FNIRSI-1013D "100MHz" tablet oscilloscope

[Felipe Lacerda]

Firmware
https://www.fnirsi.com/pages/support-and-downloads?srsltid=AfmBOoqJ__o6ILvaTPLgDZegLURHQDYwUCge2Rf8drCNHL_DWOFXsB5Q#

There is no release date, it may just be the stock firmware.
Let's wait for the experts. 

[citek]

The downloaded archive contains a folder with the date 21.09.2024, and inside the firmware 1013D.bin is dated 14.06.2024 and there is also an instruction.

We can safely assume that the firmware is updated.

[pcprogrammer]

I have to check it against the original firmware versions I have, but the name of the binary file in the archive is not what I expected it to be. Not sure if the older versions of the firmware will see this file as an update.

As far as the open source firmware, that can still be used. To clarify, the open source firmware does not use the FLASH memory in the scope, and that is why it is possible with the new boot loader to select between the two firmware's.

If one wants to try the "new" firmware from FNIRSI, it is best to use my firmware backup tool first to make a full copy of the FLASH, so that one can go back if needed. In the supplied document they state to connect to the scope via USB, open the folder, delete all the files and load the binary file in the folder.

There is, at least with the older versions I tested with, no need to delete the files on the SD card. What the original firmware does is see if there is a file with the correct name and if so clear the start sector of the firmware in the FLASH (at address 0x27000) and then reboots. The boot loader reads the binary file and writes it to the FLASH, while showing a progress bar.

[Atlan]

file rename to FSI-1013.bin, own risk

It does not display small signals as well as before.
In the long mode, they kept forgetting to multiply the values ​​by 2x

I RECOMMEND not to update

[pcprogrammer]

In the long mode, they kept forgetting to multiply the values ​​by 2x

Well that shows that they don't read this thread. 

[pcprogrammer]

The new firmware from the FNIRSI download site has a lot of binary differences with what I have in my collection, but to see the actual changes it would have to be de compiled with Ghidra which is a lot of work, so not going to do that.

I do wonder if they solved the issue with the different displays that are used in different releases of the scope. At some point they provided two separate binaries to cope with this issue, but now it is only one. 

So like Atlan wrote, rename the file and try it if you want to see for yourself what has been changed. There are always options to get back to a working scope. Best to make a backup first of course.

[Atlan]

I tried to go back through the fwb 1013.bin backup, but the oscilloscope starts updating, but it freezes at 100%

[pcprogrammer]

I tried to go back through the fwb 1013.bin backup, but the oscilloscope starts updating, but it freezes at 100%

Yeah, that part was never correctly finished. Has to do with some CRC not being correct. It is better to use the program I wrote for email hidden. https://www.eevblog.com/forum/testgear/fnirsi-1013d-100mhz-tablet-oscilloscope/msg5749727/#msg5749727

You can take your own full backup and extract the section from 0x27000 up to the end.

The source code for the restore program is in my github repository: https://github.com/pecostm32/FNIRSI-1013D-1014D-Hack/tree/main/Test%20code/fnirsi_1013d_firmware_restore

[Russell2024UK]

Hi Atlan  which Firmware do you recommend as today for our 1013D ? Thanks again 

[Atlan]

* fnirsi_1013d.bin v0.026b

[Russell2024UK]

Atlan Perfect .. will load it in Thank you again

[Atlan]

So that no one is surprised that it is moved up compared to the first versions.

[Cadstar70]

Hello! I am likely missing something really obvious, but after several days of reading this thread and experimenting with my newly acquired scope I am simply unable to get the new firmware loaded. I am a 35+ years linux admin and feel dumb for asking but I find a whole lot of this confusing for some reason and I'm stuck.

I can make a backup of my stock firmware just fine. Using dd is second nature to me so I think I understand what is going on here. I can dd the backup program over then reboot and I get the PECOS splash and it backs up all 3 files.  When I reboot all is good and I have the backups.

When I dd the firmware followed by the standard display config - I reboot and nothing has changed. I have tried sending just the firmware and not the display config. I have removed it and reinstalled multiple times. I have wiped the card completely and repartitioned several times. Each attempt has the same result - nothing, I'm still on the stock firmware and yet the backup always works.

What the heck am I doing wrong?? I don't need to build from source right? The how_to_load_scope.txt says I just need the  dd command.

Just downloaded the loader for windows and tried that - shockingly the same result! Loader says it loaded and I need to reboot but when I do I'm still on stock firmware. In the boot mode tab I selected PECO firmware and menu enabled - loader reports all set but I reboot and same thing.  Now when I select "compare to device" in loader it tells me they match. When I boot I don't have the version number in the upper right and I don't have the new menu options so I'm positive I am still on stock firmware. I think the firmware is there it is just not booting it.

I've gone over all the instructions endlessly but there is a whole lot here after 108 pages so I must be missing something.
I'm going nuts here I have no idea why this doesn't work.   

[Atlan]

What firmware versions did you try to copy?

[Cadstar70]

I have been using this link where it says "The firmware is at this location:"

https://github.com/pecostm32/FNIRSI_1013D_Firmware/tree/main/fnirsi_1013d_scope/dist/Debug/GNU_ARM-Linux

[Cadstar70]

OK, never mind.


It seems I was falling prey to downloading wrong I think. I was pulling files right from that link directly from the webpage through web browser and wget. That hasn't given me a problem in many years now. Finally remembered I should be cloning the repo from git and starting there. So I did that, navigated ton the .bin and I just now sent it over via dd as I have a hundred times, but this time just greeted with the PECOs boot screen!!

This always happens to me. Spend hours and hours determined to read everything and try everything, then finally ask for help knowing I'm doing something dumb. As soon as I do I find the answer - usually before anyone chimes in to help though.

Thanks Atlan, I really appreciate you being quick on the draw today but I finally got it.

Now to see if my cheap bench power supply is stable enough to actually calibrate this thing. I'm sure to have more questions coming, but looks like I finally have it at this point.

My version shows as V0.006 though - is that right?? I was expecting to see  v0.026b

  ....update: no this definately isn't right - I'm on the PECOs but not the right version.

[Cadstar70]

I'm getting closer, managed to get to V0.025c

Now I am new around here, but wow it wasn't easy to finally find this version. Maybe I just don't understand your organizational structure.
Still trying to figure out where the V0.026c is ....


   .....OK, I got it finally - wow it was a long road - I think the large number of zipped up firmware versions kept me from exploring those directories because they looked like incremental updates but I took a chance and yeah I'm on the V0.02c now! I jump into projects like this all the time and it doesn't generally take me this long to get up to speed. Now I can finally see if I can calibrate with my $35 power supply.

[pcprogrammer]

I'm getting closer, managed to get to V0.025c

Now I am new around here, but wow it wasn't easy to finally find this version. Maybe I just don't understand your organizational structure.
Still trying to figure out where the V0.026c is ....

You can find version V0.026b here: https://www.eevblog.com/forum/testgear/fnirsi-1013d-100mhz-tablet-oscilloscope/msg5780529/#msg5780529

It would be better if Atlan did what I did. Put a link to his github repository in his signature and upload the binaries there too, so it would be easier for new comers to find them.

[pcprogrammer]

It seems I was falling prey to downloading wrong I think. I was pulling files right from that link directly from the webpage through web browser and wget. That hasn't given me a problem in many years now.

It might be that it treated the file as ASCII text instead of binary and messed it up that way.

[Cadstar70]

Yeah I think so but I can't remember the last time that happened to me, oh well. I got there in the end, was just trying to avoid posting a dumb question.   

[Cadstar70]

Thanks, yeah I found it a bit hard to locate myself, but then again I tried to claw my way through 108 pages so I wasn't posting questions that were already answered. But there is a lot in there where you guys are reverse engineering way beyond my level and I think I got overwhelmed and just skipped by it. Figures it was ONE page back too

[Cadstar70]

OK, well that didn't last long. I set the toggle off for menu thinking that would just boot me without having to choose but no. Now I'm in FEL mode with no way out - heh.

I pulled the card and re-flashed the firmware again on another machine but that didn't fix the issue. Can I remove the package first and then reinstall or is that menu setting going to be persistent?

[Atlan]

Atlan has over 100 browser windows open, and currently 15 folders with files open. I don't have time to upload files to github, and usually the working versions are uploaded there. I created a new folder with the current firmware, the link is below in the signature.
Basically, don't try new versions outside the LAST FIRMWARE directory. It might not end well That's also the reason why the link to git hub is not there.

[Cadstar70]

Yeah I was trying hard not step on toes! I know how much hard work must have gone into all of this, please know I appreciate this like crazy! I picked up this scope second hand for $60 only because I already found out about your work and knew how much better it could be.

I got the FEL issue resolved by just formatting the sdcard in a pi and starting fresh.

[Atlan]

It was enough to edit the bytes on the SD card that select the boot mode