Author Topic: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004  (Read 809484 times)

0 Members and 1 Guest are viewing this topic.

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2550 on: September 20, 2019, 06:47:18 pm »
Just checking if this is intended behaviour - I can save a waveform file, but there seems to be no option to re-load it. (RTM3004)
IIRC you should be able to load it into a reference memory trace. But it may depend on the format in which the waveform is saved.
No - that only allows loading CSV files. Also probably not useful if you want to look at decodes on a stored trace
Well, if you save as CSV then you should be able to load it. But that does limit the possibilities. Decoding a stored trace is a rare feature -if it even exists- on an oscilloscope. On the RTM3004 the reference traces are decimated versions of the original trace anyway and I guess the same is true for the RTB2004; it would be difficult to deal with references traces at the full memory depth.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6447
  • Country: hr
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2551 on: September 20, 2019, 10:52:28 pm »
Decoding a stored trace is a rare feature -if it even exists- on an oscilloscope.
What do you mean ? That is a normal thing for Picoscope. That was one of few main reasons for me to get it.
And every scope that decodes in software can do that trivially if you can load trace. Load trace and enable decode.
 

Offline 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1570
  • Country: de
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2552 on: September 20, 2019, 11:08:27 pm »
I wouldn't say that every scope can do this but AFAIK all real LeCroys of the last ten years or so can save and restore the complete scope state including trace data and allow you to work with it after restoring as if the waveform was just captured. I used that on the WaveRunner 6Zi in 2012 or so but AFAIK this feature even exists in the WaveSurfer 3000 ("LabNotebooks").
« Last Edit: September 21, 2019, 12:43:59 am by 0xdeadbeef »
Trying is the first step towards failure - Homer J. Simpson
 
The following users thanked this post: 2N3055

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2553 on: September 20, 2019, 11:51:19 pm »
These two are probably the only exceptions then.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 13694
  • Country: gb
    • Mike's Electric Stuff
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2554 on: September 21, 2019, 09:05:44 am »
It doesn't seem to me to unreasonable to want to save & restore the exact scope state, more so with the large memories of modern scopes.
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 
The following users thanked this post: egonotto

Offline Harjit

  • Regular Contributor
  • *
  • Posts: 141
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2555 on: September 24, 2019, 04:39:36 pm »
Scope: RTB2004
FW: Current - sorry, I deleted the info. by mistake and am not near the scope
Bug: Horizontal setting doesn't follow zoom factor

Steps:
  • Capture waveform
  • Using touch screen zoom into area by drawing a rectangle
  • Rotate horizontal rotary switch

Expect: Start zooming from the "zoomed" in horizontal scale or closest "normal" value. Where "normal" means what you would get if you were using the horizontal rotary switch.
Observed behavior: Horizontal scale starts at 20microseconds or some such non-related value

 

Offline Harjit

  • Regular Contributor
  • *
  • Posts: 141
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2556 on: September 24, 2019, 04:45:29 pm »
Scope: RTB2004
FW: Current
Enhancement request: During zoom, show relative horizontal scale along with absolute value

When using the scope in non-zoom mode, the trigger point has a horizontal value of zero and every vertical graticule is marked with the relative time offset from the trigger point.

I was looking at a 10 second capture and I was zoomed into an area around eight seconds. The vertical graticules were 8.aaaa, 8.bbbb, 8.cccc, etc.

It was not fun to do the math or turn on the cursors to measure the time between events.

If there was an additional row that showed the delta time between the graticules, it would have made life a lot easier.

As I write this, I realize that it may be marginally helpful. Perhaps it is something the R&S team can play with to see if there is a way to do horizontal timing estimations easier when using zoom.
 

Offline genghisnico13

  • Regular Contributor
  • *
  • Posts: 56
  • Country: cl
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2557 on: September 25, 2019, 10:54:59 pm »
Hi Rich, I have a couple of suggestions and Bugs:

In some cases changing the Probe Factor changes HW gain (even when changing in 10^n steps) and because of this the offset can be altered also, for example when changing from 0.1:1 to 10k:1 it does it, but when changing from 1:1 to 10k:1 it doesn't. The expected behavior by me would be to still see the waveform exactly the same, but have all vertical numbers scaled accordingly. Also is ti possible to extend the Probe Factor range to 1M:1?

X-Y: I understand is not a mode that is used that often so probably not a priority, but would it be possible to use a Math channel as source?
Could Dy/Dx be added to cursors for example to measure Resistance in a I-V plot.
There is some weird behavior with High Resolution Enabled on timebases longer than 50ms/div included, it makes a couple of incomplete acquisitions and then glitches out and repeats the cycle, a couple of incomplete acquisitions and a glitched one.
Channel 1 is always shown in the CH1-t plot even when disabled(and even when being used as Y), not a big deal(at all), but different from the behavior of the other channels.
Limited resolution in the readout of cursors of y channels when using high probe factors like 800V/A, when moving the cursors you get the expected resolution in the popup readout(ex. 1.44uA), but not in the cursors band/section(Y1,Y2,DY)(ex. 1uA).

Digital Channels: Would it be possible to have a different color for 1 and 0? sometimes is not that easy to figure out fast, if you don't have a transition on screen.

FFT: Would it be possible to have the scope store the "Timebase" setting when enabling FFT and recall it after disabling it? the same with cursors position, source and type.
When setting a center frequency lower than the span/2 an error is displayed, wouldn't it be better to adjust the span to the requested center frequency*2?
As others have mentioned I would also like to have at least 1 more variable to control the acquisition time or samplerate, the function that the timebase used to do before the FFT change. in  my case I wanted to look at a portion of the spectrum but because I couldn't control the samplerate I had to deal with aliasing.
Maybe to make use of the encoder buttons in this mode you could store the current center frequency by pushing and holding the horizontal position encoder for 0.5-1 sec and if you then move the center you can recall it by pushing the encoder without holding it. the same with the Vertical position/Reference level., so it can act similar to the normal scope mode. perhaps could be applied to Bode too.

Measurements: opening the measurement type screen and then changing measurement number doesn't closes the meas. type screen or makes it change to the currently selected meas. number, so when you select a type of meas. you are modifying the previous meas. number.

Bode: Would it be possible to continue sub-dividing horizontally when zooming in? Right now you don't get a horizontal grid when you are between the Decade ticks(ex. 1.0-1.1kHz).
I would like to maybe set the number of points between start and stop frequencies, just because sometimes 500 points per decade isn't enough when you are only interested in small spans. The over-engineered solution could be to use something similar to amplitude profile, but for points per decade, allowing you to have a lot of points only in the areas of interest (this would be awesome).
Would it be possible to add an X in the Bode Tab to exit, similar to Measurements, statistics and Bus table tabs?(not a big deal)
Maybe it is by design, but when selecting the number of points per decade the menu doesn't close immediately when you select the number, requiring a double touch.
When scrolling lists of values (Bus table, Bode,etc) isn't the direction inverted? technically you could use both directions, but Clockwise to go down makes more sense because you are also scrolling through the points on the waveform to the right while you go down the table. I guess in the Bus table it makes sense because the waveform is moving to the left?

I understand this is unlikely, but would math with measurements be out of the question? ex. VRMS-CH1/VRMS-Ch2.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2558 on: September 26, 2019, 12:07:42 am »
Bode: Would it be possible to continue sub-dividing horizontally when zooming in? Right now you don't get a horizontal grid when you are between the Decade ticks(ex. 1.0-1.1kHz).
I would like to maybe set the number of points between start and stop frequencies, just because sometimes 500 points per decade isn't enough when you are only interested in small spans. The over-engineered solution could be to use something similar to amplitude profile, but for points per decade, allowing you to have a lot of points only in the areas of interest (this would be awesome).
I second this suggestion!
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline casinada

  • Frequent Contributor
  • **
  • Posts: 599
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2559 on: September 27, 2019, 09:17:41 pm »
A new version of the RTB2000 manual posted on Sept 24. It has 7 extra pages. I haven't had the time to look on what was added.  :)
https://www.rohde-schwarz.com/us/manual/rtb2000/
 
The following users thanked this post: edavid

Offline maginnovision

  • Super Contributor
  • ***
  • Posts: 1963
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2560 on: September 27, 2019, 09:26:31 pm »
Maybe it's 7 pages about the limitations of uart decoder hard limit.
 

Offline KaneTW

  • Frequent Contributor
  • **
  • Posts: 805
  • Country: de
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2561 on: September 27, 2019, 09:58:57 pm »
No changes that I can see, just formatting in the SCPI section.
 

Offline KaneTW

  • Frequent Contributor
  • **
  • Posts: 805
  • Country: de
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2562 on: September 28, 2019, 11:26:48 pm »
@Rich, would it be possible to have more than 500pt/dec for the bode plot mode? I'm measuring ultrasonic oscillators and sometimes even 500pt/dec isn't enough.
 

Offline Rich@RohdeScopesUSA

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2563 on: October 08, 2019, 04:04:01 pm »
@Rich, would it be possible to have more than 500pt/dec for the bode plot mode? I'm measuring ultrasonic oscillators and sometimes even 500pt/dec isn't enough.
Hi KaneTW - sorry I've been so slow to respond - it's been a crazy few weeks!  I'll submit the request.   :-+

-Rich
 

Offline nictinkers

  • Contributor
  • Posts: 18
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2564 on: October 08, 2019, 08:11:03 pm »
For those with one of these, on the scale of dysfunctional-pain-in-the-arse to buttered-toast, how do you rate the sigrok / PulseView integration? How fast can it stream and how big is the sample memory?

I'm strongly considering cracking opening my wallet for one of these but I don't think my electronics designs will be including fewer serial buses in the foreseeable future. Equipping a scope that has 20 input channels running at up to 2.5 GS/s with just two serial decoders limited to 3 MBit was not the best design tradeoff R&S has ever chosen.

I know post-analysis is not the same as decoding on the scope, mostly because of triggering but if I can slink off to PulseView when I need more than one UART or things a bit faster, I'd feel much better about buying the scope.

I actually think R&S should spend some engineer hours contributing comprehensive drivers to sigrok, and making any firmware changes to make it all rock solid - that would really make these killer scopes.
« Last Edit: October 08, 2019, 08:28:26 pm by nictinkers »
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2565 on: October 08, 2019, 08:50:58 pm »
OTOH you have to think about the reality of needing UARTs at >3Mbit. A UART is usually clocked at 16x the bitrate so at 3Mbit this would be 48MHz. Also from my tests on the RTM3004 it turned out the protocol decoders in general work up to much higher speeds than specified. I wouldn't get so worked up about a limit you probably never run into.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline phs

  • Contributor
  • Posts: 26
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2566 on: October 09, 2019, 10:15:22 pm »
Though I don't have time to get into this further at this point, hopefully the following observations can help folks who are considering purchasing these beautifully engineered instruments.  As most already know, it really is important to try before you buy if at all possible.  That’s the best way to make sure an instrument will be able to handle the tasks you need it to.  Also, you need to learn everything possible about whatever machine you end up with.  Test and verify every function you plan to use!!  Even the RTB2000 locks up on occasion, and has some difficult to understand bugs that are still being worked through here. 

After having used, primarily, a range of old Tek, Agilent and Rigol scopes prior to the release of the RTB2k, and having increased needs for additional MSO functions with decoding, the RTB seemed to be a potentially worthwhile addition to the stables.  Primarily, RS232 decoding was needed here (at speeds at least up to 10 Mbps), but recently I2S, I2C and SPI functionality are becoming necessary as well.  I can’t comment on I2S/I2C/SPI at this time, as I haven’t thoroughly tested these functions yet on the various scopes.  A Rigol MSO5000 is also being evaluated and compared to the RTB2k, in the hopes that it will meet current and future decoding needs that the RTB2k can’t handle at this time.

So, after having had great success over the years with the Rigol DS2000 and DS1000Z series, thanks to a rich feature-set that has been critical to solving many development and troubleshooting issues, despite various rather minor bugs, I was really looking forward to getting the RTB2k's for the logic and decoding features, especially.  At the time of the purchase, there wasn’t any way to “demo” a unit before buying, so it was understood that there was some risk involved – especially with such a new platform with limited details available.

Within minutes of firing one up though, I was quite surprised and disappointed to discover how limited the feature-set on the RTB was/is.  I had slightly expected that R&S would absolutely pack the thing full of advanced features – especially in comparison to the “lowly” Rigols.  At that time (launch edition period), there were only 4 math functions, and the wireless mouse functionality wasn’t there.  Thankfully at least those two issues were fixed subsequently, so hat's off to R&S for that, and also for adding the Bode analysis functionality. 

Somewhat sadly though, to this day, the DS2000 machines continue to get used over the RTB's because the faster decoding is necessary for so many of the projects here.  (Yes, there are LA’s used too, but most of the time it’s necessary/desirable to see the actual signal.) 

While the RTB2k is a great scope in so many ways, it turns out there are some surprising limitations for what is needed for the typical development projects done here -- especially for a scope in this price range.  Though every scope has limitations and bugs, at the end of the very long development day, the necessary features to get the job done are the priority.  Bugs, as long as there are workarounds, are just something you learn to deal with, and unless they cause lock-ups or crashes, aren’t understandable, or in some other way waste time, really aren't a big deal -- as long as you take the time to understand each bug and the workarounds.  A much bigger issue is if a critical feature is completely missing though, or if it’s so buggy that it’s not usable.

The biggest RTB2k features that are truly outstanding for typical use scenarios here, are:

o  The web interface (really well done, and the preferred way to “drive” the scope).

o  Capability to use a mouse for direct control (however, there are some UI improvements that would be welcome, to more easily enable driving the scope entirely by mouse).  One great advantage is no more achy arms from reaching out to the rack and twiddling knobs for hours on end, and also that the scope can be placed in more convenient locations.  Since you don't have to physically touch the instrument -- it can be controlled remotely.  And, a favorite here is the wireless Logitech M570, which now works very well, and allows the user to even stroll around while operating the instrument.  Brilliant!

Side note:  Personally, I still don’t like using touch screens on lab equipment for a variety of reasons, other than for entering lots of data, which isn’t too frequent with scopes in my experience.  Using a mouse is so much more convenient, if the UI has been designed properly.  And, when the scope can be placed close to the user, using knobs/buttons still works very well and can be pretty automatic, if the setup/layout is designed properly.

In the case of driving from the physical knob/button scope controls I personally find the Rigol MSO5000 intuitive and easy to use, and once you learn the menu structure (which doesn’t take long), it is an efficient and satisfying way to control the machine.  Even the touch buttons on the screen are quite snappy on the Rigol – It’s just when you want to drag stuff around on the screen via touch that it gets really laggy.  The RTB is laggy in this case, too – maybe not quite as much, though.  The Rigol also works very nicely with the Logitech M570 wireless mouse.

Driving the RTB from the physical controls seems less intuitive than the Rigol to me, but then I’ve been using the Rigols for many years.  It seems the RTB forces you to use the touch screen for some functions, rather than allowing full control via the buttons/knobs.  This is disappointing.  And not having separate vertical controls is also disappointing.

o  Useful (but very basic) functionality and display in FFT mode that visually makes it easy to see what's going on.

o  Decoding UI is really well done -- visually appealing and very functional.

o  Decoding is quite smooth, reasonably feature-rich (with the limited available protocols), despite the significant speed limitations.

o  Screen really is nice!  (Both size and image quality.)  Still, I don’t normally like using touch screens on lab equipment, except for typing in license codes or labels.  Dragging around waveforms on a scope lacks precision and is quite laggy in this case (also true of the Rigol MSO5000)

o  10-bit resolution.  Really not needed here for the vast majority of the time, but maybe it will be useful at some point.  Certainly nice to have, but not a critical function.

The most significant issues:

o  Significant (for purposes here) UART decode speed limitation.  And, it is a hard limitation as I recall, with no workarounds.  Beyond 3 Mbps, the decoding simply doesn't work.  Even the ancient Rigol DS2000 series decodes at least up to 8 Mbps -- although it can be a bit painful, it gets the job done, and it is used for this quite a lot, to this day.

o  Limited decoding types available.  No I2S, FlexRay, 1553B, if you might need any of these (MSO5000 has these available, DS2000 does not, and the DS2000 lacks LIN).  I2S, in addition to I2C and SPI, is becoming more important for projects here.  And, only about 1.5 channels of decoding at one time...ouch!  Even the DS2000 allows two fully independent TX/RX decoding channels.

o  No Nth-edge trigger mode.  This is something used a lot (on various Rigol scopes), and there is no runt trigger capability, though there is a search function for runt pulses, at least.  Also, a rather limited range of trigger options, in general, as compared to the Rigol DS2000 series, and even the lovely little DS1000Z (which has paid for itself many many times over the years).

o  No "Vertical Expansion" options, as in the Rigols.  This allows vertical zoom to operate referenced to the center of the screen, which makes it quite easy to precisely zoom in all the way to a DC-offset waveform (for checking power rail noise, for example) without having to use AC coupling.  This is used all the time here, and I haven't found a way to easily and precisely do this on the RTB2k's.  It's been mentioned before, but I don't believe this is likely even on the radar for future RTB "upgrades" for some reason.  Maybe I’m just missing something...

o  Mechanically horrible probes.  There is a lot of “stiction” that often prevents the probe “hooks” from properly grabbing a contact point, and it’s hard to rotate the probe clips on the probes easily.  They’re the worst feeling probes I’ve ever used.  Working with a limited budget here, this is quite disappointing, as buying a set of probes in addition to the relatively high cost of this scope, is quite painful.  Especially when the probes offered by much cheaper brands have far better mechanical operation and feel.

o  The Logic Probe cable is quite a crude design.  Certainly doesn’t facilitate unplugging from the scope with any confidence that there is not undue stress being placed on the cable.  For very little extra money, R&S could have designed a much better way to plug it into the scope.

Now, a Rigol MSO5000 is being evaluated, with very promising results so far, despite some minor bugs that remain, and the crude, by comparison, web interface, and slightly more limited mouse control.

Some apparent advantages of the MSO5000 (still testing, so these observations may or may not be true):

o  More comprehensive trigger functionality, including Nth edge and runt.

o  4 complete decode channels, each of which are spec’ed to operate at up to 20 Mbps (not tested here yet).

o  More decoding options, including i2s, which will be very useful if it works even reasonably well.

o  Vertical Expansion functionality.

o  Independent vertical channel controls.  Scope can be fully driven with knobs/buttons, without having to touch the screen.  Yay!!  And, to me the shape of the knobs on the Rigol feel a bit better than the RTB, but that’s likely a rather silly distinction.

o  Snappy UI (unless you’re dragging around waveforms via touchscreen or mouse).  Now, not as “snappy” as a current Keysight, no doubt, but certainly not unpleasant at all.  Running the FFT slows everything down, though, but this is also true for the RTB.  Love the operation of the knobs/buttons, and the UI is quite intuitive once you get used the the “Rigol” way.

o  Screen is plenty bright here –  pretty close to the RTB.

o  Two AWG’s (nice to have, I suppose, but not critical here).

o  HDMI port.  This is handy to have, despite and in addition to the Web UI, as it allows for another, larger, view of the scope screen.

o  Great feeling probes, as with all Rigol probes I’ve ever used.

o  Four FFTs available simultaneously.  Not sure how usable this is ultimately, but at least it’s there...

Some early observations of things I don’t like about the MSO5000:

o  FFT looks awful in comparison to the RTB unless you turn on “Color Grade”, and then it gets really slow, especially if you have all four FFTs running (which isn’t an option on the RTB).  It even looks bad compared to the DS2000, so I hope this is something Rigol addresses in the near future.

o  The decoding UI is rather crude looking compared to the RTB and lacks some of the clever UI design elements of the RTB.

o  The web UI is extremely crude in comparison to the RTB, and really needs further work.

o  The network settings are screwy and it’s hard to get static IP parameters locked in.  Luckily, Wireshark monitoring doesn’t show the instrument “phoning home” yet, despite the bizarre network settings behavior.  The online upgrade option has not been tested as this is a STUPID function for any networked lab instrument (security/privacy), and will never be used here.  Micsig is trying this BS approach to firmware updates, too.  UNACCEPTABLE!!

To wrap up – while I love much about the RTB series, it is pretty expensive for the functionality, as compared to the Rigols at least, and I’d guess as compared to Siglents and others, as well.  To get faster/additional decoding options with R&S, you’d have to pay for at least the next tier up and that’s well over $10k, I would guess. 

At this time it appears that even the low-end Rigols continue to offer critical functionality that the RTB lacks – which in this case keeps them in use here.  And, they are similarly enjoyable to operate in my experience – as long as you don’t have a passion for touch screen operation.  I also remain hopeful that the MSO5000 will fill in the rest of the functionality gaps.

So, none of this is intended to be comprehensive, and I’ve likely gotten things wrong.  It is just a basic, non-detailed overview of some of the observations based on use cases here, that I hope may be helpful to others.  I’m sure there are many folks who have requirements that are different, and for whom the RTB would be the perfect match.  If money becomes much less of a factor in the future, I will certainly be looking at some of the more advanced R&S scopes because there is so much about the RTB that I really do like.

One thing that makes the R&S products really stand out from the Rigol offerings is that Rich is here on this forum, actively listening to, and helping, users all of the time.  As for Rigol, hmm…..

Finally, and again –  don’t forget to play before you pay if at all possible, and make sure to learn the critical quirks of whatever instrument you end up with, lest you be tripped up for resting on your laurels.
 

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 13694
  • Country: gb
    • Mike's Electric Stuff
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2567 on: October 09, 2019, 10:32:49 pm »
OTOH you have to think about the reality of needing UARTs at >3Mbit. A UART is usually clocked at 16x the bitrate so at 3Mbit this would be 48MHz. Also from my tests on the RTM3004 it turned out the protocol decoders in general work up to much higher speeds than specified. I wouldn't get so worked up about a limit you probably never run into.
I regularly use 4 and 6Mbaud. The decoder max baudrate is a hard limit-you can't enter a higher value.
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline JxR

  • Supporter
  • ****
  • Posts: 352
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2568 on: October 09, 2019, 10:59:32 pm »
how do you rate the sigrok / PulseView integration?

I don't see the RTB2004 on the supported hardware list, so I would assume integration is non-existent.

I actually think R&S should spend some engineer hours contributing comprehensive drivers to sigrok, and making any firmware changes to make it all rock solid - that would really make these killer scopes.

Is there honestly any other company that has personally contributed to this open source project?  Why would R&S spend the money and resources themselves, when they can devote it to projects that do make them money?

If this is a very important feature to you, then it sounds like you have the perfect motivation to make the required drivers yourself.  All it takes is one motivated individual to get the ball rolling.
 

Offline nictinkers

  • Contributor
  • Posts: 18
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2569 on: October 10, 2019, 09:20:15 am »
There have been a couple of mentions of people in this thread working on drivers:
https://www.eevblog.com/forum/testgear/new-killer-scope-a-true-game-changer-from-rs-rtb2002-rtb2004/msg2297925/?topicseen#msg2297925
https://www.eevblog.com/forum/testgear/new-killer-scope-a-true-game-changer-from-rs-rtb2002-rtb2004/msg2534421/?topicseen#msg2534421

And it looks like a patch was written last year:
https://sourceforge.net/p/sigrok/mailman/message/36471894/
But I don’t really know how to check if the patch was accepted, and without actually having a scope it’s hard to test if it works.

I spoke to a R&S rep at Electronex four weeks ago. He was polite, took my details and promised he’d organise a loan of a unit for testing but I think I made the mistake of being honest and admitting it was a personal purchase and that I didn’t represent a giant institution. Since then, emails sent to sales.australia@rohde-schwarz.com and to the distributor Rapid Tech have been met with stony silence. It is starting to be a factor against the purchase - if it takes weeks and weeks for them to bother getting back to me with a sales enquiry, heaven help me if I have an issue and need support.

Is there honestly any other company that has personally contributed to this open source project?  Why would R&S spend the money and resources themselves, when they can devote it to projects that do make them money?

We’ve all seen vendor-created companion software tools that are janky, Windows-only abandonware. I think R&S would do better than average if they rolled something themselves but I suspect it still wouldn’t be as good as leveraging the existing work (including great protocol decoders) in sigrok and PulseView. Internal engineers would have a major advantage to someone working on it from outside, as they have much better knowledge of the instrument’s communication protocol, and could even adjust the firmware if necessary to make it as good as it could be.

Would it make them money? I admit, its hard to say for sure. Does fixing bugs and adding firmware features for people after they’ve purchased their scope make the company money? I think verified, comprehensive sigrok / PulseView integration would ameliorate a weakness in the product and would increase its value to potential buyers.
 
The following users thanked this post: Doug_T

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2570 on: October 10, 2019, 09:43:27 am »
OTOH you have to think about the reality of needing UARTs at >3Mbit. A UART is usually clocked at 16x the bitrate so at 3Mbit this would be 48MHz. Also from my tests on the RTM3004 it turned out the protocol decoders in general work up to much higher speeds than specified. I wouldn't get so worked up about a limit you probably never run into.
I regularly use 4 and 6Mbaud. The decoder max baudrate is a hard limit-you can't enter a higher value.
But having bitrates over 3Mbaud still is rare in general. But you are right that the upper limit is hard coded. I forgot that you need to set the baudrate for UART decoding.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 8605
  • Country: gb
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2571 on: October 10, 2019, 11:44:26 am »
But having bitrates over 3Mbaud still is rare in general. But you are right that the upper limit is hard coded. I forgot that you need to set the baudrate for UART decoding.
If you look at external interfaces its rare for a UART to run at anything like 3Mbps. However, if you look at device to device interfaces within boards you'll often find people cranking UARTs to the maximum they can get. Quite a few engineers looking at new devices show serious interest in how far the UARTs can be cranked.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2572 on: October 10, 2019, 12:30:56 pm »
But having bitrates over 3Mbaud still is rare in general. But you are right that the upper limit is hard coded. I forgot that you need to set the baudrate for UART decoding.
If you look at external interfaces its rare for a UART to run at anything like 3Mbps. However, if you look at device to device interfaces within boards you'll often find people cranking UARTs to the maximum they can get. Quite a few engineers looking at new devices show serious interest in how far the UARTs can be cranked.
IMHO using an SPI port would be a much better solution for such a purpose. Often these have buffers and support 16 or even 32 bit words so less overhead for transmitting / receiving. Also the send / receive bitrates wouldn't have to be a very close match.
« Last Edit: October 10, 2019, 12:48:49 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline JxR

  • Supporter
  • ****
  • Posts: 352
  • Country: us
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2573 on: October 10, 2019, 01:50:13 pm »
We’ve all seen vendor-created companion software tools that are janky, Windows-only abandonware. I think R&S would do better than average if they rolled something themselves but I suspect it still wouldn’t be as good as leveraging the existing work (including great protocol decoders) in sigrok and PulseView. Internal engineers would have a major advantage to someone working on it from outside, as they have much better knowledge of the instrument’s communication protocol, and could even adjust the firmware if necessary to make it as good as it could be.

Would it make them money? I admit, its hard to say for sure. Does fixing bugs and adding firmware features for people after they’ve purchased their scope make the company money? I think verified, comprehensive sigrok / PulseView integration would ameliorate a weakness in the product and would increase its value to potential buyers.

I can certainly understand the desire to have such integration.  But, you have to also understand that R&S doesn't provide any free protocol decoders in their oscilloscopes.  Every single one of them is sold as a license you have to purchase individually, or in a bundle for the RTB/RTM/RTA series.  Sigrok would be competing against their current licensing model. Providing free drivers for Sigrok would most likely hurt the sale of these licenses.

R&S currently doesn't even allow RTB owners to purchase the decoder licenses that are available in the RTM and RTA oscilloscopes (which I personally think is a mistake).  Sigrok provides these decoders, and others that even their more expensive models don't have.
« Last Edit: October 10, 2019, 02:00:53 pm by JxR »
 

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 13694
  • Country: gb
    • Mike's Electric Stuff
Re: new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004
« Reply #2574 on: October 10, 2019, 03:26:35 pm »
But having bitrates over 3Mbaud still is rare in general. But you are right that the upper limit is hard coded. I forgot that you need to set the baudrate for UART decoding.
If you look at external interfaces its rare for a UART to run at anything like 3Mbps. However, if you look at device to device interfaces within boards you'll often find people cranking UARTs to the maximum they can get. Quite a few engineers looking at new devices show serious interest in how far the UARTs can be cranked.r
Rs485 can easily do 10Mbaud. Another reason for fast uart rates is when sending realtime debug data out
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf