At last, Siglent's SDS5054X touchscreen DSO...

[Martin72]

You mean SDS2000X+ without %$@*, I guess..
Indeed it would be interesting - I got the demoboard from siglent here (with spi signal) if this could help.

[nctnico]

You'll need SPI at 20+ MHz with about 200ms of data to do a real check. No doubt it will work with the demo board; that is carefully tuned so that it always works. A digital pattern generator (like I'm using for my oscilloscope tests / reviews) is a much more useful tool to find out what the real limitations of a DSO are.

[kcbrown]

You'll need SPI at 20+ MHz with about 200ms of data to do a real check. No doubt it will work with the demo board; that is carefully tuned so that it always works. A digital pattern generator (like I'm using for my oscilloscope tests / reviews) is a much more useful tool to find out what the real limitations of a DSO are.

I performed this experiment on my SDS-2104X+.  I set up my Raspberry Pi to do SPI and used spidev-test to send data over it at 20 MHz.  The SPI driver in the Pi has a default length limit of 4K bytes, so I used a file with 4K of data as the data source and ran the command in a loop.  I set up the scope with a timebase of 20 milliseconds per division (thus getting me 200 milliseconds of time coverage), and set up 200M points for the capture buffer.  I moved the trigger point to the beginning of the screen so as to maximize the amount of post-trigger signal captured.  I used the digital channels for the capture, with the lines connected directly to the Pi's GPIO pins.

This got me 9 bursts of 4K worth of data per burst (the loop I ran it in was in the shell).

It turns out that the decoding is limited to 15,000 bytes.  It wouldn't decode anything beyond that.  So it stopped decoding most of the way into the 4th burst.  But it did work, though the interface for examining it is slow and clunky (you can see it trying to align the decoded data with the signals, and it sometimes misaligns the results).

I was able to get it to work properly up to 30 MHz.  I got incorrectly decoded values above that.  This may well be due to environment more than the limits of the logic pod, since the lines to the pod are unshielded, and we're now in the territory where probing technique matters.

[Martin72]

It turns out that the decoding is limited to 15,000 bytes.

Yepp, you can find these limits in the datasheet for the sds2k+, for all decoder types(Page12).
For the sds5k, the values aren´t avaible in it´s sheet or manual.

[bson]

I there even a limit for the SDS5000X?  It seems to decode the whole record, at least up to 12.5M points (which is all I've tried).

[kcbrown]

I there even a limit for the SDS5000X?  It seems to decode the whole record, at least up to 12.5M points (which is all I've tried).

Does it decode anything that's not shown on the screen?  A way to test that is to set up SPI decoding with a chip select line included.  If the line doesn't go active within the visible portion of the capture, it will fail to decode the data if it's limited to what's visible (note that I'm talking about when you're not using zoom mode, but where the visible portion of the capture is a subset of the total capture).

My 2104X+ will do a decoding pass on the entirety of a 200M points capture (the maximum size of the capture buffer), but the decoded item count (e.g., bytes in SPI) is limited to 15K.  I'd think the limits of the 5000X would be higher.  No idea.

[bson]

I've only used it for parallel bus decodes so far.  Here's an example (the four analog channels are CS#, RD#, WR#, and C/D# to an SSD1963 display controller).  The zoomed in portion is barely discernible on the trace at the top.  I don't know if it's always decoded regardless of where the zoom is panned to, but it behaves as it is.  Maybe it does a higher level survey to find decode key points and then only does more detailed decodes when you look close enough.  I dunno.  But for all practical purposes it shows all 12.5M points decoded in the trace window, at least for bus traffic.

[tautech]

I've only used it for parallel bus decodes so far.  Here's an example (the four analog channels are CS#, RD#, WR#, and C/D# to an SSD1963 display controller).  The zoomed in portion is barely discernible on the trace at the top.  I don't know if it's always decoded regardless of where the zoom is panned to, but it behaves as it is.  Maybe it does a higher level survey to find decode key points and then only does more detailed decodes when you look close enough.  I dunno.  But for all practical purposes it shows all 12.5M points decoded in the trace window, at least for bus traffic.

D = Digital not Decode. 

In Decode menu turn List to ON and scroll to end for the decoded byte/packet count.

[bson]

The bus B1 is decoded.

[kcbrown]

The bus B1 is decoded.

B1 is composed of the digital channels, right?  So that would in fact be 12.5M points as you say.

But that's the number of samples (per channel, of course).  The question is how many decoded values it's limited to.  You can't determine that from the number of points in the sample buffer.

So Tautech is right: enable the list and scroll to its bottom.  That'll tell you how many values it decoded.

Note that on my SDS2000X+, if I'm in zoom mode when I attempt to scroll to the bottom, it'll annoyingly shift me back to the portion of the decoded section that the zoom window is showing.  I regard this as a usability bug.  I should be able to scroll anywhere in the decoded list and view the values and, on top of that, I should be able to double click on a decoded value and have the zoom window shift to that portion of the buffer.

It's not the end of the world.  I can turn off zoom mode if I want to scroll freely through the decode list, and you may need to do that in order to scroll to the bottom of it.

[oewean]

New firmware for SDS5000X, 0.9.7R2
https://www.siglenteu.com/download/12562/

1. Acquire: supported Fixed Sample Rate and Fixed Memory Depth modes

2. Connectivity:
a) Supported LXI
b) Supported Network Storage

3. Measure: improved the AIM limit from 1,000 to up to 25,000 (AIM limit: the
upper limit of horizontal parameters measure statistics in one frame )

4. Power Analysis: supported MOSFET SOA (Safe Operating Area)

5. Channel:
a) Two custom probe ratio options supported
b) Supported CP030 current probe (with LPA10 adapter)

6. UX: Added support for mouse wheel when using a mouse

7. Optimized UI

8. Remote Control: Supported to read sequence segments by WF command

9. Optimized the Bode Plot

10. Fixed several bugs
a) Incorrect zoom trace in Roll mode (Stop)
b) Bode Plot draw error (missing draw)
c) Bode Plot zig-zag phase
d) [Measure] Gating settings cannot be typed in

[FlPr]

Hello.
SDS 5034X.
When asked :syst:board? gives 4СХ_500М_LIC-V2
And the bandwidth.txt file is not empty.
But Time scale settings are limited to 1 ns/div
Owners of 5054 and 5104 confirm that you have Time scale 500 ps/div and 200 ps/div?
I don’t understand why he identifies himself like that and why the bandwidth.txt file is not empty

[tautech]

Hello.
SDS 5034X.

But Time scale settings are limited to 1 ns/div
Owners of 5054 and 5104 confirm that you have Time scale 500 ps/div and 200 ps/div?

SDS5054X with 1 GHz BW installed = 200ps/div....previously 500ps/div.

[bson]

Can someone else who knows confirm that installing the 4/27 release on a SDS5034X with the 1G option won't roll that back since it was never intended for this hardware?  All other options I'm sure will carry forward, but since there's no undoing it I just want to make certain.  I'd rather keep the improved bandwidth (even if slightly less than 1G) than pick up the various improvements.  (Although explicit support for the LeCroy CP030 is a very attractive addition.)

[tautech]

New firmware for SDS5000X models.

Version： V0.9.7R5
https://int.siglent.com/upload_file/zip/firmware/Oscilloscope/SDS5000X_V0.9.7R5_EN.zip
60MB

Release notes
Optimized frequency response of 350 MHz model
Fixed several bugs:
Screenshot Save Info Box does not close automatically
Random +/-400 ps skew between channels after power cycle
No display from Web Server if the scope boots with a USB device connected

[Martin72]

Screenshot Save Info Box does not close automatically

Ah, sounds good... 

Since the last firmware, the 2000Xplus got this box (and this bug) too, but not for long anymore I guess.

[tautech]

These are on special promotion until years end.

It entails a free BW upgrade and all the advanced decoding options for free:
SDS-5000X-FlexRay
SDS-5000X-1553B
SDS-5000X-CANFD
SDS-5000X-I2S
SDS-5000X-SENT
SDS-5000X-Manch

https://int.siglent.com/info/detail-62.html
https://www.siglenteu.com/news-article/free-oscilloscopes-bandwidth-upgrade-and-free-option-bundles-with-sds5000x-sds2000x-hd-purchase/

[Abvelv]

This promotion expired so I'm looking for alternative ways to upgrade my recently purchased SDS5034X. The scripts I find here don't seem to work. I did buy mine from an EU retailer, I saw some chatter about differences between western/european markets.

I searched the forums here but can't seem to find anything regarding keygens in the past 6 months or so. Perhaps recent firmware releases block keygen unlocks?

[tautech]

This promotion expired so I'm looking for alternative ways to upgrade my recently purchased SDS5034X. The scripts I find here don't seem to work. I did buy mine from an EU retailer, I saw some chatter about differences between western/european markets.

I searched the forums here but can't seem to find anything regarding keygens in the past 6 months or so. Perhaps recent firmware releases block keygen unlocks?

Yes AFAIK the algorithm for license creation has changed for some products and also be aware the limit for improving the 350 MHz model is to 500 MHz.
Early in this thread is the reason why. 

[0xdeadbeef]

New firmware V0.9.8R1
https://www.siglenteu.com/service-and-support/firmware-software/digital-oscilloscopes/

6/1/2023 0.9.8R1
1. Increased the data width of saved Average and ERES data from to 16-bit
2. Decode: added ARINC429
3. Math: added RBW display for FFT
4. Supported 12-hour time with AM and PM
5. Remote control:
   a) Supported the USB-GPIB adapter
   b) Added command supporting to query average count Acquisition = Average
   c) Added command supporting to enable/disable axis labels
6. Fixed several bugs
   a) Incorrect prompt when saving .csv with “Save all channels U-disk as specified file name
   b) Channel ON / OFF labels are inconsistent
   c) Freezing issue caused by command CURSor:XDELta?

[tautech]

New firmware for SDS5000X models

Version： V0.9.8R2
61.8MB
https://int.siglent.com/upload_file/zip/firmware/Oscilloscope/SDS5000X_V0.9.8R2_EN.zip

Release note
Fixed an issue in production process.

[JPortici]

Will we ever get the digital filters?

[Martin72]

Did the scope have the fixed mem/samplerate function ?
Edit:
It has (V0.9.7.R2).
Afaik this is necessary for the filters.

[markus_jlrb]

Dear EEVBlogger,

have someone a script/procedure to dump the FW flash (all partitions mtd0..X)
of the SDS5kX models in the same way I have done it for my SSA3kX few years ago
via telnet login and dd.

Thanks in advance for any hints.

Markus

[Martin72]

Batronix offers the 1Ghz model for 5405/6431€ (excl./incl. VAT):

https://www.batronix.com/versand/oszilloskope/Siglent-SDS5104X.html

Stock overstock sale, only while supplies last!

Only while supplies last....Hm-hm..
The last time they did that was to get rid of the Rigol HDO4000 because they were replaced by DHO.
And now they want to get rid of the 5104X...
Or is this just a promotion that doesn't mean anything?