Siglent SDS1104X-E vs. Rigol DS1054Z Advice?

[KungFuJosh]

Maybe it's just laziness, but I like hitting the autoset button. Yes, it's annoying waiting the 1 or 2 seconds while it's frozen, but it's less annoying than twiddling 3 knobs for 10 to 20 seconds to come to the same conclusion. Or maybe that's fun? Once it's set you can still twiddle the knobs to adjust the display or settings beyond that.

Pressing Storage then Default sets everything back to default on the Rigol if you're worried about the auto settings left over, and there's other undos also. But that wasn't an issue for my basic uses. I wasn't doing math functions or anything like that, just watching sine waves pooping around with it. The autoset seemed to work pretty well, even the triggering worked pretty well with some tricky stuff.

It couldn't freeze everything, but neither could I. Sometimes I had to hit the Stop button to print the ugliness I was seeing with the parasitic I'll hopefully eventually kill.

[BillB]

Right now, I see one of the vendors on eBay selling MicSig 100MHz 1400V differential probes has them for $120:

https://www.ebay.com/sch/i.html?_from=R40&_nkw=micsig+dp10013&_sacat=0&_sop=15

I just ordered one for $115 (best offer). Hopefully the thing works well. Thanks for the ebay idea, saved me $45 vs. the same thing I was looking at on Amazon.

Sure thing. Was that from the same vendor that had it discounted to $119 (looks like it went up a bit)? It sure is a good price. I'm tempted to get one, but I already have an HVP-70 (got it before Micsig released theirs).

$115 !?  That's a great price.  There are a few review threads of it floating around here somewhere.  It's probably more like a 60MHz probe, but for the price, it's a nice little unit.

[KungFuJosh]

$115 !?  That's a great price.  There are a few review threads of it floating around here somewhere.  It's probably more like a 60MHz probe, but for the price, it's a nice little unit.

Haha, yeah, not too bad. I started my offer at $80. For my needs, 60Mhz is plenty. I could probably leave the 20Mhz bandwidth limit on and be fine. Maybe I should? I already got the tracking number, they ship pretty quick it seems. I'm in MA, and it's coming from NJ, so it'll probably be here in a day or two...before I even have a new scope to play with.

[rf-loop]

Analog front end signal noise is also some diference between Rigol and Siglent.
These tests are comparable as much as possible.
(Note difference between Siglent Stdev and Rigol RMS (Rigol do not have Stdev so only way is use RMS and it include of course also DC offset)
Here Rigol is 100MHz BW and Siglent is 200MHz. If we reduce Siglent BW to 100MHz it reduce noise level around 3dB. As can see Siglent have lot of less noise even when it have double bandwidth.

Siglent have full resolution 500uV/div and Rigol have 5mV/div full vertical resolution.
With 10x probe it mean that highest sensitivity with full vertical resolution with Rigol is 50mV/div and with Siglent 5mV/div.

Other thing is that Rigol use highly decimated data for measurements. (also it map all samples to display, example here all 7Msample is mapped to display memory without decimation also for better intensity gradation so that it is dependent about one wfm data density on display and also sequential overlayed traces specially when use settings what give more wfm/s speed but less horizontal scrunch)

In this image Siglent can measure horizontally using full 1ns resolution.
But Rigol time resolution is perhaps roughly something like 10000 times less. (somewhere in forum is old thread where is exact data)

Of course all depends user needs what are important things but differencies between SDS1000X-E and DS1000Z is really huge if start looking all things. Is it even fair to compare so different things, even if some main parameters are same. What are same. Max 1GSa/s, 7" display, 4 channels and shared vertical controls. Siglent is more expensive, yes. And it is LOT of more if look its power as tool.
But only IF user need. Of course if example this Rigol 1kZ is more than enough for user needs then why pay more. It is still today lot of scope with its price!

But then, there is possible also things what user can not think before he find how useful some features are if he . Just example always background working waveform history buffer. Of course it is perhaps widely underestimated because most have never used this kind of feature and then no one talk about it - how talk things what do not have and/or know. Because most of scopes (specially low end scopes) do not have it at all or have some nearly useless "frame recorder" without full raw acquisition data.




I think difference is tiny bit more than negligible.

[KungFuJosh]

But then, there is possible also things what user can not think before he find how useful some features are

This is exactly right. This is a big motivator for me in purchasing the Siglent over the Rigol. A lot of people point out that the Siglent costs 40% more but I see 3 important factors:

1. Neither one of them are that expensive as oscilloscopes go. Not at all.
2. In some cases the Siglent performs 200% better than the Rigol.
3. It will cost me much less to buy the Siglent now, than it would to keep the Rigol, be disappointed by it, and then need to buy the Siglent or something else superior to the Rigol anyway. ("buy cheap, buy twice")

If it turns out I hate the Siglent, I can return it, but that's not likely.

If it turns out I never benefit from any of the better performance or features, oh well, but that's not likely either.

Now the only problem is most dealers seem to be out of stock.

[Fungus]

It's not "unknown" at all. The Rigol does all calculations, serial decodes, etc., based on the block of 1200 samples that's currently displayed on screen (squished into 600 pixels). You want a better answer? Zoom in on the area of interest (or press 'auto'  ).

PS: Any RMS calculation of non-bandwidth-limited white noise is always a bit 'suspicious' and a very suspect way to demonstrate a 'scopes supposed superiority. The VPP number is the one to focus on.

1.3mV of noise @1GHz with the probe in 1x mode? Turn on the bandwidth limiter, that's what it's for.

[rf-loop]

It's not "unknown" at all. The Rigol does all calculations, serial decodes, etc., based on the block of 1200 samples that's currently displayed on screen (squished into 600 pixels). You want a better answer? Zoom in on the area of interest (or press 'auto'  ).

1) PS: Any RMS calculation of non-bandwidth-limited white noise is always a bit 'suspicious' and a very suspect way to demonstrate a 'scopes supposed superiority. The VPP number is the one to focus on.

2)  1.3mV of noise @1GHz with the probe in 1x mode? Turn on the bandwidth limiter, that's what it's for.

1) Perhaps Keysight engineers are wrong and you are right? But here is Keysight:

Understanding Noise and How it Should Be Measured:
Random noise, sometimes referred to as white noise, is theoretically unbounded and exhibits a Gaussian distribution. Unbounded simply means that because of the random nature
of noise, the more data you collect in noise characterization measurements, the higher the peak-to-peak excursions will grow. For this reason, random phenomenon such as vertical noise and random jitter should be measured and specified as an RMS (one standard deviation) value.

2)

1GHz 

I still believe that Rigol have 100MHz "gaussian like" BW (if you understand how BW affect noise...) and here compared with 200MHz BW Siglent what also can calculate RMS and it do it right.  (here used Stdev for remove possible DC offset for get noise RMS)
Of course we can also compare with more narrow BW. But if I need 100MHz scope for measurements I need know how it works with 100MHz. If I measure unrepetitive  1Hz square wave what have 4ns risetime I can not reduce BW. I need full 100 MHz BW or better if more. So simple.
Of course when want less noise and IF can reduce BW then just reduce BW.
Also it is not so important what are these absolute values... but as you can see there is two scope face to face and using enough same setup. They are comparable. You can also see peak to peak just with eyes from screen image if you want and it is big.
Of course there is lot of differences. This is only just one.

[KungFuJosh]

Turn on the bandwidth limiter, that's what it's for.

Do you think that's good to use in general for audio stuff? Is there any reason I would need to see anything above 20Mhz probing a tube amp? Is that a dumb question?

[nctnico]

It doesn't matter whether you turn the BW limit on or off because the Rigol 1000Z is noisy like hell. The more recent oscilloscopes from GW Instek, MicSig and Siglent use a very low noise analog front-end and ADC (IIRC they all use the same so their performance when it comes to noise is comparable).

[KungFuJosh]

It doesn't matter whether you turn the BW limit on or off because the Rigol 1000Z is noisy like hell. The more recent oscilloscopes from GW Instek, MicSig and Siglent use a very low noise analog front-end and ADC (IIRC they all use the same so their performance when it comes to noise is comparable).

I'm returning the Rigol. I'll have a Siglent on Saturday. So what then?

[rf-loop]

It doesn't matter whether you turn the BW limit on or off because the Rigol 1000Z is noisy like hell. The more recent oscilloscopes from GW Instek, MicSig and Siglent use a very low noise analog front-end and ADC (IIRC they all use the same so their performance when it comes to noise is comparable).

Yes, nearly all are better than this Rigol 1kZ if think its front end signal noise. Example Owon, GoodWill (GWI), Siglent, MicSiglent, etc...

[Fungus]

Turn on the bandwidth limiter, that's what it's for.

Do you think that's good to use in general for audio stuff? Is there any reason I would need to see anything above 20Mhz probing a tube amp? Is that a dumb question?

With 10x probe it's not really needed.

[Fungus]

Siglent what also can calculate RMS and it do it right.

1) The Rigol measurement method isn't "unknown" as you claim. Just zoom in on the signal and you'll get a better (and much lower!) number.

It's not difficult to do (in fact it's quite natural to do this when you're working with a signal - you tend to zoom in to look at it).

The only thing 'suspicious' about those measurements was subtracting 3dB from the Siglent's noise level just because you felt like it. Noise is noise, the VPP of the noise is what people see on screen, not the RMS.

The "apparent" noise also depends on the 'scope settings, display brightness, display persistance, sample rate, number of channels, etc. If you watch this video at around the 7:30 mark you can see Dave change the (apparent) noise levels by changing the memory depth:

[tautech]

Different series, different processor, older model.

Good to see you watching videos about Siglents. 

[Fungus]

Different series, different processor, older model.

Still relevent to making claims about "noise" levels.

Good to see you watching videos about Siglents.

Rigol will soon have their new ASIC in a low-end model, so... meh.

[rf-loop]

The only thing 'suspicious' about those measurements was subtracting 3dB from the Siglent's noise level just because you felt like it.

Still it looks like you do not understand that there in image was two oscilloscopes with as same situation and settings as can. These can compare. No need look measurements results as all. And Siglent measurement level can see in image. There is not any 3dB subtraction.
Onlöy what I tell was that BW difference is 1:2 and you know (or then you do not know) where from this 3dB come.
Also I have knowledge and experience well enough for claim around 3dB noise level rise when turn Siglent 100MHz to 200MHz BW.  Not of course exatly due to fact that there is also some noise components what are not related to analog side BW.  3dB tip was only from common 10 Log(200 /100)

If we compare Rigol and Siglent then take 100MHz Rigol and 100MHz Siglent and then Siglent show even less  noise.

If you do not believe it is your own problem.

Here compared same Siglent scope with 100MHz BW and 200MHz


100MHz out from box

This new test have same settings so they are quite highly comparable and also same individual scope.

Same individual scope mod to 200MHz
As can see theory and practice is as same as it can be. (because there is also other things related to noise)

If we look how much it rise and take randomly example CH4. Level (Vrms ¹)  rise 3.14dB
And if we look how our BW shape and BW width change,  it can say that it is just as expected.
1. V Stdev = V rms - possible DC offset

Different scopes are difficult to compare if look noise p-p because then we need collect same amount of data for measurement and sorry but it looks like with Rigol effective amount of measured data points are really low compared to Siglent due to slower wfm/s in faster times and with slow time bases highly decimated data points. With random gaussian distribute noise this is very important. Probability for catch highest p-p values with Rigol is really low if compare Siglent. If we compare random noise p-p we need collect and analyze same amount of measured data. Example here previously as you told Rigol use 1200 points for measurements. And in this image prevciously where was Rigol and Siglent this Siglent use 7M points. If with this setting turn p-p measure on ... well Rigol tell p-p from random noise using 5800 times less data. Of course propbability to find rare highes peaks rise lot of.
This is why we need compare apples to apples and not apple to shoe.

[Fungus]

Onlöy what I tell was that BW difference is 1:2 and you know (or then you do not know) where from this 3dB come.

I know where it came from, I'm just saying that it shouldn't be "compensated" or reduced in any way because of $THEORY$.

The noise you see on screen is going to be there every time you look at a signal, even if a different brand of oscilloscope has less bandwidth than yours(!)

[Fungus]

FWIW the displayed RMS value will halve (approximately) if you zoom in on the Rigol. This halved value is a more correct representation of the noise.

I have two problems with the comparison above, they are:
a) Not knowing (or wilfully misrepresenting) how Rigols work, and
b) "Compensating" the Siglent down by 3dB (because, hey, why not?)

Don't get me wrong: I believe the Siglent probably has a lower noise than the Rigol (but then it ought to, it costs more).

I'm just questioning the data being presented here and trying to point out that trying to measure it is almost impossible. If you watch Dave's video I posted above you can clearly see he's not willing to make a definitive statement because it depends on a lot of variables.

PS: The Rigol makes no claims to having a 500uV range. A more realistic comparison would be the visible noise with them both on their lowest range, ie. Siglent on 500uV and the Rigol on 1mV. That's the noise that people interested in those ranges are actually going to see (in real life!) whenever they look at a signal.

Define your requirements: If you need 500uV then buy a Siglent (or better... eg. a 10 bit R&S).

I know we're all volt-heads around here but for most people/jobs, 1mV of noise (eg. Rigol) isn't a problem.

[KungFuJosh]

Rigol will soon have their new ASIC in a low-end model, so... meh.

Do you have any specifics about this? Hopefully they have a new design for the front panel too.

I think both companies are in a funny position.

Only Rigol's highest-end scopes are relatively new (from what I could see), and they still don't have a good front panel design. I hate the arbitrary, asymmetrical goofy outlines. Looks better than Instek, but that's a low bar.

It also seems like only the SDS1004X-E series is relatively new from Siglent, and everything higher-end is trailing behind. I like the look, larger screens, and individual vertical controls of some of the higher-end Siglent models. But why would I spend 4 times as much money for less scope ability than 1104X-E? Yes, I know there's higher bandwidth and sampling available, but the rest of the scope needs to catch up.

[Fungus]

Rigol will soon have their new ASIC in a low-end model, so... meh.

Do you have any specifics about this?

Obviously not. No company leaks information like that.

(not until a few weeks before launch).

Hopefully they have a new design for the front panel too. I hate the arbitrary, asymmetrical goofy outlines. Looks better than Instek, but that's a low bar.

Meh. You probably wouldn't notice it after a day of use.

It also seems like only the SDS1004X-E series is relatively new from Siglent, and everything higher-end is trailing behind. I like the look, larger screens, and individual vertical controls of some of the higher-end Siglent models. But why would I spend 4 times as much money for less scope ability than 1104X-E? Yes, I know there's higher bandwidth and sampling available, but the rest of the scope needs to catch up.

Siglent has been investing in the low-range (to compete with the DS1054Z, which has made Rigol a ton of money) while Rigol has been investing in the the mid-range.

The situation should flip around now the respective new models have been launched. Let's see if Rigol startes dropping some hints a few weeks from now.

(to make anybody who's thinking of going from DS1000Z to SDS1000X-E wait a while before deciding )

[KungFuJosh]

Obviously not. No company leaks information like that.

So you're speculating?

Hopefully they have a new design for the front panel too. I hate the arbitrary, asymmetrical goofy outlines. Looks better than Instek, but that's a low bar.

Meh. You probably wouldn't notice it after a day of use.

I had the DS1054Z for 3 weeks. I noticed it every single time I looked at it. I hated the design. It annoys me. My primary job is web/graphic design; I can't help but notice that stuff. It drives me batty when designs are that bad. The Instek also seems like a better choice than the 1054Z, but I simply won't put it on my bench because I can't look at it.

The situation should flip around now the respective new models have been launched. Let's see if Rigol startes dropping some hints a few weeks from now.

(to make anybody who's thinking of going from DS1000Z to SDS1000X-E wait a while before deciding )

Given the business model (from most of these companies) of releasing the scopes and letting the customers do the debugging, I think I'll pass. It'll be a minimum of a year before any new scopes are reliable at all. 

[Fungus]

I had the DS1054Z for 3 weeks. I noticed it every single time I looked at it. I hated the design. It annoys me. My primary job is web/graphic design; I can't help but notice that stuff. It drives me batty when designs are that bad. The Instek also seems like a better choice than the 1054Z, but I simply won't put it on my bench because I can't look at it.

Is the Siglent really that much better? 

Maybe you should get a Micsig tBook or something.


I assume you won't want the optional fugly rubber frame but you can even hide the thing away and use it via your phone.


Prices for 4-channel, 100MHz are comparable to the Siglent.

[Mr. Scram]

It's funny, because I actually quite like the design of the DS1054Z and loathe that of the Siglent. The oval channel selection buttons drive me nuts. It's completely out of tune with the rest of the design.

Admittedly, Rigol jumped the shark on some of the newer designs that followed the same basic idea of the DS1054Z. Regardless, there's no accounting for taste. There's plenty of professionally designed junk around to prove that once and for all.

[KungFuJosh]

I had the DS1054Z for 3 weeks. I noticed it every single time I looked at it. I hated the design. It annoys me. My primary job is web/graphic design; I can't help but notice that stuff. It drives me batty when designs are that bad. The Instek also seems like a better choice than the 1054Z, but I simply won't put it on my bench because I can't look at it.

Is the Siglent really that much better? 

Maybe you should get a Micsig tBook or something.

I assume you won't want the rubber bumper frame but you can also use it via your phone.

The Siglent is that much better. It looks boring, but it isn't hideous. It also looks like the front panel graphic overlay is detachable, so I'm going to try popping it out, scanning it, and designing my own overlay if I decide I like the scope.

I considered the MicSig, but the reviews for the model in the same price range said the MicSig was slow, and the screen was laggy. So, no thanks on that one. I love the idea of the touchscreen, but the form factor isn't more convenient for me than the bench scopes. It would actually be a hindrance.

[nctnico]

I don't get why design is so important. As long as the user interface is not cluttered and easy to use all is well. IMHO it is way more important a tool does a good job.