Author Topic: DS1054Z record mode and serial decode  (Read 2881 times)

0 Members and 1 Guest are viewing this topic.

Offline xyzzy

  • Contributor
  • Posts: 13
DS1054Z record mode and serial decode
« on: October 13, 2017, 05:26:20 am »
Greetings!

I'm in the process of debugging some SPI code using my DS1054Z.  The serial decode feature works ok (seems to work only on data displayed on-screen rather than on full memory depth, but that's not holding me back this time), but when I tried to use the record (aka segmented memory) function to capture SPI packets, I discovered that I can't use record and serial decode at the same time.  Whenever one is enabled, the option to enable the other is grayed out!

Is this a PEBKAC error, or is it really impossible to have the scope tell me the contents of the packets it just collected?  Segmented memory is of limited value if I have to decipher all the bits in each packet manually.

Please let me know if anyone can recreate this bug(?) or if I'm just doing it wrong.

Thanks!
 

Offline ebclr

  • Super Contributor
  • ***
  • Posts: 1620
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #1 on: October 13, 2017, 05:37:00 am »
 

Online CustomEngineerer

  • Frequent Contributor
  • **
  • Posts: 448
  • Country: us
Re: DS1054Z record mode and serial decode
« Reply #2 on: October 13, 2017, 01:51:13 pm »
Don't have a DS1000Z so can't directly answer your question, however the DS2000A that I have does allow decoding on segmented memory, however it also allows decoding on the full memory, not just whats on screen like the DS1000Z. My guess would be this is intentional to help make the more expensive models more appealing. If the cheapest line did everything the more expensive scopes did, nobody would bother buying their more expensive lines (though have a feeling this pretty much happens to Rigol anyways). Seriously doubt this is just a bug.
 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 1224
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #3 on: October 21, 2017, 11:31:51 pm »
One option is to use the "Wave Inspector" of DSRemote. That way you can use SPI-decoding using the whole 24Ms on your pc.
You can zoom in and out and scroll through the whole buffer with serial decoding enabled.
The difference between theory and practice is less in theory than
the difference between theory and practice in practice.
Expensive tools cannot compensate for lack of experience.
 

Offline H.O

  • Frequent Contributor
  • **
  • Posts: 568
  • Country: se
Re: DS1054Z record mode and serial decode
« Reply #4 on: October 22, 2017, 01:15:14 am »
I don't have a 1054Z and I might be stating the obvious but just in case, here goes:
You say that one option is disabled as soon the other is enabled. To me that sounds like you're trying to capture AND decode segments at the same time. But what if you capture data into segmented memory first and THEN enable the decoder as you "browse thru" the captured segments?

Or perhaps that is what you're already doing in which case I applogize.


 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 1224
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #5 on: October 22, 2017, 02:56:47 am »
Here's a video that demonstrates a way to do serial decoding with a DS1054Z using full memory:


The difference between theory and practice is less in theory than
the difference between theory and practice in practice.
Expensive tools cannot compensate for lack of experience.
 

Offline mos6502

  • Frequent Contributor
  • **
  • Posts: 531
  • Country: aq
Re: DS1054Z record mode and serial decode
« Reply #6 on: April 27, 2018, 06:59:19 am »
I just ran into the same freaking problem with my DS1054Z. This is bullshit. First the stupid serial decode only works on a single screen. Then they obviously deliberately block what would be a way around that limitation. Also, saving CSV data only works on a single screen as well. Why the hell do I have all this memory if I can't do anything with it? Fuck you, Rigol. Luckily, I can still send this POS back for a refund.

Yeah, I tried DSRemote. It's slow and the decoding only produces garbage, it's even worse than the scope's internal decoding.
for(;;);
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8489
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #7 on: April 27, 2018, 07:19:15 am »
Yeah, I tried DSRemote.

But have you tried sigrok?

...or one of those $6 logic analyzers off eBay. If you'd asked anybody here they'd have told you about the limitations of DS1054Z serial decode for large amounts of data.

 

Offline mos6502

  • Frequent Contributor
  • **
  • Posts: 531
  • Country: aq
Re: DS1054Z record mode and serial decode
« Reply #8 on: April 27, 2018, 07:29:59 am »
I haven't tried sigrok. I've had one of those Saleae clone LAs for years. It's cool, it works. But the scope is supposed to be able to do that. I mean Rigol is actually selling those decode options. They're not cheap. Considering the Siglent SDS1104X-E is able to do the decoding properly, combined with other improvements like better multi-channel sample rate, I'm seriously considering dropping the extra 100 eurobucks and getting that one.
for(;;);
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 8108
  • Country: my
  • reassessing directives...
Re: DS1054Z record mode and serial decode
« Reply #9 on: April 27, 2018, 09:30:03 am »
I just ran into the same freaking problem with my DS1054Z. This is bullshit. First the stupid serial decode only works on a single screen. Then they obviously deliberately block what would be a way around that limitation. Also, saving CSV data only works on a single screen as well. Why the hell do I have all this memory if I can't do anything with it? Fuck you, Rigol. Luckily, I can still send this POS back for a refund.

Yeah, I tried DSRemote. It's slow and the decoding only produces garbage, it's even worse than the scope's internal decoding.
long memory 8bit analog has other uses, mainly its advertised feature as a 'dso'. You may return your rigol and get siglent for better decoding, then again for us its not a clear winner. Rigol excel in something other brand doesnt..
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 12705
  • Country: nz
  • NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: DS1054Z record mode and serial decode
« Reply #10 on: April 27, 2018, 10:38:47 am »
I just ran into the same freaking problem with my DS1054Z. This is bullshit. First the stupid serial decode only works on a single screen. Then they obviously deliberately block what would be a way around that limitation. Also, saving CSV data only works on a single screen as well. Why the hell do I have all this memory if I can't do anything with it? Fuck you, Rigol. Luckily, I can still send this POS back for a refund.

Yeah, I tried DSRemote. It's slow and the decoding only produces garbage, it's even worse than the scope's internal decoding.
long memory 8bit analog has other uses, mainly its advertised feature as a 'dso'. You may return your rigol and get siglent for better decoding, then again for us its not a clear winner. Rigol excel in something other brand doesnt..
I'll bite.

Pluses ?
Avid Rabid Hobbyist
 

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 1446
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #11 on: April 27, 2018, 11:14:32 am »
Take a look on this linkl

https://sigrok.org/wiki/Rigol_DS1000Z_series

Is that new? It does not say what version of sigrok supports it, and it did not seem to say that it supports DS1104Z.

I tried PulseView and connect with USB, and when I click scan, nothing happens (the button changes color, yay!).

I also seem to recall trying to install some driver. It was weird and then totally broke any communication I once had with my Rigol (using ultrasigma, etc).

My scope connects to my PC and I have a USB Instrument device from IVI foundation.

sigrok does not work.

edit: I just downloaded the nightly build and same same  :--

I also remember looking at the menu, it zadag, and that thing had me hose up my SDR too... :--
« Last Edit: April 27, 2018, 11:22:18 am by metrologist »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8489
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #12 on: April 27, 2018, 06:34:59 pm »
I've had one of those Saleae clone LAs for years. It's cool, it works.

SO why are you messing about with a DS1054Z?

But the scope is supposed to be able to do that.

Says who?

The DS1054Z decodes serial data perfectly, just not in the way you expected it to.

The limitations are no secret. There's a hundred posts and entire threads on here dedicated to them.

eg. http://www.eevblog.com/forum/testgear/ds1054z-rs232-decoding/

Considering the Siglent SDS1104X-E is able to do the decoding properly

Says who?
a) It might not do what you think it does, either.
b) Twisty knobs and buttons are simply the wrong paradigm for large amounts of serial decoding. eg. Setting up serial triggering or searching for data in along transmission is a couple of clicks on a PC but a total pain in the ass on an oscilloscope. That's ANY oscilloscope, not just a DS1054Z.

(unless you get a super expensive one with mouse/keyboard, obviously).

Bottom line: Use your scope's serial decoding feature to make sure the signals are present/correct. Use your $6 USB logic analyzer to process/decode large amounts of data.
« Last Edit: April 27, 2018, 07:41:41 pm by Fungus »
 
The following users thanked this post: theirishscion, Jacon

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8489
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #13 on: April 27, 2018, 06:42:56 pm »
I also seem to recall trying to install some driver. It was weird and then totally broke any communication I once had with my Rigol (using ultrasigma, etc).

My heart and prayers are with you. I hope you recover soon.

The DS1054Z doesn't actually need any drivers to work. Putting that 500Mb "driver" in the DS1054Z's box isn't doing anybody any favors.

(It's a bit like the CDs you get with printers, the only sane way to use them is as a frisbee or to scare birds away).
« Last Edit: April 27, 2018, 07:45:08 pm by Fungus »
 

Offline Gandalf_Sr

  • Frequent Contributor
  • **
  • Posts: 570
  • Country: us
Re: DS1054Z record mode and serial decode
« Reply #14 on: April 27, 2018, 07:36:08 pm »
I own a DS1054Z but I've also worked with Agilent and Tektronix scopes that had serial decoding. The simple fact is that scopes are not the right tool to decode yards of serial message data -  Being able to trigger off specific packets is useful but really the role of the scope is to see that the waveform/voltages are correct and that every part of the SPI/I2C/UART system is responding correctly.
Once you've got the basic ability to send packets between devices done then you're much better off using a logic probe like the Saleae Logic 8/16.
 
The following users thanked this post: Fungus, theirishscion, Jacon

Offline Bob Sava

  • Regular Contributor
  • *
  • Posts: 84
  • Country: us
Re: DS1054Z record mode and serial decode
« Reply #15 on: April 27, 2018, 11:42:50 pm »
With digital scopes it does not have to be that way.  If oscilloscope can can share it's deep memory with sigrok (via LXI tools), which SDS1104X-E is supposed to do in near future, you get capable 4 lead LA. 
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 2215
  • Country: it
Re: DS1054Z record mode and serial decode
« Reply #16 on: April 28, 2018, 12:02:48 am »
while that is true, i also want to be able to zoom in and out because i want to see the bigger picture.. or a detail. I also want to not become crazy if i want to time-correlate the serial bus with other analog waveforms.
of course doing this with the rigol 1000 is impossible  highly impractical.
move the horizontal and some part of the signal goes off screen? decoded data become garbage
zoom out so it can't decode from pixels anymore? decoded data become garbage
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8489
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #17 on: April 28, 2018, 12:29:03 am »
With digital scopes it does not have to be that way.  If oscilloscope can can share it's deep memory with sigrok (via LXI tools), which SDS1104X-E is supposed to do in near future, you get capable 4 lead LA.

You can download all DS1054Z memory today.

Use ":WAVeform:DATA?" command. I assume Sigrok can do that.
 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 1224
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #18 on: April 28, 2018, 02:19:59 am »
Yep, the video posted above shows that it's possible. It's just annoying that downloading the internal memory takes so much time.
The difference between theory and practice is less in theory than
the difference between theory and practice in practice.
Expensive tools cannot compensate for lack of experience.
 
The following users thanked this post: JPortici

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 1446
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #19 on: April 28, 2018, 03:26:15 am »
I also seem to recall trying to install some driver. It was weird and then totally broke any communication I once had with my Rigol (using ultrasigma, etc).

My heart and prayers are with you. I hope you recover soon.

The DS1054Z doesn't actually need any drivers to work. Putting that 500Mb "driver" in the DS1054Z's box isn't doing anybody any favors.

(It's a bit like the CDs you get with printers, the only sane way to use them is as a frisbee or to scare birds away).

Well, thanks for helping me figure it all out. Time will heal all wounds. Nice new shiny things help too...
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 2215
  • Country: it
Re: DS1054Z record mode and serial decode
« Reply #20 on: April 28, 2018, 03:34:41 am »
Yep, the video posted above shows that it's possible. It's just annoying that downloading the internal memory takes so much time.

yep. better use a scope that does it on its own
 

Offline Gandalf_Sr

  • Frequent Contributor
  • **
  • Posts: 570
  • Country: us
Re: DS1054Z record mode and serial decode
« Reply #21 on: April 28, 2018, 04:25:58 am »
You could get this one on eBay https://www.ebay.com/itm/Keysight-Used-DSOZ634A-Oscilloscope-63-GHz-160-GSa-S-4-channel-Agilent-/192507981698?hash=item2cd25eaf82

You'll save nearly $226,000 but I'm not sure it does decoding though.
 

Offline Bob Sava

  • Regular Contributor
  • *
  • Posts: 84
  • Country: us
Re: DS1054Z record mode and serial decode
« Reply #22 on: April 28, 2018, 04:27:55 am »
Siglent SD1104X-E is 10x faster than SD1054z (using LXI tools) so I'm counting that it'll also be 10x faster with Sigrok when supported.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8489
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #23 on: April 28, 2018, 04:40:56 am »
Yep, the video posted above shows that it's possible. It's just annoying that downloading the internal memory takes so much time.

You need better software. It's possible to grab data about twice as fast as in that video:

http://www.eevblog.com/forum/testgear/ethernet-transfer-speed-for-rigol-ds2072a-and-ds4014/msg1087482/#msg1087482


« Last Edit: April 28, 2018, 04:44:02 am by Fungus »
 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 1224
  • Country: 00
Re: DS1054Z record mode and serial decode
« Reply #24 on: April 28, 2018, 04:54:05 am »
Yep, the video posted above shows that it's possible. It's just annoying that downloading the internal memory takes so much time.

You need better software. It's possible to grab data about twice as fast as in that video:

http://www.eevblog.com/forum/testgear/ethernet-transfer-speed-for-rigol-ds2072a-and-ds4014/msg1087482/#msg1087482

That test doesn't follow Rigols' programming guide. There could be a reason why the Rigol engineers created
those long timeouts between packets. Possible buffer underruns could be one of them.
Nice if you are trying to trace down some strange artefact and after hours or days you discover it's the software that reads
the data from your scope.
The difference between theory and practice is less in theory than
the difference between theory and practice in practice.
Expensive tools cannot compensate for lack of experience.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf