Author Topic: Best 350MHz scope in a hackable world (Siglent SDS2104X Plus or Rigol MSO5072)  (Read 36603 times)

0 Members and 1 Guest are viewing this topic.

Offline pascal_swedenTopic starter

  • Super Contributor
  • ***
  • Posts: 1539
  • Country: no
Bandwidth savings in a hackable world:
We will always choose the lowest bandwidth model, as we know that we can upgrade it to 350MHz anyhow.
Siglent starts with 100MHz. There is no 70MHz model. So we can not go any lower to save costs.
Rigol starts with 70MHz. This saves us some extra money compared to the 100MHz model.

Number of channels and possible savings in a hackable world:
We need to make sure that the physical connectors are already there, as this can not be fixed with a software hack.
Siglent SDS2102X Plus does not come with 4 physical connectors, so we really have to go for Siglent SDS2104X Plus.
Rigol MSO5072 has only 2 channels enabled by default but comes with 4 physical connectors, so we can choose Rigol MSO5072 instead of Rigol MSO5074.

Conclusion: If we want a 350MHz scope with 4 channels in a hackable world, we can choose between Siglent SDS2104X Plus or Rigol MSO5072:

Siglent SDS2014X Plus can be upgraded to Siglent SDS2354X Plus.
Rigol MSO5072 can be upgraded to Rigol MSO5354.

But what is the best option of these two oscilloscopes when comparing them side by side?

Sample rate: Siglent only supports up to 2GSa/s sample rate, while Rigol supports up to 8GSa/s sample rate.
In practice we can consider Rigol has only double sample rate of Siglent if one uses at least 2 channels.
Do we really need 4Gsa/s for 350MHz? Or would 2Gsa/s be enough?

Waveform capture rate: Almost the same for both units. Siglent: 480.000 Wfms/s, Rigol: 500.000 Wfms/s.
But do we know if this value can be achieved for real or not? What are the limitations on the time base settings?
Maybe Rigol or Siglent perform very different on different time base settings. Are there any practical tests available about this?

Sample memory: Siglent supports 400MPts in total, while Rigol supports 200MPts in total.
There is no detail if the Siglent can achieve 400MPts when using only 1 channel in total.

Logic analyzer: The 16 channel Logic Analyzer costs about the same for both units.
However the Siglent Logic Analyzer looks of higher quality.
Do they support same voltage levels? What about the various trigger options?

Display output: The Rigol has an HDMI port while the Siglent has a VGA port.
Obviously HDMI provides better picture quality than VGA. But is the HDMI output 720P or 1080P?

Overall winner of both oscilloscopes based on the above comparisons:

Would that mean that Rigol MSO5072 (hackable to MSO5354) is the big winner?
Or what features (besides memory depth) would favor the Siglent SDS2014X Plus (hackable to SDS2354X Plus) instead of the Rigol?

The Siglent can be further upgraded to 500MHz, but wouldn't that reach the limits given that sample rate is max. 2GSa/s?
« Last Edit: June 11, 2020, 10:31:56 am by pascal_sweden »
 
The following users thanked this post: decybelq

Offline Wade2019

  • Contributor
  • Posts: 18
  • Country: cn
The best performing oscilloscope must be MSO5072, but MSO5072 is terrible in terms of user experience
 
The following users thanked this post: NCG

Offline KeBeNe

  • Regular Contributor
  • *
  • Posts: 73
not only look at the technical data, the feel is also an important point.
I had the Rigol 5k and replaced it with the Siglent 2104X+ without regrets
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
 

Offline pascal_swedenTopic starter

  • Super Contributor
  • ***
  • Posts: 1539
  • Country: no
About the described limitation in the Siglent SDS2000X Plus with regards to the 200MPts memory.

Can someone test if the Rigol MSO5000 can capture the full 200MPts memory at all time bases?
 

Offline tv84

  • Super Contributor
  • ***
  • Posts: 3212
  • Country: pt
About the described limitation in the Siglent SDS2000X Plus with regards to the 200MPts memory.

Can someone test if the Rigol MSO5000 can capture the full 200MPts memory at all time bases?

Siglent captures the 200Mpts at all time bases. Just not the way you think they are being captured...  :palm:

As such, claiming 200 Mpts (WITHOUT a BIG disclaimer) seems deceitful.

(Nonetheless I like the scope.)
 
The following users thanked this post: Simon_RL

Offline pascal_swedenTopic starter

  • Super Contributor
  • ***
  • Posts: 1539
  • Country: no
Obviously you did not look at the video!
Check from 00:35 and beyond!

It is perfectly possible to capture 200Mpts at ALL time base settings.
The scope just has to sample all the time, and the moment it triggers it will display that part on the screen, but it means that it has captured data before and after.

You can compare it with the time shifting buffer in a Set-Top Box for Digital Television.
If you are watching on a TV channel you can use time shifting to go back and forth.
« Last Edit: June 11, 2020, 11:06:51 am by pascal_sweden »
 

Offline Gandalf_Sr

  • Super Contributor
  • ***
  • Posts: 1729
  • Country: us
The best performing oscilloscope must be MSO5072, but MSO5072 is terrible in terms of user experience
Emphasis added is mine.

How, exactly, is MSO5072 terrible in terms of user experience? 

I own an MSO5074 and it does everything I need, the screen is sharp and clear, the touch screen is useful, the mouse works great, I can zoom out, it has bode plot, up to 8 Gsps.

I'd give mine 9/10 for user experience and 10/10 for value.
If at first you don't succeed, get a bigger hammer
 
The following users thanked this post: egonotto, Fungus, luma

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Obviously you did not look at the video!
Check from 00:35 and beyond!

It is perfectly possible to capture 200Mpts at ALL time base settings.
The scope just has to sample all the time, and the moment it triggers it will display that part on the screen, but it means that it has captured data before and after.

Why doesn't it show that data when Dave zooms out? Even my Rigol DS1054Z can do that.


 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
The best performing oscilloscope must be MSO5072, but MSO5072 is terrible in terms of user experience
Emphasis added is mine.

How, exactly, is MSO5072 terrible in terms of user experience? 

I'd give mine 9/10 for user experience and 10/10 for value.

We live in a world where user interface "experts" are busy making grey on grey web sites, "dark" themes for our computers, etc. I really don't see it myself but maybe the Siglent owners have a special sensitivity to something in the interface that's completely invisible to you and me.

The only advice I can give is to ignore all posts that claim a user interface is good/bad but without giving any evidence for that claim. It could be based on something esoteric idea that makes no difference in real use
 
The following users thanked this post: Gandalf_Sr

Offline luma

  • Regular Contributor
  • *
  • Posts: 130
  • Country: us
The best performing oscilloscope must be MSO5072, but MSO5072 is terrible in terms of user experience

Can you explain a bit about your experience with the scope and what exactly is terrible about it?
 
The following users thanked this post: Gandalf_Sr, Fungus

Offline Gandalf_Sr

  • Super Contributor
  • ***
  • Posts: 1729
  • Country: us
The only advice I can give is to ignore all posts that claim a user interface is good/bad but without giving any evidence for that claim. It could be based on something esoteric idea that makes no difference in real use
I concur, it's meaningless to say the Snigdent FUS2354X Minus has a "terrible UI" unless I qualify it in some way.

You can say I think it's terrible that the Snigdent FUS2354X Minus has a slow response or it doesn't allow zooming out in standard usage patterns.
If at first you don't succeed, get a bigger hammer
 

Offline doppelgrau

  • Contributor
  • Posts: 37
  • Country: de
Siglent:
 - 10 Bit mode (100 Mhz)
 - 50 Mhz Signal generator (1 Channel)
 - FFT up to 2 Mpts
 - Easier to hack (Keygen)

Rigol:
 - 25 MHz Signal generator (2 Channel)
 - FFT "only" up to 1 MPts
 

Offline Elasia

  • Frequent Contributor
  • **
  • Posts: 726
  • Country: us
Obviously you did not look at the video!
Check from 00:35 and beyond!

It is perfectly possible to capture 200Mpts at ALL time base settings.
The scope just has to sample all the time, and the moment it triggers it will display that part on the screen, but it means that it has captured data before and after.

Why doesn't it show that data when Dave zooms out? Even my Rigol DS1054Z can do that.

(Attachment Link)

They made an oddball design choice, its there and works, just not the way people are used to doing it and it requires setup vs what most would expect of just normal zoom in / out with no setup

As far as comparing to rigol, they are similar enough that getting one or the other will get the job done for the most part

The main difference comes down to how you use it and personal preference / feature set

The pic is the setup to do zoom out, its.. interesting... but it is kinda growing on me, i still wish it could be hidden though or even allow a selectable option for hide / small slider / preview slider
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Siglent:
 - 10 Bit mode (100 Mhz)
 - 50 Mhz Signal generator (1 Channel)
 - FFT up to 2 Mpts
 - Easier to hack (Keygen)

Rigol:
 - 25 MHz Signal generator (2 Channel)
 - FFT "only" up to 1 MPts

Rigol:
30% Cheaper
400Mpts memory

Siglent:
200Mpts memory

 

Offline Noy

  • Frequent Contributor
  • **
  • Posts: 361
  • Country: de
Siglent:
Ugly UI (Font, Symbols) looks "cheap Chinese"
Shared Channel knobs (biggest issue for me)

Rigol:
Single Channel knobs!! Love them..
50Ohm missing (passthrough termination works, solves the issue.)
Color of the channels ist a bit "questioning"
 
The following users thanked this post: doppelgrau

Offline doppelgrau

  • Contributor
  • Posts: 37
  • Country: de
Where did @Fungus find the 400Mpts memory for the Rigol?

I find in the datasheet (https://www.batronix.com/files/Rigol/Oszilloskope/MSO5000/MSO5000_DataSheet_EN-V2.0.pdf)
Rigol:
 - 200 Mpts (1 Channel), 100 Mpts (2 Channel), 50 Mpts (3-4 Channel)

Siglent:
 - 200 Mpts (up to two channels), 100 Mpts (3-4 Channels)
 - automatic switching between 1x/10x (as selected on the probe)
« Last Edit: June 11, 2020, 02:44:34 pm by doppelgrau »
 

Offline jemangedeslolos

  • Frequent Contributor
  • **
  • Posts: 386
  • Country: fr
It is not really 8GS/s, it's 2x4 and 2 x 100 Mpts for the Rigol.
They can do interleaving so they are able to do 1x8GS/s and 1x200 Mpts which is good but you have to be careful when comparing with others scope.

It is not 1x2 GS/s and 1x200 Mpts for the Siglent but 2x2GS/s and 2x200 Mpts. It can put things in other perspective.
I don't know why they can't do interleaving, if it's by choice or if the architecture don't allow that.

The Siglent also has a 10inch screen, the bigger the better.
I don't have the Siglent but I don't find the IU cheap.

Rigol is trying to make a modern user interface with unusual colors on this kind of instrument.
I wouldn't say I like the interface but I don't hate it either. I would prefer that there is less space lost.
At least they try something without trying to copy someone else.

One thing that I don't understand is that we can find 10inch full HD touch screens for nothing today.
Why are oscilloscope manufacturers limited to such low resolutions ?

But to answer the question, it is difficult to choose between these two. The price difference is huge. 35-50% more depending on whether we compare to MSO5072 or 5074.
« Last Edit: June 11, 2020, 03:06:40 pm by jemangedeslolos »
 

Offline Elasia

  • Frequent Contributor
  • **
  • Posts: 726
  • Country: us
One thing that I don't understand is that we can find 10inch full HD touch screens for nothing today.
Why are oscilloscope manufacturers limited to such low resolutions ?

More pixels the more processing power / time you need to fill those pixels.. more cost.. whats the trade off gain? questionable for a budget scope

Still think at least the 2k+ line should have had an hdmi port... the web interface is nice but the refresh rate is bleh
 

Offline doppelgrau

  • Contributor
  • Posts: 37
  • Country: de
Since I'm currently unsure which I want to buy to replace my MSO1104Z I try to sum up all "hard" difference

Rigol:
  • Higher sample rate (twice, for one channel four times as high)
  • 350MHz probes included
  • Signal generator with two channels, higher sample rate, lower frequency (25MHz), lower amplitude (+-5V/+-2.5V)
  • Single channel knobs
  • about 30% cheaper
  • hdmi port

Siglent:
  • Twice the memory if using 2-4 channels
  • 10 bit mode
  • Signal generator with only one channel, higher frequency (50MHz), higher amplitude (+-6V/+-3V)
  • Bit better FFT (2Mpts instead of 1)
  • Automatically switches between 1x/10x setting (as set on the probes)
  • up to 500MHz (but included probes only good for 200MHz)
  • internal 50Ohm termination possible
  • Keygen (easier)

So besides differences in the GUI, both seem to have different strength.
And I'm still unsure which is better, but currently slightly leaning towards the Siglent I think it may fit my needs a bit better (10 bit mode, more memory on multiple channels, guess won't need the higher frequencies above 200 MHz so the probes won't be an issue), but to be honest, that's mostly guessing.
 

Offline tv84

  • Super Contributor
  • ***
  • Posts: 3212
  • Country: pt
(I had promised myself that I wouldn't participate but... I'll repeat myself...)

The only difference is price! If you have $1500 buy the Siglent. If you only have $1000 buy the Rigol 5074.
 
The following users thanked this post: Fungus, Elasia

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Where did @Fungus find the 400Mpts memory for the Rigol?

My bad, it's 200mpts.

OK, take that out and the difference is price. One costs hundreds of $$$ more than the other so you'd expect it to be better, comparing them directly is silly*.

eg. We could easily compare the Siglent with something more expensive from R&S and find the R&S is the one to buy. It's turtles all the way up.


(*) That never stopped people directly comparing the Rigol DS1054Z with the Siglent SDS1104X-E though :-//
 

Offline doppelgrau

  • Contributor
  • Posts: 37
  • Country: de
Well, the price difference is partly since we compare a 70MHz scope with an 100MHz scope (same Bandwidth bought less than 10% difference - so I guess they have the same target group).
Also both are the two cheapest models if you want high memory and sample rate and touch screen, so in my eyes it makes sense to compare them, is it worth to spend the extra money.
 

Offline tv84

  • Super Contributor
  • ***
  • Posts: 3212
  • Country: pt
Well, the price difference is partly since we compare a 70MHz scope with an 100MHz scope (same Bandwidth bought less than 10% difference - so I guess they have the same target group).
Also both are the two cheapest models if you want high memory and sample rate and touch screen, so in my eyes it makes sense to compare them, is it worth to spend the extra money.

Don't look at the sticker! You know perfectly well that both scopes can be software enhanced so compare them in full glory.

Again: does the MSO cover all your needs? Forget which is better, because they are alike but different. If the MSO covers your needs, stay with the MSO. If you have the $$$ burning your pocket go for the Siglent.

PS: If the Siglent 2074X+ was available outside China it would be a no-brainer!
« Last Edit: June 12, 2020, 07:46:45 am by tv84 »
 

Offline macaba

  • Regular Contributor
  • *
  • Posts: 203
  • Country: gb
MSO5074 owner here. I had the same question and it just came down to price. Ideally I would have been able to compare the two side by side in real life.

Mostly happy with the MSO5074, there are some things that could be better; UI lag is one that bothers me. On a stopped acquisition, you can move the trace up and down fairly smoothly with the knob but the UI display of the channel offset takes ages to update. Also, start doing SCPI over TCP causes the UI to really lag (the UI has crashed a couple of times but it’s rare). I’ve seen small anecdotal reports that the Siglent suffers similar UI issues but we’d have to establish a standard testing methodology to compare the two.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf