Siglent SDS1104X-E and SDS1204X-E Mixed Signal Oscilloscopes

[tautech]

As a note/addition I installed the FW and then OS from a freshly formatted and only 512 MB USB stick.

So it's confirmed that we can do it in any order we like.

Not entirely sure as some of the improvements you've listed in the Review thread I've not seen yet. More study needed. Got to whip out and get another shipment so it'll be later before I look at this all in detail.
Like Arny, I'll be back. 

Please don't alienate the folks here

Verifying the listed fixes/improvements is a completely different matter and I will certainly take a closer look at them soon - even though I generally have very little time right now.

Whenever the system status displays 7.1.6.1.25R1 as the software version, then this proves that both (OS and FW) updates have worked okay. No further confirmation needed.

Yep. it's all good.

Not sure what you guys are doing to not have success and HW version is immaterial to success. I've done 2 units the oldest being a beta unit with SN# 0012 and another built ~4 weeks back. Downgrade to 20R1 and back to 25R1 with no probs. It doesn't even seen to matter which order to do it either, OS first or FW first.
It does help to have a USB stick with LED indication so you know the processes are happening. Just be patient and let it happen.
I've been using 2 USB sticks both FAT formatted, one 8G and another 512M.

Other than to try downloading again and/or another USB stick I have nothing else to suggest, sorry.

[Performa01]

For me this confirms that the USB-stick and its formatting is fine and I cannot think of any reason why the automatic OS update wouldn't work.

I've dropped Siglent a note - hopefully they can come up with some more suggestions, but I guess they would have to be able to reproduce the issue, which might be a problem.

[tautech]

For me this confirms that the USB-stick and its formatting is fine and I cannot think of any reason why the automatic OS update wouldn't work.

One possibility is inserting the OS upgrade without the DSO OFF. Instruction clearly states it should be and then after inserted power ON. It'll reboot a time or two but when finished stop at the normal display.

[Performa01]

For me this confirms that the USB-stick and its formatting is fine and I cannot think of any reason why the automatic OS update wouldn't work.

One possibility is inserting the OS upgrade without the DSO OFF. Instruction clearly states it should be and then after inserted power ON. It'll reboot a time or two but when finished stop at the normal display.

As I understand it, member "nez" has already tried that...

[tv84]

I only have Windows available at the moment, so I can't try formatting through linux. Sorry!  It is an 8GB drive formatted as Fat32.

I've tried inserting the usb stick while the scope was powered off, and also while it was on and then rebooting it.  Same result either way for the OS not installing.  The firmware installed fine through the normal Utility menu on the scope.

OK. Linux formatting would be a "remote" possibility.

My last suggestion is:

Format the stick with FAT. Ensure that the scope recognizes the stick and save some files (captures) from the scope to the stick.

Then, power off the scope, remove the stick and copy the 4 files to the stick as the manual says (don't remove the captures).

Ensure that you are doing clean ejections from Win.

Insert in the scope and power on.

Also, check with some hash prog that the files you have in the stick have the correct hash (crc32 is enough).

[cardre]

The 4 files match in name, timestamp, and file size as those shown in the instructions pdf file.

Hmm, not sure they do exactly match the sizes. Maybe try downloading a fresh copy of the OS file, possibly with a different browser, if you have one available (Chrome, Firefox) just to be sure. Then try again.

I updated mine successfully and did the firmware first then OS second. Apparently though the order is not important.

[Performa01]

Okay, I just got a reply from Siglent and they would like to know the result when a different USB stick is used.

As unlikely as it seems to me, there aren't many possibilities left, so you should definitely try a different USB stick, preferably a different brand as well. maybe you could also provide the information what you actually use.

I would also strongly consider the advice given by cadre above. Corrupted files would certainly prevent the update process.

EDIT: oh yes, and tv84's suggestion should be considered as well ...

[Performa01]

I updated mine successfully and did the firmware first then OS second. Apparently though the order is not important.

Please let us know whether the new custom probes fit your needs, once you've tried it out.

[Old Printer]

My next serious purchase is an entry level 4 ch scope. I have been closely watching offerings from Rigol, Siglent and GW Instek. I have to say, I am impressed with the communication and response from Siglent. New release bugs and questions about how quickly they would be addressed were probably the most negative comments about these two new scopes. So far I am impressed with Siglent's effort to get this done quickly and with an open mind as to requested changes.

[firstcolle]

- Kingston data traveler 100 G3 32Gb: no way to get it works
- Transcend SD card 8Gb with USB adapter: work at first try

[radiolistener]

New release bugs and questions about how quickly they would be addressed were probably the most negative comments about these two new scopes. So far I am impressed with Siglent's effort to get this done quickly and with an open mind as to requested changes.

On 8 Febrary 2018 I sent the issue about signal distortions (which was added in the latest firmware for SDS1102X) to Siglent Support.
But there is still no fix for this annoying bug. And I still stay with old firmware where this bug is missing.

[nctnico]

New release bugs and questions about how quickly they would be addressed were probably the most negative comments about these two new scopes. So far I am impressed with Siglent's effort to get this done quickly and with an open mind as to requested changes.

On 8 Febrary 2018 I sent the issue about signal distortions (which was added in the latest firmware for SDS1102X) to Siglent Support.
But there is still no fix for this annoying bug. And I still stay with old firmware where this bug is missing.

People make a lot of noise and praise new firmware releases but they fail to realise they bought a flawed device to begin with, not all bugs are fixed and some bugs may never be fixed. 

[tv84]

Noob question: the siglent and root passwords of this scope (new OS) are known?

[tautech]

Noob question: the siglent and root passwords of this scope (new OS) are known?

One would guess they're the same as they've always been. I know not otherwise.

[Old Printer]

New release bugs and questions about how quickly they would be addressed were probably the most negative comments about these two new scopes. So far I am impressed with Siglent's effort to get this done quickly and with an open mind as to requested changes.

On 8 Febrary 2018 I sent the issue about signal distortions (which was added in the latest firmware for SDS1102X) to Siglent Support.
But there is still no fix for this annoying bug. And I still stay with old firmware where this bug is missing.

People make a lot of noise and praise new firmware releases but they fail to realise they bought a flawed device to begin with, not all bugs are fixed and some bugs may never be fixed. 

I understand what you are saying, but this may be the new "normal" in this class of scope. Releasing "rough" product seems to be a trend. Time will tell if someone can release a polished product with these specs and at this price point. If not, hobbyists may well be willing to be a part of the development process, and live with some unfixable  bugs that professional users would not stand for, all in the name of price.

[nez]

Okay, I just got a reply from Siglent and they would like to know the result when a different USB stick is used.

As unlikely as it seems to me, there aren't many possibilities left, so you should definitely try a different USB stick, preferably a different brand as well. maybe you could also provide the information what you actually use.

I would also strongly consider the advice given by cadre above. Corrupted files would certainly prevent the update process.

EDIT: oh yes, and tv84's suggestion should be considered as well ...

Ok, I made sure the files were fine and tried a different usb stick, a Kingston DataTraveler 8GB I just bought rather than the junky no-name stick I had available previously (a free handout, from a convention or something).

With the new usb stick, the OS updated successfully right away.

Thanks all for the various suggestions to troubleshoot!

[tv84]

With the new usb stick, the OS updated successfully right away.

Thanks all for the various suggestions to troubleshoot!

nez,

Please read https://www.eevblog.com/forum/testgear/siglent-sdm3045x-boot-hang/msg1565089/#msg1565089 and provide us with your 2 USB controllers info.

The one that didn't work and the one that worked.

[17_29bis]

Hello,

First thing that I noticed in the newest firmware is that AC trigger level indicator is still missing.
----------------------------------------------------------------------------------------------------------------
Added later:  I have no clue how the fix of IC2 bug passed QA test and was added into the new firmware but this is what we get now:



1) what is that suffix 'A' doing in the hex output  (before they used H)?
2) what is the point in using the prefix '0x' and the suffix (I assume Siglent meant 'H' but ended up with 'A' due to some strange reasons)? If you want to save space then either use the prefix or the suffix but not both because the screen is small and space is precious. I would say - use  none, since after I  have set the format to 'Hex' I don't need any reminders/hints embedded into the data  since I am shown what I asked for - the hexadecimal data;
3) it is a nice idea to have a dedicated window to show the missing part of the string although I would implement it differently. Anyway, I would be happy with it too but just look at the last hex number in the first line (19A) and the first hex number in the second line (1BA) - the guys lost 0x1A or 1AH (a matter of taste);
4)again, I think consistency is important. Look at the last hex number in the packet (it's 0x1F) displayed in the 'Long Data' window - it is there. Now look at the SDA and SCL lines, the end of the packet - 0x1F is there. Now look at the decoded data (the line at the bottom of the screen) - the last part is  '0AA 0' - it must be 0x1F and nothing else to be synchronized with other data.

I think I will downgrade the firmware for now.

[iMo]

I understand what you are saying, but this may be the new "normal" in this class of scope. Releasing "rough" product seems to be a trend.

Releasing a "rough product" is a trend today everywhere, indeed. It is the result of the new "Agile mindset" especially with the sw community, which considers a "waterfall" sw development project approach obsolete.
Moreover, the number of sw developers involved in this kind of products is rather small, my guess is the number of sw developers who actually mess with the above stuff at Rigol or Siglent is 1.53 in average.
Soon or later their willingness to provide fast fixes will decline, as the number of newly introduced bugs will be higher than the number of bugs fixed..
PS: I saw that with the largest IT internationals I worked with in past. The staffing ratio of their hw_developers:sw_developers:sales&marketing people with their strategic products was typically 4:1:20000 world-wide..

[Performa01]

First thing that I noticed in the newest firmware is that AC trigger level indicator is still missing.

Okay, that doesn’t surprise me as there was absolutely no evidence we would get that at all, let alone for this intermediate update. Yet I’ll ask for once why this request is persistently ignored.

Added later:  I have no clue how the fix of IC2 bug passed QA test and was added into the new firmware but this is what we get now:

First let me say thank you for trying out the new IIC decoder and documenting its issues so thoroughly. I’m really sad to see this mess.

1) what is that suffix 'A' doing in the hex output  (before they used H)?
2) what is the point in using the prefix '0x' and the suffix (I assume Siglent meant 'H' but ended up with 'A' due to some strange reasons)? If you want to save space then either use the prefix or the suffix but not both because the screen is small and space is precious. I would say - use  none, since after I  have set the format to 'Hex' I don't need any reminders/hints embedded into the data  since I am shown what I asked for - the hexadecimal data;

It is quite obvious that the decoder is messed up. The ‘0x’ doesn’t belong there and the ‘A’ should be ‘H’. I think a (non capital!) ‘h’ suffix might still be useful to clearly indicate the numeric format used when taking screenshots.

3) it is a nice idea to have a dedicated window to show the missing part of the string although I would implement it differently. Anyway, I would be happy with it too but just look at the last hex number in the first line (19A) and the first hex number in the second line (1BA) - the guys lost 0x1A or 1AH (a matter of taste);

Yes, the text doesn’t quite fit the window width.
How would you implement it?

4)again, I think consistency is important. Look at the last hex number in the packet (it's 0x1F) displayed in the 'Long Data' window - it is there. Now look at the SDA and SCL lines, the end of the packet - 0x1F is there. Now look at the decoded data (the line at the bottom of the screen) - the last part is  '0AA 0' - it must be 0x1F and nothing else to be synchronized with other data.
 

It is quite obvious that the packet is severely misaligned and truncated in the bottom decoding line. You get the first 11 bytes at a wrong position – and a red dot at the right end of the bracket, indicating that there is more data – how true.

What happens if you zoom out? Is the bottom decoder line still misaligned and/or truncated?

[rf-loop]

It is quite obvious that the decoder is messed up. The ‘0x’ doesn’t belong there and the ‘A’ should be ‘H’. I think a (non capital!) ‘h’ suffix might still be useful to clearly indicate the numeric format used when taking screenshots.

What a mess.

0x belongs there, (example here data 00) in first place as 0x00. It tell that it is hex. 0x is standard method for tell data is in hex format. Without any exeption.
Also it is enough in this case it show it with first byte. After then who do not understand next bytes are something other than hex. I can not imagine any. So after each data there do not need anything for tell every byte it is hex. User have selected hex and first 0x also show it if some other people look only image.

It can be as in image example: 0x00 01 02 03 04 05 06..... etc. It is totally stupid waste of tiny resources if add H or h or what ever after every data byte.  if 0xdd (dd=data)  in fisrt place do not make clear to everyone that displayed data is hex..... unprintable.

For Siglent: Just remove these stupidities (H and now accidentally this weird A) from displayed data string. Without any question. No one miss these waste of display area. First place 0x tell hex is in use. If one who use serial decoder do not understand 0x it is not scope manufacturer problem at all. Period.

[Performa01]

It can be as in image example: 0x00 01 02 03 04 05 06..... etc. It is totally stupid waste of tiny resources if add H or h or what ever after every data byte.  if 0xdd (dd=data)  in fisrt place do not make clear to everyone that displayed data is hex..... unprintable.

I fully agree.

I just didn't think that far when I posted my last reply. After the huge disappointment about this quite obviously totally untested piece of code I would just have been happy to get the previous appearance back.

Now I begin to understand - have you suggested that change? And the attempt to implement it produced the mess we have now?

[nctnico]

  The A stands for 'Acknowledge' of the I2C bus. There will probably an 'N' or '~A' for not-acknowledge .

[kerouanton]

I fully agree too.

0x at the beginning of the string to remind it's hex, and the rest using just 2 hex-digit pairs.


By the way, I'm fitting my hobby lab with the full range of Siglent equipment (PSU, Scope, SA, Meter, FuncGen), and noticed a few bugs/issues for example on the SA32xx. Is this forum a good way to let Siglent know about those issues in the hope they will profide fixes?

[TK]

Siglent could use GREEN for ACK, RED for NACK and use lower case 'h', 'd', 'a' at the end of the data to indicate HEX, DECIMAL, ASCII