

I've done it a dozen times. It's whack-a-mole, they just come back with something else...

...or even the same old crap. The infamous "Siglent vs. Rigol" spreadsheet is still being reposted three years after it was debunked (see page two of this thread).

It's less relevant than ever because you don't even have to hack your Rigol to get all those "optional" features that are in there, they come as standard now. Doesn't stop it being reposted though.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 12241
  • Country: 00
Re: the eternal question about rigol vs siglent
« Reply #124 on: March 26, 2019, 03:28:47 pm »
eblaster, this will never end, because they cherry pick their demonstration cases. For example, here he went for a 25 Ohm signal generator output with settings beyond what most signal generators can do. Furthermore he selected an offset voltage above 2V, because that's the limit that most manufacturers will allow as offset voltage for the mV ranges (though there are easy tricks around that one). Then he selected a 10 mV pulse, well knowing that many signal generators can't generate clean 100 ns pulses to begin with, so what you get is noisy by default. The images he showed is mostly a demonstration of how good his signal generator is, it doesn't actually show the performance of the scope.

No, no, no! These are "real life" examples!

It is typical that after technical arguments or examples in in real life...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf