SDS800X HD Bug Reports + Firmware

[mikes]

Leading zeros would add the benefit of that they can be used in scripts. I believe the same lame scheme is used with filenames of other data types...

Your scripting-fu is weak. Leading zeros have the detriment of being of somewhat arbitrary length. You suggested 6 digits. Do we know what filesystem they use? FAT32 has a limit of 65534 files/directory (or less if filenames are longer than 8.3), so no more than 5 would be needed. OTOH, if you can plug in a exFAT USB drive, the limit is over 2,000,000, so then you'd need 7. But ext4 can be formatted to allow over 10,000,000, so 8+ digits?

[Furna]

   

Can no one work with what they already have without the need to muck with it and impact on a whole product line ? 

Don't like internal/local memory filename structure ?
Connect a USB keyboard and properly name your filenames......or is that too hard ? 

As I wrote ... "If anything has to be suggested to Siglent"
I save files directly on a NAS and then move/rename the files via PC ... saving a keyboard and some real estate on the bench

[tautech]

   

Can no one work with what they already have without the need to muck with it and impact on a whole product line ? 

Don't like internal/local memory filename structure ?
Connect a USB keyboard and properly name your filenames......or is that too hard ? 

As I wrote ... "If anything has to be suggested to Siglent"
I save files directly on a NAS and then move/rename the files via PC ... saving a keyboard and some real estate on the bench

A wireless keyboard can be used on your lap and stored on a shelf just as I do at my workstation.
Sorry, I don't buy that one......

[Furna]

Leading zeros would add the benefit of that they can be used in scripts. I believe the same lame scheme is used with filenames of other data types...

Your scripting-fu is weak. Leading zeros have the detriment of being of somewhat arbitrary length. You suggested 6 digits. Do we know what filesystem they use? FAT32 has a limit of 65534 files/directory (or less if filenames are longer than 8.3), so no more than 5 would be needed. OTOH, if you can plug in a exFAT USB drive, the limit is over 2,000,000, so then you'd need 7. But ext4 can be formatted to allow over 10,000,000, so 8+ digits?

I agree on the scripting-fu skills ...
I do not agree on the remaining sentences ... local disk is by design a temporary disk.
Do you really want to save thousand of files on the local disk?? Using it as a long term storage? With no backup?
C'mon this entire discussion is BS to make eTbobey eyes happy ... there is no real topic here.
 

PS a wireless keyboard is a keyboard too much; a keyboard should stay where it belongs: wired to a PC.

[2N3055]

In a topic about bugs, there is more than hundred of posts about things that has nothing to do with instruments bugs.
People arguing sort orders and all kinds of other things off topic.
I cannot decide if it is annoying how some are desperate to create drama where there is none, or that we are so lucky there are so little real problems they have to be invented artificially.

There is topic about suggestions for improvements of instrument.
Scope sorts exactly the same as my Ubuntu WKS. So no bug there.
As for filename choice,  those are automatic names that scope gives because user didn't want to name them themselves.
If they can be made a bit more user friendly fine. But it is filenames, people. Really?

[mikes]

I do not agree on the remaining sentences ... local disk is by design a temporary disk.
Do you really want to save thousand of files on the local disk?? Using it as a long term storage? With no backup?

The scope has an SMB client.

[Furna]

I do not agree on the remaining sentences ... local disk is by design a temporary disk.
Do you really want to save thousand of files on the local disk?? Using it as a long term storage? With no backup?

The scope has an SMB client.

Exactly what I use.
Then I reorder/rename/move the files via PC where they belong.
A project folder or whatever your want; not thousand of files on a single directory.
Sometimes even the trash bin.
Anyway as someone already pointed out, this topic should be about BUGS and FIRMWARE.
The topic for genereal discussiuon is at https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/
Unfortunately, sometimes I cannot resist to comment. My bad.

[shapirus]

Leading zeros have the detriment of being of somewhat arbitrary length. You suggested 6 digits. Do we know what filesystem they use? FAT32 has a limit of 65534 files/directory (or less if filenames are longer than 8.3), so no more than 5 would be needed. OTOH, if you can plug in a exFAT USB drive, the limit is over 2,000,000, so then you'd need 7. But ext4 can be formatted to allow over 10,000,000, so 8+ digits?

I guess this is where common sense comes into play. 6 digits should be enough for everyone™. Nobody's ever gonna have that many screenshots (we're still discussing auto-generated screenshot file names, don't we?) on a scope.
And in the improbable (but theoretically possible) case when the counter hits 999999 the suffix can be extended with another "-000001" and so on. Sorting order won't be affected.

[2N3055]

Leading zeros have the detriment of being of somewhat arbitrary length. You suggested 6 digits. Do we know what filesystem they use? FAT32 has a limit of 65534 files/directory (or less if filenames are longer than 8.3), so no more than 5 would be needed. OTOH, if you can plug in a exFAT USB drive, the limit is over 2,000,000, so then you'd need 7. But ext4 can be formatted to allow over 10,000,000, so 8+ digits?

I guess this is where common sense comes into play. 6 digits should be enough for everyone™. Nobody's ever gonna have that many screenshots (we're still discussing auto-generated screenshot file names, don't we?) on a scope.
And in the improbable (but theoretically possible) case when the counter hits 999999 the suffix can be extended with another "-000001" and so on. Sorting order won't be affected.

To me common sense would be 001xxxx to 999xxxx.
That is 998 files in one directory.
If you are doing thousands upon thousands file saves in one directory then maybe there is a better way to do things...

But again, these are musical wishes and greetings type of radio program. 
Ask by all means , but not here, please. These are not bugs...

[eTobey]

Leading zeros have the detriment of being of somewhat arbitrary length.

Arbitrary length with leading zeros?? You mean "filenames without leading zeros" have arbitrary length?

I guess this is where common sense comes into play. ...

Common sense... something that would really help, but is not as easy to attain like knowledge.

Its not just the screenshots, but primarly other files that one would work with, or use in a script.


What i really find interresting:
I showed the proof of how "unconventional" the sorting of the filenames is with regards to letter case, but no one acknowledged or disprooved, and focused on less important things they can add their few cents. To me there is nothing to discuss about leading zeros. Its a pain in the ass to work with filenames in scripts that dont have leading zeros, and windows cant handle it either. Talking about filesystem limitations when we are are talking about 3-5 leading zeros, is beyond me...

But as usual, its always the messenger...

[mikes]

I guess this is where common sense comes into play. 6 digits should be enough for everyone™. Nobody's ever gonna have that many screenshots (we're still discussing auto-generated screenshot file names, don't we?) on a scope.
And in the improbable (but theoretically possible) case when the counter hits 999999 the suffix can be extended with another "-000001" and so on. Sorting order won't be affected.

I don't think anyone is going to have over 9 screenshots, and if they do, when the counter hits 9 the suffix can be extended with another "-1".etc.

[newbrain]

I showed the proof of how "unconventional" the sorting of the filenames is with regards to letter case

Ok let's see how "unconventional" this is"

Linux, bash on Ubuntu 24.04 (or anything else and FreeBSD default shell etc . etc.):

Code: [Select]

>ls
1  10  11  2  Aa  a  aa  b
>
Windows, powershell Get-ChildItem (i.e. 'dir'):

Code: [Select]

>Get-ChildItem

    Directory: C:\Users\teifezm\OneDrive - Ericsson\Documents\Sort

Mode                 LastWriteTime         Length Name
----                 -------------         ------ ----
la---          2025-02-11    12:58              0 1
la---          2025-02-11    12:58              0 10
la---          2025-02-11    12:58              0 11
la---          2025-02-11    12:58              0 2
la---          2025-02-11    12:58              0 a1
la---          2025-02-11    12:58              0 a10
la---          2025-02-11    12:58              0 A11
la---          2025-02-11    12:58              0 a2
Windows, cmd shell, 'dir'

Code: [Select]

#dir
 Volume in drive C is Windows
 Volume Serial Number is E6D0-4949

 Directory of C:\[none of your business]

2025-02-11  13:13    <DIR>          .
2025-02-11  13:13    <DIR>          ..
2025-02-11  12:58                 0 1
2025-02-11  12:58                 0 10
2025-02-11  12:58                 0 11
2025-02-11  12:58                 0 2
2025-02-11  12:58                 0 a1
2025-02-11  12:58                 0 a10
2025-02-11  12:58                 0 A11
2025-02-11  12:58                 0 a2
               8 File(s)              0 bytes

Windows File Manager, the only one to use lexicogrpahic order rather than purely alphabetical.

But as usual, its always the messenger...

As the boy who cried wolf used to say.

[eTobey]

I showed the proof of how "unconventional" the sorting of the filenames is with regards to letter case

Ok let's see how "unconventional" this is"
...

As i used no numbers, and more than just one letter in both letter cases, your post is kinda... uhm... what should is say... nonsense. 

Have you actually looked at the screenshot?
I even wrote:

Notice the "B" and its place (notice the case also).

Or are you just chiming in to bashing someone?

[newbrain]

Or are you just chiming in to bashing someone?

I really would like to be able to write with the rudeness that exudes from your every comment, unfortunately that's not my way of behaving.

Your screenshot only shows the UPPER CASE is sorted before lower case.
As 'ls' does:

Code: [Select]

1  10  11  2  Aa  B  a  aa  b  c
Don't bother to answer.

[awakephd]

I showed the proof of how "unconventional" the sorting of the filenames is with regards to letter case ...

As someone who has spent the last 40+ years playing with computers and micro-controllers of all sorts ... the sorting order you are showing looks perfectly conventional to me. This is how ASCII works. I guess that is why I was confused by your comments - every example you list, I think, "but that's perfectly sorted according to ASCII coding."

I do understand what you desire, and I even agree that it would be a nice feature, but having written a fair bit of UI software, I can attest that it is, relatively speaking, a lot of work. Very doable, of course, but typical of the adage that software development requires 1% correct calculations and 99% UI tweaks ... the latter especially because each person, in different cultures, has different expectations about what is conventional. Should the numbers be reported as 99,01 or as 99.01? Should the date be 2025-02-11, or 2-11-2025, or ... ? Yes, one can write the UI to provide choices and flexibility, but that means one is focusing 99% of attention on UI and only 1% on actual function.

When I made the choice between the Rigol 804 and the Siglent 804 last year, I did give thought to the UI. There seems to be widespread (though of course not universal) agreement that the Rigol UI is easier for newbies. As a complete newbie to DSOs, I worried that I would need that extra help. As it turns out, the Siglent UI, while it makes a few odd choices to my way of thinking, has generally been perfectly straightforward. Maybe that is just because of my long experience with ASCII, Linux, etc. (FWIW, I haven't ever encountered a UI that doesn't include some odd choices, or at least not the choices I would make! Which is where one might usefully look at the extensive hacking of the UI going on in the Rigol threads - at least two different individuals are reworking the UI to their own expectations and desires ... which, if I understand correctly, are not mutually compatible!)

Since I don't have a dog in this fight, I'm going to bow out of this conversation. I had hoped to make a helpful observation, but since we have different expectations for what sorting looks like, I can see that it was not helpful after all. Ah, well ...

[2N3055]

I showed the proof of how "unconventional" the sorting of the filenames is with regards to letter case ...

As someone who has spent the last 40+ years playing with computers and micro-controllers of all sorts ... the sorting order you are showing looks perfectly conventional to me. This is how ASCII works. I guess that is why I was confused by your comments - every example you list, I think, "but that's perfectly sorted according to ASCII coding."

I do understand what you desire, and I even agree that it would be a nice feature, but having written a fair bit of UI software, I can attest that it is, relatively speaking, a lot of work. Very doable, of course, but typical of the adage that software development requires 1% correct calculations and 99% UI tweaks ... the latter especially because each person, in different cultures, has different expectations about what is conventional. Should the numbers be reported as 99,01 or as 99.01? Should the date be 2025-02-11, or 2-11-2025, or ... ? Yes, one can write the UI to provide choices and flexibility, but that means one is focusing 99% of attention on UI and only 1% on actual function.

When I made the choice between the Rigol 804 and the Siglent 804 last year, I did give thought to the UI. There seems to be widespread (though of course not universal) agreement that the Rigol UI is easier for newbies. As a complete newbie to DSOs, I worried that I would need that extra help. As it turns out, the Siglent UI, while it makes a few odd choices to my way of thinking, has generally been perfectly straightforward. Maybe that is just because of my long experience with ASCII, Linux, etc. (FWIW, I haven't ever encountered a UI that doesn't include some odd choices, or at least not the choices I would make! Which is where one might usefully look at the extensive hacking of the UI going on in the Rigol threads - at least two different individuals are reworking the UI to their own expectations and desires ... which, if I understand correctly, are not mutually compatible!)

Since I don't have a dog in this fight, I'm going to bow out of this conversation. I had hoped to make a helpful observation, but since we have different expectations for what sorting looks like, I can see that it was not helpful after all. Ah, well ...

Thank you for your comment. I find it very helpful and well said.

[eTobey]

Or are you just chiming in to bashing someone?

I really would like to be able to write with the rudeness that exudes from your every comment, unfortunately that's not my way of behaving.

Your screenshot only shows the UPPER CASE is sorted before lower case.
As 'ls' does:

Code: [Select]

1  10  11  2  Aa  B  a  aa  b  c

You left the crucial item out, so it looked much different. Since this is what certain indivuduals do here too, i looked like you were in on that too.

My screenshot shows where the "B" is sorted, and that is sufficient (should be) to understand how this sorting works.

Since people mostly use Windows, which sorts as shown in the screenshot, i would consider this as "conventional".

[Furna]

eTobey you are right and we are wrong.
Happy now?
The topic for genereal discussiuon is at https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/

[eTobey]

eTobey you are right and we are wrong.
Happy now?

I am not happy at all, i got a scope, that differentiates between 2 ABC`s (among all the other issues). 

And i tested the sorting order with my older ubuntu, and it does like windows, and how my understanding of a proper sorting order would be. And its the opposite to newbrains sorting order...

[Furna]

I am not happy at all, i got a scope, that differentiates between 2 ABC`s (among all the other issues). 

And i tested the sorting order with my older ubuntu, and it does like windows, and how my understanding of a proper sorting order would be. And its the opposite to newbrains sorting order...

The topic for genereal discussiuon is at https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/
I see you are also aware of the "Wanted features" https://www.eevblog.com/forum/testgear/sds800x-hd-bugswanted-features/

[eTobey]

File letter case issue

I have again investigated in, what others call a "non issue", or a "user error".
I have created 2 files with the same letters, but different case. My confusion in the quoted post above, stems from the fact, that my windows cant handle different letter cases: You can copy a upper-case filename, that has the upper-case letters, but it actually copies the contents of the file with the lower-case.

Further more, after more investigation, i found, that you can delete the lower-case filename on the scope itself, but in windows, it still shows them both. After the restart of the scope, there is only the one file left as it shoult be.

(Connection was made with a WIFI bridge and SMB server)

Moral of the story: Do not use upper-case.

[Furna]

Moral of the story: Do not use upper-case.

The topic for genereal discussiuon is at https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/

[eTobey]

I just got to work (or at least i am trying to) with the scope again. I could complain, for the feature to only to be able to detect 1000 events, but instead i am gonna report something.

Search feature navigation issue (call it what ever you like, i call it a bug):
In continuous navigation mode, the search events are only iterated through at a rate of about 1.2 per second, no matter how small the interval is set to. Also multiple search events are omitted in between. With "singe step", i am able to achieve an iteration rate of about 5-10x of that of the continuous, by hand.

Another example of how a seemingly great feature will give you missleading information, if you wouldnt notice, that it has actually omitted events. 

EDIT:
LOOOL would you belive it? I found another bug, that - at the same time - is like a workaround for the bug above: 
When in single step mode, and you press and hold on the navigation buttons in the menu, the iteration works faster than continuous mode and even without omitting events.
The bug i consider here, is that the intention is only to move the menu up, but an iteration starts immediately (Even in single step mode). It is not possible to move the menu on this button without iterating the events.
Thankfully this bug can be worked around, by not moving the menu on those buttons, but this is something one has to know, because it may just lead to confusion, as the position changes unintentionally.

[Furna]

I just got to work (or at least i am trying to) with the scope again. I could complain, for the feature to only to be able to detect 1000 events, but instead i am gonna report something.

Search feature navigation issue (call it what ever you like, i call it a bug):
In continuous navigation mode, the search events are only iterated through at a rate of about 1.2 per second, no matter how small the interval is set to. Also multiple search events are omitted in between. With "singe step", i am able to achieve an iteration rate of about 5-10x of that of the continuous, by hand.

Another example of how a seemingly great feature will give you missleading information, if you wouldnt notice, that it has actually omitted events. 

EDIT:
LOOOL would you belive it? I found another bug, that - at the same time - is like a workaround for the bug above: 
When in single step mode, and you press and hold on the navigation buttons in the menu, the iteration works faster than continuous mode and even without omitting events.
The bug i consider here, is that the intention is only to move the menu up, but an iteration starts immediately (Even in single step mode). It is not possible to move the menu on this button without iterating the events.
Thankfully this bug can be worked around, by not moving the menu on those buttons, but this is something one has to know, because it may just lead to confusion, as the position changes unintentionally.

The topic for genereal discussiuon is at https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/

[mikes]

I don't know if this is a bug, or just an oversight.

The scope supports SMB server, but only SMB client is documented in the user manual. For the client, both username and password can be set. But for the server, there seems to be no way to change the username from the default "ShareUser". I'd certainly prefer if it could be changed to a username of my own choice.