Rigol 5074 vs Siglent SDS2104X Plus vs Keysight DSOX1204G

[Electro Fan]

2N3055 & nctnico,

>pause

If it makes you guys each and both feel any better, your debates are fairly entertaining and definitely educational.  Many of us have watched lots of rounds in this mostly good natured TE boxing match, and each time you guys get into it a few more insights about TE, electricity, analog/digital, and workflow becomes attainable - at least in my case.  So, just wanted to say thanks for sharing your knowledge and passion.  You guys both help make this a very educational place.   

>end pause

[Bassman59]

Zooming out is much more useful feature. But the Siglent can't do that very basic operation.

Yep. If I count the number of Bode plots I've done in my life vs. the number of times I've zoomed out on captured data, the zoom wins.

Siglents don't do that most basic thing.

The Tek DPO3054 to my right on my bench has a zoom feature that I use literally all the time. It's a vital feature of an oscilloscope and I'm honestly surprised that the Siglent doesn't have zoom.

Given that 99% of my work is digital design, I don't need a Bode plot feature.

Everyone's uses are different, of course.

[2N3055]

Zooming out is much more useful feature. But the Siglent can't do that very basic operation.

Yep. If I count the number of Bode plots I've done in my life vs. the number of times I've zoomed out on captured data, the zoom wins.

Siglents don't do that most basic thing.

The Tek DPO3054 to my right on my bench has a zoom feature that I use literally all the time. It's a vital feature of an oscilloscope and I'm honestly surprised that the Siglent doesn't have zoom.

Given that 99% of my work is digital design, I don't need a Bode plot feature.

Everyone's uses are different, of course.

I'm sure we have communication problem here. Siglent SDS2000X+ has zoom. It is implemented very similar to the Tek DPO3054, with additional benefit of touch screen with gestures to move viewport around. If I remember correctly, you can also zoom vertically.

Here we are ad nauseam discussing something that unfortunately got called "zoom out" but in fact is a practice to use manually set long buffer to force scope to capture long buffer (say 100 ms),  while you're looking at signal with short timebase (say 500 ns/div) to deliberately avoid use of zoom. And if you stop capture, since you did, in fact, capture very long capture, you can (by changing timebase button) "zoom out" and move around with horizontal position button.

Nico swears by it and claims it is faster and easier to use than the other way around, by using zoom mode and have dual time base at the same time on screen and navigate that way. I disagree with him with  my whole heart (as you can see )

Hope that explains it..
Regards,

[2N3055]

2N3055 & nctnico,

>pause

If it makes you guys each and both feel any better, your debates are fairly entertaining and definitely educational.  Many of us have watched lots of rounds in this mostly good natured TE boxing match, and each time you guys get into it a few more insights about TE, electricity, analog/digital, and workflow becomes attainable - at least in my case.  So, just wanted to say thanks for sharing your knowledge and passion.  You guys both help make this a very educational place.   

>end pause

Yep, at times veritable Monty Python.. Glad you like it... 

[Fungus]

Here we are ad nauseam discussing something that unfortunately got called "zoom out"

Doesn't matter what you call it, one name is as good as another. Fact it: Siglents don't zoom out.

Other 'scopes do zoom out and it's a useful, logical, natural way to work.

Siglents have no plausible reason for not doing it. It's a design choice. A bad design choice.

in fact is a practice to use manually set long buffer to force scope to capture long buffer (say 100 ms),

a) The buffer is often bigger then the screen in AUTO mode but Siglents don't even fill it when it is. They only show what was on screen.
b) Siglents don't obey you even when you manually set the buffer size. See the image below, 200M points configured, 10.0 (ten point zero!) points captured.

The vertical grid is evenly spaced, isn't it? How comes that we get 5 dB/div in the bottom row, then 4 dB /div above

If I had to guess I'd say the numbers have fractions and they're using a round() function to display them.

And ... is it any sillier than having 10.0 points of memory? eg. Can Siglents capture 10.2 points of memory?   

[2N3055]

Here we are ad nauseam discussing something that unfortunately got called "zoom out"

Doesn't matter what you call it, one name is as good as another. Fact it: Siglents don't zoom out.

Other 'scopes do zoom out and it's a useful, logical, natural way to work.

Siglents have no plausible reason for not doing it. It's a design choice. A bad design choice.

in fact is a practice to use manually set long buffer to force scope to capture long buffer (say 100 ms),

a) The buffer is often bigger then the screen but Siglents don't even fill it when it is. They only show what was on screen.
b) Siglents don't obey you even when you manually set the buffer size. See the image below, 200M points configured, 10 (ten!) points captured.

The vertical grid is evenly spaced, isn't it? How comes that we get 5 dB/div in the bottom row, then 4 dB /div above

If I had to guess I'd say the numbers have fractions and they're using a round() function to display them.

And is it any sillier than capturing 10.0 points of memory? Can Siglents capture 10.2 points of memory...?   

Like I said,  Monty Python. 

[Fungus]

Like I said,  Monty Python. 

I'll be here all week.

[2N3055]

Like I said,  Monty Python. 

I'll be here all week.

Of course you will, like a proper circus should!! 

[Electro Fan]

Presumably it would cost oscilloscope manufacturers too much to offer a price competitive oscilloscope that enables users to have both zoom in and zoom out?  And good search/navigate?

[nctnico]

Presumably it would cost oscilloscope manufacturers too much to offer a price competitive oscilloscope that enables users to have both zoom in and zoom out?  And good search/navigate?

Zoom in/out is basically the same function. So from a functional point of view it only depends on the UI. Search OTOH does take some effort because this involves analysing the acquired data either through software or using part of the trigger system. Search is not a high-end scope feature BTW. The sub $500 GW Instek GDS1054B has search for example.

[TK]

Presumably it would cost oscilloscope manufacturers too much to offer a price competitive oscilloscope that enables users to have both zoom in and zoom out?  And good search/navigate?

Zoom in/out is basically the same function. So from a functional point of view it only depends on the UI. Search OTOH does take some effort because this involves analysing the acquired data either through software or using part of the trigger system. Search is not a high-end scope feature BTW. The sub $500 GW Instek GDS1054B has search for example.

Sub $300 in the US with EEVBLOG discount at tequipment

[Fungus]

Presumably it would cost oscilloscope manufacturers too much to offer a price competitive oscilloscope that enables users to have both zoom in and zoom out?

The Rigol DS1054Z had zoom in and out and that's a cheap 'scope. $300 Insteks can do it as well. Dave's video shows a long list of scopes that do it, even an Owon and a Uni-T.

So, no, price has nothing to do with it.

[tautech]

Some very simple examples of zooming out in a capture:

Manually initiated capture at full mem depth SDS5000X:


Engage Zoom mode with a single press of Zoom or timebase:


Zoom out using timebase:


And further:


Siglent's can't zoom out......yeah right !   
Same principles work for all their X/X Plus and X-E/U model DSO's.

[TK]

@tautech: you are not zooming out from the original capture, you are capturing, zooming in then out

[tautech]

@tautech: you are not zooming out from the original capture, you are capturing, zooming in then out

Serves exactly the same purpose......it's still a capture.
Like I said it's a very simple example and manually instigated whereas if you were serious about waveform analysis you might single shot to some particular trigger condition in a slow timebase and then use the real power of a DSO...zooming in !

[Electro Fan]

Presumably it would cost oscilloscope manufacturers too much to offer a price competitive oscilloscope that enables users to have both zoom in and zoom out?

The Rigol DS1054Z had zoom in and out and that's a cheap 'scope. $300 Insteks can do it as well. Dave's video shows a long list of scopes that do it, even an Owon and a Uni-T.

So, no, price has nothing to do with it.

ok, over simplified on the terminology

Capture just what's on the screen and adjust the time base to show more detail vs...... ?

Fungus, since you are a good distiller of info (and have a good sense of humor), maybe you can propose names for the two alternatives that have been debated.

[Electro Fan]

@tautech: you are not zooming out from the original capture, you are capturing, zooming in then out

An example of how we use similar words to discuss the same stuff over and over but the terminology of the sender gets confused by the receiver and then we have nearly half the debate over misunderstandings vs. intended content.

[Fungus]

Capture just what's on the screen and adjust the time base to show more detail vs...... ?

It's true that it's not just zooming. On other 'scopes you can move the display left and right and more signal will scroll into view as you do it. On a Siglent you just get blackness.

Any name needs to take that into account, too. 

[Electro Fan]

Presumably it would cost oscilloscope manufacturers too much to offer a price competitive oscilloscope that enables users to have both zoom in and zoom out?

The Rigol DS1054Z had zoom in and out and that's a cheap 'scope. $300 Insteks can do it as well. Dave's video shows a long list of scopes that do it, even an Owon and a Uni-T.

So, no, price has nothing to do with it.

Yes, purely at face value zoom (in or out) in the narrowest sense is not inhibited by price.  However, when we zoom out (metaphorically haha) the issue under discussion might well be somewhat cost (of NRE and manufacturing) driven as we see that one of the brands held up as an example of how to approach the workflow is Agilent/Keysight - and it’s generally believed that they invested early on in the development of ASICs that enabled some extra spiffy performance. 

So, once we come up with some naming conventions that represent the alternatives being discussed we might see that one of the alternatives does in fact cost more (or did at the time).  Or, once we run this to ground (50/50 on that since this debate has been going in various forms for awhile) we might find that with current technology cost (and therefore price) is not a driver.  Although even if hw NRE has come down, sw NRE probably hasn’t.  And if it turns out economics is not a driver, why wouldn’t scope manufacturers provide functionality that supports both approaches (whatever we call them)? 

Net, net:  it might be too early to say price has nothing to do with the discussion on the table.

[2N3055]

Capture just what's on the screen and adjust the time base to show more detail vs...... ?

It's true that it's not just zooming. On other 'scopes you can move the display left and right and more signal will scroll into view as you do it. On a Siglent you just get blackness.

Any name needs to take that into account, too.

No that is not true. On all scopes you get same 100ms of capture, and can change what you see on the screen with timebase and horizontal position.
Difference is that for initial capture you need to use different procedure.
Which have pros and cons. 

Those that capture fixed buffer size will have very slow update rate. Even if you look at nanoseconds on the screen, trigger rate will be slow and be dependent on buffer size and current sample rate. That might be whatever it is based on memory size, timebase, number of channels used etc. and it will change all the time.  If you don't need all that data all the time, you will need to keep managing memory size manually all the time. Or set it to auto, which will make it work like Siglent does.
Those things are reasons why some hate some Tektronix scopes so much. With scopes that do auto memory management you just use scope, no need to think about these things too.
OTOH it will give you more data than you saw on screen originally, which might be useful or not, depending how you work with scope (how deliberate and planned your work is, or you just twiddle with knobs until you manage to see something of interest. It will also depend on whether your work is routine or you are measuring something first time and have no clue what to expect).

Those that acquire only amount of data that fits to time span on the screen (which is what you have with CRT analog scopes), will never have extraneous data to process (acquire, move, ,measure or do math with it) and that will enable scope to be fast and responsive. Also there are no hidden rules. WYSIWYG, and if you need longer capture, just set it for that. It is that simple, really. But there are moments where you need/want to look at one shorter period of whole waveform but still capture the lot. Like, for instance, some startup transient inside startup sequence, that once in a while will have a glitch on one particular time. With this scope, to do this you simply use zoom function to achieve dual timebase, in one (main) window you still have whole long capture, and in zoom window you look at "magnified portion" at fast timebase. That is for both stopped and running scope. Some people say "that is wasting my screen...thats baaad..", some people say "that is awesome, I can see what I'm doing, that is so good .."

But once you got your 100 ms worth of data in the scope, with both you can "zoom in and out" with timebase and move left and right with horizontal position knob.

And if you set your "manual memory scope" to anything less that 100 Mpts, and have a high sampling rate scope  you will also get "just get blackness" outside what you captured..
In fact, that would make scope very unresponsive for general work, so people will keep it somewhere shorter. So scope will start dropping sample rate very quickly and alias and loos information.
Auto mode was invented to solve this problem.

Seriously, I'm contemplating making a cartoon about this....

[tatel]

I'm just a hobbyist. I'm just a newbie here. I even don't have any scope yet.

But, while looking at different threads to get information about what could I buy, I have found most have one thing in common: Siglent-liking people bullying other people that points weak features in Siglent devices.

Guys, you don't need to do this and it's even counter productive. Siglent products look nice by themselves and is up to us to decide if we want to own one or not.

I don't even know how this zoom out thing has to be called, but after watching Dave's video (12:23) I call it Siglent-scopes-laugh-at-your-face-when-you-set-memory-depth-manually-while-other-brands-don't

Please note I'm not even considering if this is technically useful/right or not. It doesn't matter right now. What matters is other brands do it. Cheaper scopes do it. Again it's up to us prospective buyers to decide what we are going to purchase after knowing it.

Would that stop me considering Siglent products? No, if I think they are still worth the money. But it's good to know as much as possible before buying. It's specially good to know about the weak points. The strong points mostly are already touted by the brands themselves.

So Siglent decided to put a manual memory depth setting but anyway the scope sets it automatically no matter what you choose? Good to know, that's information that can be, more or less, useful, and nctnico and fungus are helpful letting know about it. Siglent people could do the same by just pointing that increasing memory depth can have a big impact in wfms which is a bad thing too. Then us poor ignorant bastards would have being taught something so we could make our own decision.

Instead we get thread after thread hijacked by arguments spiced with laughs, ad hominem attacks, incivic behaviour, etc, which is neither nice nor informative.

nctnico and fungus (and any other that could express the same opinion) have perfect right to do so. Should Siglent be worried about that, they could make some changes. They decided to give "low priority" to it? Be it, but then they must take any criticism.  We all know what "low priority" means and we all know all brands do the same sometimes. Still we have to choose among what's available.

When I registered in this forum, I was thinking it was as grassroots as a forum can be. Now it does look more like a hijacked forum becoming a Siglent astroturfing effort. At the very least, all this bullying looks very much the same that an apple fanboy flame war. I myself was an apple fanboy in the 1990s so I can identify the trend just by the BS smell.

Siglent not taking some user feedback into account could or could not make me stop considering their products, but there's no amount of bold text nor laughing smiles nor bullying nor ad hominem attacks that could made me change what I think after seeing somebody being assured that connecting SLA1016 to his SDS1104X-E wouldn't have  any performance penalty at all, then getting this report from a user:

https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg3528422/#msg3528422

Now, in this same thread, even not having any scope yet, I was quite suspicious about the screenshots posted, then TK confirmed what I was thinking.

Again: guys, this is counter-productive.

I don't think this argument is being civil so I'm getting out right now.

Have a nice day

[Fungus]

I'm just a hobbyist. I'm just a newbie here. I even don't have any scope yet.

But, while looking at different threads to get information about what could I buy, I have found most have one thing in common: Siglent-liking people bullying other people that points weak features in Siglent devices.

Guys, you don't need to do this and it's even counter productive. Siglent products look nice by themselves and is up to us to decide if we want to own one or not.

You'd be even madder at them if you'd been here when they were constantly bashing the Rigol DS1054Z in every single thread where it was ever mentioned.

[Electro Fan]

I'm just a hobbyist. I'm just a newbie here. I even don't have any scope yet.

But, while looking at different threads to get information about what could I buy, I have found most have one thing in common: Siglent-liking people bullying other people that points weak features in Siglent devices.

Guys, you don't need to do this and it's even counter productive. Siglent products look nice by themselves and is up to us to decide if we want to own one or not.

I don't even know how this zoom out thing has to be called, but after watching Dave's video (12:23) I call it Siglent-scopes-laugh-at-your-face-when-you-set-memory-depth-manually-while-other-brands-don't

Please note I'm not even considering if this is technically useful/right or not. It doesn't matter right now. What matters is other brands do it. Cheaper scopes do it. Again it's up to us prospective buyers to decide what we are going to purchase after knowing it.

Would that stop me considering Siglent products? No, if I think they are still worth the money. But it's good to know as much as possible before buying. It's specially good to know about the weak points. The strong points mostly are already touted by the brands themselves.

So Siglent decided to put a manual memory depth setting but anyway the scope sets it automatically no matter what you choose? Good to know, that's information that can be, more or less, useful, and nctnico and fungus are helpful letting know about it. Siglent people could do the same by just pointing that increasing memory depth can have a big impact in wfms which is a bad thing too. Then us poor ignorant bastards would have being taught something so we could make our own decision.

Instead we get thread after thread hijacked by arguments spiced with laughs, ad hominem attacks, incivic behaviour, etc, which is neither nice nor informative.

nctnico and fungus (and any other that could express the same opinion) have perfect right to do so. Should Siglent be worried about that, they could make some changes. They decided to give "low priority" to it? Be it, but then they must take any criticism.  We all know what "low priority" means and we all know all brands do the same sometimes. Still we have to choose among what's available.

When I registered in this forum, I was thinking it was as grassroots as a forum can be. Now it does look more like a hijacked forum becoming a Siglent astroturfing effort. At the very least, all this bullying looks very much the same that an apple fanboy flame war. I myself was an apple fanboy in the 1990s so I can identify the trend just by the BS smell.

Siglent not taking some user feedback into account could or could not make me stop considering their products, but there's no amount of bold text nor laughing smiles nor bullying nor ad hominem attacks that could made me change what I think after seeing somebody being assured that connecting SLA1016 to his SDS1104X-E wouldn't have  any performance penalty at all, then getting this report from a user:

https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg3528422/#msg3528422

Now, in this same thread, even not having any scope yet, I was quite suspicious about the screenshots posted, then TK confirmed what I was thinking.

Again: guys, this is counter-productive.

I don't think this argument is being civil so I'm getting out right now.

Have a nice day

Hi tatel,

Sorry for the less than great first month but if you give this place a little more time you will see that your initial sample and interpretation might not have been truly representative.  The Test Equipment forum is an excellent forum and EEVblog is a great site.

Since you don't seem to mind a long post (me too) here is a quick story to set the stage.

In one of the Star Trek adventures the crew made a landing on Earth maybe in the 1980s or so while their ship was having technical difficulties and their systems were mostly down.  So at one point (I think it was) Scotty and the captain are reduced to using a regular (to them ancient) Mac.  At first they try to issue some voice commands but at the time the computer has no way to accept voice input; one of the Trek guys grabs a wired mouse thinking it might be a microphone and speaks into it.  The person who owned the computer looks at Scotty and the captain like "Who are these guys?  They don't know bupkis about technology."  Pretty quickly though the Mac owner realizes he's dealing with some pretty good engineering talent.  At another point in the movie realizing how they appear out of sorts one of the Trek guys says to another incredulous 1980s Earthling, "Well, you're not really catching us at our best."  Or something like that.  I'm not a big Star Trek fan but those scenes and that line always stuck with me.

Turns out you didn't catch this forum at it's best.  We've all seen other web sites where personalities beat on each other; it's not the case here; people here do have personalities but to a very high extent it's a super respectful and friendly place.  Dave set a culture of friendly with an occasional touch of humor and it's the norm, so if a particular thread seems off the rails or is not your cup of tea, just find another thread on the same topic and it will probably be more to your liking.  There is a lot of depth and breadth here and it's generally easy to find; the search feature is pretty good.  And there is SO MUCH good content here you can often find it fastest by adding your search terms plus EEV and Google will help you find it.

Overall, I don't think you will find any web site much less a test equipment forum anywhere that has more knowledgeable and helpful and friendly people - but if you do, people here will be happy to hear about it because hardly anyone here has just one axe to grind whether it's a particular product or brand or technology or any one web site.  Many of the contributors here have excellent web sites and/or make great youtube videos, etc.  What people have around here is a wide range of experience (from relatively little to a ton) across a lot of technologies and use cases, and what you are seeing is a passion for "figuring out" (learning and teaching) and sharing in general.

What you happened to come across was more like some commercials in between good TV shows.  There are some topics that come up from time to time, and how to manage data capture with logic analyzer functionality built into oscilloscopes happens to bring out one of our oldest running commercials.  I've been here almost 8 years and it's a commercial that has evolved and keeps running but it's just a sliver of what's on TV here.  In the process it might look like people selling their brand but it's really not.  Ok, well maybe one or two but even the sales people here are knowledgeable, helpful, and always friendly.  Mostly, people here ebb and flow in generally focused but sometimes wide ranging discussions in search of ever better tools, and an increased understanding of electricity including everything analog or digital, or mixed-use.  How to capture, measure, visualize, and manage signals that are comprised of amplitude, frequency, phase and other attributes is not easy to describe in the few words of a text post and it can get more challenging when you throw in how to do it for ever lower prices. 

Currently Siglent gets held up as a reference point for good and/or bad because over the last couple years or so they have come from being in the running to maybe being in the lead on price/performance for various product categories of interest.  A few years ago Rigol was consistently the reference point, and they are still in the mix, of course - along with all the others whether you are talking about price or performance or functionality, or whatever the criteria.  I think relatively few people here will buy or recommend a product only because of the logo.     

In any event you won't find more people willing to discuss more aspects of oscilloscopes (and other test equipment) across all brands and models anywhere on the planet, I think.  And what that gives users here is the opportunity to frame any question or any thought and get a reply from interested, passionate, and often highly experienced people.  And sometimes you get responses from not so experienced people, or people who are just not fully informed.  But even then someone usually jumps in to try to add helpfully clarifying info to put things back on track.  And sometimes the replies are not right or wrong, they are just opinions.  But thankfully, there really is very little bullying going on here.  You will get some strong opinions but bullying is pretty rare, very rare here.  What might appear as bullying is people trying hard with text to represent complex technology with subtle UI issues, and sometimes (often?) people are writing/reading in English which is not their primary language.  So, sometimes it's subtle stuff that gets people tangled up.

In any event, hope you stick around and feel free to talk about your questions, answers, observations, interests, or whatever you think is relevant.

[bestel]

This thread is really interesting, because I think this price range concern a lot of people.

So maybe we can try to sum up the good/bad for all scopes?

Rigol 5074 :
+ 1x8GSa/s, 2x4GSa/s, 4x2GSa/s
+ Lot of memory (4x50Mpts)
+ Advanced math
+ AWG 2x25 MHz in option (50/Hi-Z = 2.5Vpp/5Vpp)
+ Advanced decoding options
+ 4 controls knobs
- Bad input noise (compared to the others)
- Only real 4mV/div sensitivity
- Low res bode plot

Keysight DSOX1204G
+ Hardware decoding (fast update rate)
+ AWG 20 MHz (50ohm/Hi-Z = 9Vpp/12Vpp)
+ High res bode plot ? (after firmware update)
+ Responsive & easy to use (UI, UX, ...)
- Price range
- Lack of memory (4x1Mpts)
- 64k FFT

Siglent SDS2104X Plus
+ 10" touch screen
+ Lot of memory (4x100Mpts)
+ AWG 1x50 MHz in option (50/Hi-Z = 3Vpp/6Vpp)
+ High res bode plot
+ real 500µV/div
+ 2Mpts FFT
- No "zoom out" hahaha (we should call it "masked" or "forced" acquisition)
- No internal multi-meter and frequency counter? (did not see in datasheet)


Tell me if you think I loosing a point somewhere.

It seems that the Siglent is the best option for "analog works", and rigol best option for "digital works".
Keysight lacks in deep memory, but provide a good integrated AWG voltage output swing, and a solid Megazoom IV experience.

In Europe :
- Rigol 5074 4x70Mhz + Bundle MSO5000-BND = +/- 1100€ (limited time offer)
- Keysight DSOX1204G 4x70Mhz = +/- 1500€ (normal price)
- Siglent SDS2104X Plus 4x100Mhz + AWG = +/- 1700€ (normal price)

[tautech]

SDS2000X Plus HW frequency counter is always ON at top right of display plus several other frequency related measurements can be selected from the Measurements menu.

Up until March 31 most options including AWG were free in a sales promotion that has now expired however if you shop around there will still be units available with the free options package.