Owon XDS3202E vs Siglent SDS1202X-E

[oleh]

oleh said a bit back that the Siglent was "hanging on to death" while doing a USB save.  I own an SDS1202X-E and there were some issues with the USB drive but, after I applied the first software update, there have been zero issues using USB.

I was trying to save longer waveforms in the CVS format. I didn't find how to decode Siglent's binary format, it had quite big, also binary, header. I think it hung with more than 700,000 points.

[oleh]

My old 150MHz analog scope was giving better picture and better trigger synchronization.

You have a much more powerful trigger suite in a DSO.........learn how to make the most of it. 

Thank you! I'm learning

[rf-loop]

I was trying to save longer waveforms in the CVS format. I didn't find how to decode Siglent's binary format, it had quite big, also binary, header. I think it hung with more than 700,000 points.

Header is from 0x0 to 0x2B83  This is header.
signal data starts from 0x2B84  to end of data.  This is normal waveform data where one byte is one sample.
If memory length is 700000 ksample this part length is 700000 bytes
0x2B83 is  11139 and then + 1 byte (0x0) header length is 11140 bytes. (fixed)

So, of course if you want 700k wfm data it hung bit over 700k
If you save 14 data points there is 14 data points in memory, exactly. And before these 14 data bytes there is 11140 bytes header.
If you know your settings it is not nessessary to know this header at all, Just use this data. I mean if you know V/div, offset, t/div, sample interval and so on, youi can just use these 8 bit data values.

[oleh]

I was trying to save longer waveforms in the CVS format. I didn't find how to decode Siglent's binary format, it had quite big, also binary, header. I think it hung with more than 700,000 points.

Header is from 0x0 to 0x2B83  This is header.
signal data starts from 0x2B84  to end of data.  This is normal waveform data where one byte is one sample.
If memory length is 700000 ksample this part length is 700000 bytes
0x2B83 is  11139 and then + 1 byte (0x0) header length is 11140 bytes. (fixed)

So, of course if you want 700k wfm data it hung bit over 700k
If you save 14 data points there is 14 data points in memory, exactly. And before these 14 data bytes there is 11140 bytes header.
If you know your settings it is not nessessary to know this header at all, Just use this data. I mean if you know V/div, offset, t/div, sample interval and so on, youi can just use these 8 bit data values.

Yes, I know. I found where data starts right away. But I'm too lazy to make descriptions for every file manually. Also, I live in a unix world and here file timestamps  are important. "ls -lrt" is done pretty often during a day.

[oleh]

I'm thinking about loading custom software to this scope. SoC is similar to one used in Beaglebone board but without 3D and PRUs. Process of loading software is not so complicated in Owon. I found that capacity of existing flash ROM is used for about 12% and that is almost 80% uncompressed graphic elements. The memory layout in Owon is much better than in Siglent. Owon has three banks of memory: two 256MB chips are attached to FPGA and SoC has its own 128MB memory. Shared memory in Siglent can be a bottleneck. I hope that Owon will leak FPGA constraints file somewhere, it would be great. They don't have enough developers anyway. With this hardware and good collection of open source software this scope will be able to blow any other scope out  of water.

[MrW0lf]

With this hardware and good collection of open source software this scope will be able to blow any other scope out  of water.

Cannot be fully excluded other scopes will jump out of the water themselves and ROFL on the beach if see those childish knobs on Owon. If gonna write new software maybe improved looks also?

[boggis the cat]

But I'm too lazy to make descriptions for every file manually.

I'm thinking about loading custom software to this scope.

That may be a bit ambitious.  A script to automate renaming / time stamping files would seem a much easier undertaking.

With this hardware and good collection of open source software this scope will be able to blow any other scope out  of water.

I have suggested that Rigol would do well to open up an 'end of life' model to allow people to try to work on it (and/or Siglent, or anyone else with a significant hobbyist user base).

If you believe that you can do this with this newer Owon then that would be very interesting -- however, I think this is unlikely to be achievable.  Best of luck if you do have a go at this.

[oleh]

With this hardware and good collection of open source software this scope will be able to blow any other scope out  of water.

Cannot be fully excluded other scopes will jump out of the water themselves and ROFL on the beach if see those childish knobs on Owon. If gonna write new software maybe improved looks also?

There are no P&G brand baby products colors on the front panel.

[boggis the cat]

There are no P&G brand baby products colors on the front panel.

Beauty -- or interface appropriateness -- is in the eye of the beholder.

I don't much care for the Owon design, myself, and consider the Siglent to be fine.  Not the best, but perfectly acceptable.

(The decision to make the Owon knobs that design, with the inset colour, seems ill-considered.  The rest of the UI seems accceptable.)

[nctnico]

With this hardware and good collection of open source software this scope will be able to blow any other scope out  of water.

People have been saying that for years but writing firmware for a DSO with measurements, decoding, etc is a massive project and without a decent CPU in the hardware lots needs to be done in FPGA which means development is even more slow. IOW: I don't see this happening soon.

[oleh]

With this hardware and good collection of open source software this scope will be able to blow any other scope out  of water.

People have been saying that for years but writing firmware for a DSO with measurements, decoding, etc is a massive project and without a decent CPU in the hardware lots needs to be done in FPGA which means development is even more slow. IOW: I don't see this happening soon.

I'm looking at the Comedi project. It is quite old, mature and sill alive. I hope existing fpga bitfile can do the job.

http://comedi.org/ and quite old article in the Linux Journal:
http://www.linuxjournal.com/article/7332?page=0,2

[oleh]

There are no P&G brand baby products colors on the front panel.

Beauty -- or interface appropriateness -- is in the eye of the beholder.

I don't much care for the Owon design, myself, and consider the Siglent to be fine.  Not the best, but perfectly acceptable.

(The decision to make the Owon knobs that design, with the inset colour, seems ill-considered.  The rest of the UI seems accceptable.)

Maybe I can buy better knobs on eBay...

[Diabolo]

Problèmes ?
-

Regards,
Diabolo

[oleh]

Problèmes ?
-

Regards,
Diabolo

Yes, problems. That was all about $1300 scope, here we were talking about almost the same but for $360.