SDS800X HD Bug Reports + Firmware

[tautech]

Where have you seen a V3 Uboot ?

I bought my SDS804X HD (upgraded to SDS824X) around March or April last year (2024)...  Can't recall the exact date but I do remember it was early after release when some stock was hard to come by.   I never paid attention if the U-Boot upgraded in the process, but it's currently showing as version 3.8.12

Thanks.
New unit just updated, as we always do before dispatch, also shows Uboot as 3.8.12.

With my demo unit I'll roll back the FW a couple of versions then reapply the latest update to see if it remains as 2.8.12 or gets to V3......been running on beta FW versions for a bit so need to check these things.... 

[2N3055]

Hey, everyone. This is my first post here.

I downloaded the firmware update to an 8GB USB stick freshly formatted as FAT32. I added the .ads file to it and ejected it from my PC. Then I plugged it in to my SDS814X, went to upgrade, chose the file, and pressed the reboot button. The oscilloscope rebooted, but it is now stuck on the Siglent splash screen. It's been that way for 30 minutes. I haven't tried to touch anything, but I'm unsure what to do. This isn't the first time I've updated the firmware on it, so I'm not sure what's going on. This unit is "unlocked" to an SDS824X, if that matters.

The firmware file I downloaded is named "SDS800X-HD_1.1.6.2.ADS"

Any ideas? Thanks!

These aren't intended to update with a reboot.

The FW update process is user selected and takes a minute or 2 to upload the FW update file and a progress bar is displayed.
Then the DSO must reboot for the new FW to be installed and the reboot either happens after an auto timeout or the user can select to reboot immediately.

The same thing happened in my case. Factory SDS824X, SW 1.1.3.8, UBoot-OS 2.8.12, FPGA 2024-04-11, HW rev. 00.
My oscilloscope was allowed to warm up for a sufficient amount of time.

Then the new firmware was loaded from the web interface and I let the instrument reboot itself when the reboot timer expired.
After reboot it stuck on the Siglent logo for an indefinite time.
No message like "please wait in progress" is displayed (considering that in the background Siglent is installing something or doing some cleanup.)

After waiting for 15 minutes, I decided to turn off the oscilloscope then a simple power on brought the instrument back to life.
I followed the advice to perform a self-calibration routine after the firmware update and the operation took about 5 minutes.
The new firmware seems to have been installed successfully, FPGA is now ver. 2024-11-26, but UBoot-OS remains at ver. 2.8.12.
Why are there two versions of UBoot (2.x.x versus 3.x.x), is this related to the hardware revision? I saw SDS800X with the same HW rev. but with different UBoot OS versions.

As does my SDS814X HD, an early build (#0316) from early 2024.
Just updated to 6.2 and previously running on the beta version 5.0.

Where have you seen a V3 Uboot ?

Regarding UBoot ver. 3, User @Furna post here:

https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/msg5586747/#msg5586747


On the other hand, I think I found the reason why the oscilloscope got stuck on the Siglent logo after installing the new firmware.
I always performed the firmware upgrade from the web interface. No problems up to version 1.1.6.2.
I couldn't call this a bug, because it doesn't prevent it from working, but...

- With the LAN interface connected, DHCP and NTP configured to auto, I turned on the oscilloscope and it remains stuck on the startup logo.
- I suspected something related to NTP, so I configured NTP to manual. Still stuck on Siglent logo.
- If I call the Utility - Reboot menu, everything works normally, but if I select normal shutdown, after next power on the oscilloscope remains stuck in the startup screen.
- Without LAN connection to wired network, the oscilloscope starts normally.

Maybe it's a specific malfunction only for me, it remains to be seen if it can be reproduced by other users who have installed firmware 1.1.6.2

Try the factory reset  by pressing quickly and persistently a Math button during boot.

I tried several variations of DHCP, NTP and DNS settings, and it works all the time.
But I did reset the scope after FW update.. That assures that all the new variables and all old ones are reset to good starting point For new FW.

[Furna]

Try the factory reset  by pressing quickly and persistently a Math button during boot.

I tried several variations of DHCP, NTP and DNS settings, and it works all the time.
But I did reset the scope after FW update.. That assures that all the new variables and all old ones are reset to good starting point For new FW.

After the fw upgrade (I had no issue using a USB stick), I just did a facotry reset with the Math button during boot ... the info about the SMB server were NOT deleted, user and password included.
This is obviously NOT what you want if you perform a factory reset before selling the device.

[KungFuJosh]

After the fw upgrade (I had no issue using a USB stick), I just did a facotry reset with the Math button during boot ... the info about the SMB server were NOT deleted, user and password included.
This is obviously NOT what you want if you perform a factory reset before selling the device.

I believe that's to recall factory settings for scope functions, not the same as resetting a phone. If you want to be lazy about deleting those settings, the easiest thing might be to request a recovery package.

[Furna]

After the fw upgrade (I had no issue using a USB stick), I just did a facotry reset with the Math button during boot ... the info about the SMB server were NOT deleted, user and password included.
This is obviously NOT what you want if you perform a factory reset before selling the device.

I believe that's to recall factory settings for scope functions, not the same as resetting a phone. If you want to be lazy about deleting those settings, the easiest thing might be to request a recovery package.

Sorry KungFuJosh, I won't buy this 
When the scope exited the factory my SMB server, username and password were not there.
Also the "normal" Facotry Recall of the scope parameter is under Utility/Save-Recall/Recall/Type:Facotry Default

If I go through a key during boot I mean a true "Factory Reset" ...

[2N3055]

After the fw upgrade (I had no issue using a USB stick), I just did a facotry reset with the Math button during boot ... the info about the SMB server were NOT deleted, user and password included.
This is obviously NOT what you want if you perform a factory reset before selling the device.

I believe that's to recall factory settings for scope functions, not the same as resetting a phone. If you want to be lazy about deleting those settings, the easiest thing might be to request a recovery package.

Sorry KungFuJosh, I won't buy this 
When the scope exited the factory my SMB server, username and password were not there.
Also the "normal" Facotry Recall of the scope parameter is under Utility/Save-Recall/Recall/Type:Facotry Default

If I go through a key during boot I mean a true "Factory Reset" ...

No need for "word police".
English is not native language to either of us.

Pressing Math will restore scope internal state variables, like Factory default does.

It does not delete SMB settings, true. Which is useful to me because those do not influence work of scope, and I don't need to retype them again if I reset scope. It  maybe should delete it but I actually like it this way. I often do reset scope when testing something, and that saves me typing.

To delete SMB Client settings perform Secure erase. That is procedure that is there for that scenario.

[KungFuJosh]

Exactly. Resetting Factory scope settings is not the same thing as Secure Erase or Declassification. Nor would I want it to be the same, for the same reason.

It's often a good idea to apply factory settings after a firmware update. I wouldn't want to have to reset my network settings or other general settings (like screen brightness, screen saver, etc.) every time I did a firmware update.

[TomKatt]

Exactly. Resetting Factory scope settings is not the same thing as Secure Erase or Declassification. Nor would I want it to be the same, for the same reason.

It's often a good idea to apply factory settings after a firmware update. I wouldn't want to have to reset my network settings or other general settings (like screen brightness, screen saver, etc.) every time I did a firmware update.

I usually do this myself after upgrading firmware - though I wonder if I am doing it correctly...  When powering on the scope and repeatedly pressing the MATH button, the scope beeps 2-3 times (I think normal boot is a single beep).  But there is never any confirmation on the screen that factory defaults have been applied.  Is this the correct procedure?

[KungFuJosh]

Exactly. Resetting Factory scope settings is not the same thing as Secure Erase or Declassification. Nor would I want it to be the same, for the same reason.

It's often a good idea to apply factory settings after a firmware update. I wouldn't want to have to reset my network settings or other general settings (like screen brightness, screen saver, etc.) every time I did a firmware update.

I usually do this myself after upgrading firmware - though I wonder if I am doing it correctly...  When powering on the scope and repeatedly pressing the MATH button, the scope beeps 2-3 times (I think normal boot is a single beep).  But there is never any confirmation on the screen that factory defaults have been applied.  Is this the correct procedure?

No, that trick is for when you need a hardware reset if the scope is unresponsive on bootup.

Go to save/recall and apply factory settings from there.

[watchmaker]

I upgraded to the new FW.  Now when the wakes up from screensaver, the mouse moves its pointer, but has no function.  Will only reset with a reboot.  Tried this several times. 

Is this user error or a FW issue?

THANKS!!

[2N3055]

I upgraded to the new FW.  Now when the wakes up from screensaver, the mouse moves its pointer, but has no function.  Will only reset with a reboot.  Tried this several times. 

Is this user error or a FW issue?

THANKS!!

I quickly tried with a wired Logitech mouse, screensaver set at 1min and it worked fine...

please provide more details. For mouse and scope (from scope System Info). I understand it is latest FW, but to see HW and other detail.

[watchmaker]

I upgraded to the new FW.  Now when the wakes up from screensaver, the mouse moves its pointer, but has no function.  Will only reset with a reboot.  Tried this several times. 

Is this user error or a FW issue?

THANKS!!

I quickly tried with a wired Logitech mouse, screensaver set at 1min and it worked fine...

please provide more details. For mouse and scope (from scope System Info). I understand it is latest FW, but to see HW and other detail.

Thanks.  I am using a wireless mouse to reduce cable clutter.  But I just realized it may also be a source of noise.  Attached is a screenshot of my system info.

[xiaomian]

I upgraded my oscilloscope firmware to version 1.1.6.2, but soon encountered an issue.
When the storage depth exceeds 10Mpts (e.g., 20Mpts), the interface continuously displays an "Acquiring" progress bar, which is extremely frustrating.
To verify this behavior, I downgraded the firmware back to version 1.1.3.8 and tested under the same settings—this time, the problem did not occur.
Have you experienced similar issues? I’ve attached screenshots comparing both versions for reference.

[Joe1vm]

I upgraded my oscilloscope firmware to version 1.1.6.2, but soon encountered an issue.
When the storage depth exceeds 10Mpts (e.g., 20Mpts), the interface continuously displays an "Acquiring" progress bar, which is extremely frustrating.
To verify this behavior, I downgraded the firmware back to version 1.1.3.8 and tested under the same settings—this time, the problem did not occur.
Have you experienced similar issues? I’ve attached screenshots comparing both versions for reference.

Hi. I can confirm similar behaviour. The longer aquisition time for lower frequencies and higher memory depth was the first thing I noticed after the update.

[TomKatt]

Hi. I can confirm similar behaviour. The longer aquisition time for lower frequencies and higher memory depth was the first thing I noticed after the update.

Now that this topic came up, I thought slow speeds acted funny after upgrade - I typically don't use very long timescales, but the same 'Acquiring' progress bar displayed on mine as well.  I chalked this up to user error after I changed to smaller timescale and the progress bar cleared.  It is definitely different operation. 

I will have to check when I get back to my scope.

[2N3055]

I upgraded my oscilloscope firmware to version 1.1.6.2, but soon encountered an issue.
When the storage depth exceeds 10Mpts (e.g., 20Mpts), the interface continuously displays an "Acquiring" progress bar, which is extremely frustrating.
To verify this behavior, I downgraded the firmware back to version 1.1.3.8 and tested under the same settings—this time, the problem did not occur.
Have you experienced similar issues? I’ve attached screenshots comparing both versions for reference.

Hi. I can confirm similar behaviour. The longer aquisition time for lower frequencies and higher memory depth was the first thing I noticed after the update.

There are no longer acquisition times.
Triggering frequency at 1ms/div is exactly the same in v1.1.36 and the v1.1.6.2.
Also triggering frequency is same at 10Mpts and 20 Mpts.

Yes, I measured, just now...

It is only that Acquiring progress bar that used to activate at 100ms/div,  now (unnecessarily) activates at 1ms/div.
I will report it as visual annoyance to be fixed.
But it bears no impact to function or speed.

[xiaomian]

Thank you to confirm that issue.
I have observed another similar issue in firmware version 1.1.6.2: When setting channel 1's bandwidth limit to 20MHz with 20Mpts memory depth, a pop-up warning appears stating that enabling bandwidth limitation above 10Mpts memory depth may slow processing speed. (This prompt was absent in the earlier 1.1.3.8 version)
In my opinion, this warning appears unreasonable because I believe the 20MHz bandwidth limitation occurs at the analog front-end section, which should be unrelated to the processing speed of the digital section following ADC conversion.
Practical testing also revealed that enabling bandwidth limitation under deep memory conditions showed no observable impact on waveform refresh rates. (This conclusion is based on my simple observation of the waveform update speed on the screen with 20Mpts depth)

[eTobey]

I couldn't call this a bug, because it doesn't prevent it from working, but...

- With the LAN interface connected, DHCP and NTP configured to auto, I turned on the oscilloscope and it remains stuck on the startup logo.

I dont get it... you dont call this a bug, but it remains stuck on the startup logo in this case?

[2N3055]

I couldn't call this a bug, because it doesn't prevent it from working, but...

- With the LAN interface connected, DHCP and NTP configured to auto, I turned on the oscilloscope and it remains stuck on the startup logo.

I dont get it... you dont call this a bug, but it remains stuck on the startup logo in this case?

Because he is smart. Until someone else can reproduce, it is some fluke that happened to him.
He never got back if this still happens to him after scope settings reset.
Nobody else managed to reproduce this so far.

[eTobey]

After the fw upgrade (I had no issue using a USB stick), I just did a facotry reset with the Math button during boot ... the info about the SMB server were NOT deleted, user and password included.
This is obviously NOT what you want if you perform a factory reset before selling the device.

There is so much, that is not wanted. Get used to it. 

When setting channel 1's bandwidth limit to 20MHz with 20Mpts memory depth, a pop-up warning appears stating that enabling bandwidth limitation above 10Mpts memory depth may slow processing speed. (This prompt was absent in the earlier 1.1.3.8 version)

Too bad, that there is no indication, that decoding does give you wrong data from a certain timescale onwards. Also there is no indication whatsoever, when you decrease the sample rate, that the trigger rate decreases as well.

[TomKatt]

Too bad, that there is no indication, that decoding does give you wrong data from a certain timescale onwards. Also there is no indication whatsoever, when you decrease the sample rate, that the trigger rate decreases as well.

At some point the equipment can only be made so foolproof and the expectation is that the user understands the device and it's limitations.  This is what manuals are for and why I always doublecheck my setup when I encounter unusual measurements.

[2N3055]

Too bad, that there is no indication, that decoding does give you wrong data from a certain timescale onwards. Also there is no indication whatsoever, when you decrease the sample rate, that the trigger rate decreases as well.

There is an indication, it is called timebase and sample rate.......

[tautech]

Too bad, that there is no indication, that decoding does give you wrong data from a certain timescale onwards. Also there is no indication whatsoever, when you decrease the sample rate, that the trigger rate decreases as well.

There is an indication, it is called timebase and sample rate.......

[eTobey]

At some point the equipment can only be made so foolproof and the expectation is that the user understands the device and it's limitations.  This is what manuals are for and why I always doublecheck my setup when I encounter unusual measurements.

Yes, doublechecking is important. It is however in some cases tedious. For example going through decoded packets. And one might get fooled into thinking he could use the search feature for this, but that has a flaw when using "continues run mode": It does skip events... 

The thing is: the data looks good, and the sampled data and sample rate is sufficient for proper decoding, but the scope just fails to decode properly at some scale (at stop mode).

Using another device might be better here.

[TomKatt]

Using another device might be better here.

I occasionally use the built in decoders when I'm too lazy to set up my logic analyzer.  The built in is OK for quick checks, but for anything that requires long acquisitions, complex triggers or scrolling and searching the LA software (Sigrok fork) is infinitely easier to use and more powerful.

I'm glad the decoders are available in the scope, but they're not really replacements for a proper LA.  Even a cheap LA clone running Sigrok or Pulseview is probably better in many circumstances.