Rohde & Schwarz RTB2004, 100MHz or Siglent SDS1104X-E

[Harjit]

Will be used for robotics i.e. microcontrollers, brushless dc motor driver and control, analog sensors, power measurements, signal integrity measurements on high speed SPI (10MHz to 30MHz) and UART busses (6Mbps).

I have a USB logic analyzer that I can use for the SPI and UART work, so integrating them into the scope isn't mandatory.

I know there is a big price difference but I'm tempted by the larger display and the 10 bit ADC VS 8 bit ADC on the R&S VS Siglent.

Would love to get people's perspectives.

Thanks in advance.
Harjit

[Rich@RohdeScopesUSA]

Hi Harjit - I'll let the others give you advice on which way to go, but I will say - wait until Monday (5/20) before you make any decisions 

-Rich

[exe]

Hi Harjit - I'll let the others give you advice on which way to go, but I will say - wait until Monday (5/20) before you make any decisions 

I'm intrigued, will you post here an update?

[Rich@RohdeScopesUSA]

Hi Harjit - I'll let the others give you advice on which way to go, but I will say - wait until Monday (5/20) before you make any decisions 

I'm intrigued, will you post here an update?

Happily!  One more teaser - we've heard feedback from the EEVBlog community (and others) loud and clear.  Is what's coming go to be for everyone?  No.  But I think it will get a lot of folks pretty excited.  Stay tuned 

-Rich

[Hydron]

Would this have something to do with the fact that it's ~6 months since the last RTB2k firmware drop (and that there was a very recent 3k series FW release)?

Edit: I might be a bit off the mark about a simple firmware release (thought the last one came with nice new goodies) given Rich's comment in the Tek thread. Upgraded RTB2k FW would always be welcome though!

[jemangedeslolos]

So if I planned to buy Siglent SDS5034x this week, I better have to wait  ? 

[Rich@RohdeScopesUSA]

Would this have something to do with the fact that it's ~6 months since the last RTB2k firmware drop (and that there was a very recent 3k series FW release)?

Edit: I might be a bit off the mark about a simple firmware release (thought the last one came with nice new goodies) given Rich's comment in the Tek thread. Upgraded RTB2k FW would always be welcome though!

Different.  Rest assured there will more RTB2K FW updates.  But this is broader (and way bigger IMHO)

-Rich

[Rich@RohdeScopesUSA]

So if I planned to buy Siglent SDS5034x this week, I better have to wait  ? 

I would.  But I'm biased 

-Rich

[jemangedeslolos]

 

I hate any kind of teasing, Im sweating right know.
I asked for a price for a RTB2004 at Batronix fews weeks ago but there are feature that are missing vs SDS5000x series.
If there is a significant promotion on RTM3000 and if I am mad enough...i will be so happy to work with R&S everyday but Im sure I will not have the money.
Do you know if it is possible to have an RTM3000 few days for test in France ?
Sorry Harjit for hacking your topic.

[Hydron]

Back to the OP's topic - I really like my RTB2004 (the screen especially is miles ahead of anything else I've used in the price range) but given you're requirements I'd be aware that it has some frequency limits on it's decode functions (possibly lower than those on the siglent).

Signal integrity wise, you're much less likely to get artifacts measuring fast signals on the R&S (with a minimum sample rate of ~twice the nyquist rate with all channels used in the 300MHz model) than the siglent (both 100 and 200MHz models in 3-4 channel mode sample at very close to the nyquist rate - the 100MHz limit is done in software).
If this doesn't make a lot of sense then have a look at measurements in the other Siglent/Rigol threads - one of the ways they manage such great value is skimping a little on sample rate; not an issue in many use cases but fast digital signals can have energy that gets through the (gentle rolloff) analogue filters and cause aliasing and other nastiness. I bring it up because you specifically mentioned signal integrity of fast digital buses as a key point (not trying to bash the cheaper options - I've owned a Rigol in the past and if I didn't have the RTB2004 launch promo scope I bet a SDS1104X-E would be on my desk instead!).

Finally see here for the "big news":
https://www.eevblog.com/forum/testgear/new-killer-scope-a-true-game-changer-from-rs-rtb2002-rtb2004/msg2418216/#msg2418216
(big promo on fully loaded R&S scopes)

[TK]

The price difference between the 2 scopes is huge, if you don't have $$$ restriction, go for the R&S

[Harjit]

Rich - thank you for the heads up.

Hydron, when you zoom into a waveform, does it zoom at the current center or at the trigger? I'm asking this poorly. Hopefully you get what I'm asking.

I follow your comment about Nyquist and oversample. I'm sure you will agree that the RTB2004's 1.25GS/s relative to 100MHz bandwidth, is a good oversample ratio. IMHO, a 500MS/s for a 100MHz waveform is pushing it a bit but for a 50MHz or lower waveform, it should be fine.

I think the Siglent is a great scope / value which is why I'm on the fence. I can probably do everything I need with the Siglent but can get the $'s allocated to get the R&S. The larger display is (40+% on each axis) is also attractive. I'm on the fence about touch. I've used Tek scopes with it and it didn't blow me away. I worry about smudges on the display from touching it.

Sounds like the launch promo was like the new promo but $1500 or so less?

In looking at the really high bandwidth scopes, they all use 8bit ADCs. This tells me that 8bits is enough and if you think about it, looking at a waveform on a screen, 8 bits is fine. If you have a multimeter, you really want the resolution there.

[genghisnico13]

If by Zoom you mean changing the Timebase, then it zooms to one of 3 defined horizontal positions (Center, Left or Right) of the screen, the sides are 1 or 2 divisions from the edge, I don't remember how many exactly.

[EEVblog]

The RTB2004 is a vastly better scope. You ordinarily wouldn't even compare these two as they are totally different price brackets.
If someone else is paying, buy the R&S 

[EEVblog]

In looking at the really high bandwidth scopes, they all use 8bit ADCs. This tells me that 8bits is enough and if you think about it, looking at a waveform on a screen, 8 bits is fine. If you have a multimeter, you really want the resolution there.

The 10 bits is really nice. You can zoom in and see extra detail after capture, among other benefits.

[Harjit]

@EEVblog - thank you for the insights. Since you have both, you are probably one of the few who has experience with both.

@genghisnico13 - yes, I meant timebase. Thank you for figuring out what I meant and not what I said :^)

I believe on the Tek, if the trigger isn't in the center of the display and you change the timebase, the trigger "will go flying out to the left or right".

I think on Agilent, the trigger will stay put at the screen location and changing the timebase will change the timebase of what you are looking at.

I was reading a post and on the Siglent SDS1104X-E, IIRC, this is selectable on "page 4 of the utility menu".

[EEVblog]

@EEVblog - thank you for the insights. Since you have both, you are probably one of the few who has experience with both.

As I said, they are vastly different, and the price difference reflects this of course.
Comparing them is really apples and oranges.

[tautech]

I was reading a post and on the Siglent SDS1104X-E, IIRC, this is selectable on "page 4 of the utility menu".

Exactly right however keeping the trigger position on the display is only an issue when you have moved it off the default 0s position.
For the X-E you can select it to be fixed to any set position where it remains irrespective of timebase settings.

Before you leap and buy that nice R&S check some of Daves videos and the terrible screen glare they have.
You'd be best to plan to get a screen overlay as part of your budget.

[EEVblog]

Before you leap and buy that nice R&S check some of Daves videos and the terrible screen glare they have.
You'd be best to plan to get a screen overlay as part of your budget.

Yes, it's the worst glare of any of my dozens's of scopes.

[tautech]

Before you leap and buy that nice R&S check some of Daves videos and the terrible screen glare they have.
You'd be best to plan to get a screen overlay as part of your budget.

Yes, it's the worst glare of any of my dozens's of scopes.

Didn't they also have questionable input BNC microphonics ?

[exe]

Didn't they also have questionable input BNC microphonics ?

Just don't bash your scope

[nctnico]

Before you leap and buy that nice R&S check some of Daves videos and the terrible screen glare they have.

Nonsense. In practical use the glossy screen on the RTB2000 and RTM3000 is not a problem at all even with a lamp from above. The display is bright with good contrast.

From a functional point of view there isn't much difference between the oscilloscopes discussed here. The RTB2000 however has a more polished user interface, a bigger screen and 10bit versus 8 bit. I think the OP should consider whether this is worth the extra money (even with the 42% off deal which R&S seems to have at this moment).

[Pinkus]

I should mention that the RTB2000 has one major limitation which draw me away from it quickly after realizing it: while it says it allows two decoders at the same time, this is cheated, as each data directions counts at one decoder (Rx+Tx at RS232=2 decoders). I know no other manufacturer counting decoder this way - for all other I know, one decoder is always a full duplex decoder.
In fact this made me really angry when realizing it, as it just an intentional crippling to sell more of the RTB3000 series scopes. Without this limitation I would have kept my RTB2004.
Examples: you cannot work e.g. on something with two SPI devices at the SPI bus as you would need 2x MISO (one active on CS1 the other on CS2), 1xMOSI which counts as 3 decoders.
Or working on a MCP2515 CAN controller: here would would need to see SPI MOSI (first decoder) MISO (second decoder) and CAN (third decoder).
Or think of a system with RS232 and I2C- you would need Rx, TX, I2C = 3 decoders
or RS232 and SPI: 4 decoders needed.
or 2 SPI devices on the SPI bus: 4 decoders needed
All other manufacturers (I know) will only need 2 decoder for the above.
Thus, for some hobbyists this limitation might be OK, but with with higher requirements these limitation can quickly be a problem.
Even the lowest series of Siglent, Rigol, Keysight (I assume even the 1000 series, but I don't know) all have at least 2 full duplex decoder - thus no problem with all these examples above; the RTB2000 can't (RTB3000 series can, so you might take a closer look there).
If you are working a lot on digital projects, think what you might need now and in the future.

[nctnico]

Most of the low-end scopes only have one full-duplex decoder which can be set to one protocol at a time so it is not really a limitation that the RTB2004 supports only two decoder channels if you compare it to the competition. At least you can look at two different protocols in half duplex.

[2N3055]

On MSO5000 you have 4  duplex decoders. On little 1000Z you have 2 duplex decoders..

I also have problems with feature set choices with R&S. Originally RTB2000 had math that was same as DS1000Z, they realized they went too far so they added more functions.
Also, there is no search function that can search on basic I2C UART or SPI protocols on any of the 2000/3000/4000 series..  And no search through segments. So you have 400 MPts of memory, you grab 20000 segments and then you have to twidle MANUALLY through 20000 segments and VISUALLY check in decode if there is packet you are looking for...

So pretty much you won't use any of that huge memory...