mib-std2-pq-zr-toolbox: startup_x.boot colors missing
Not sure if the extract-startup_x.boot.py script works correctly. Here’s an example of an extracted file img_31.png from startup_4.boot container (took the file from MST2_EU_VW_ZR_P0252T software package). I believe that this should have some blue elements over here.

At the same time script extract-mcf.py worked as expected and image files look correct.
About this issue
- Original URL
- State: closed
- Created 3 years ago
- Comments: 60 (30 by maintainers)
Commits related to this issue
- Updated with improvements from https://github.com/olli991/mib-std2-pq-zr-toolbox/issues/2 — committed to olli991/mib-std2-pq-zr-toolbox by deleted user 2 years ago
Step-by-step tutorial (might have some bugs). TechniSat custom boot screen - VAG MIB2.pdf
Added inverse transformation. Just update the plugin and reload gimp. There is one change. In order not to conflict with classic “png” files, the file extension must be changed to “.mib”. This applies to both opening and export. To open the original file in color - copy it with the .mib extension and just open it in gimp File \ open menu mib2image.zip (outdated version. see the last one in the next posts)
Updated: Fixed version
The images inside the boot-file all have 2 bytes per pixel. That means: 1 byte for transparency level, 1 byte for color. So there’s max 256 transparency levels, and max 256 shades of gray. Working with a palette didn’t suffice.
So I made the images half the size, so we have more bytes per pixel. This is the result when interpreting the RGBA data as a PNG:
However, if I use the Blue channel as the Green channel, (so the Blue is duplicated), it looks like this:
I’m no yet happy with this outcome, but it’s something. At least there’s some color!
I don’t mean to be rude, but be patient (or feel free to fix it yourself 😉 ) It’s done when it’s done.
small update.
I’m ashamed. Didn’t check the changes properly. the result is bug. Fiexed&Improved (added option to extract label) mib2image.zip (outdated version. see the last one in the next posts)
Yes, original pictures will be opened as gray. It was difficult for me to match the label with the axial background, so I redrawn it entirely. I think it’s not a problem to make a plugin for opening original pictures in color. If you look at the code, the conversion is quite simple. I might try it next week (just getting started with python and GIMP). I can suggest my template from Photoshop. MAIN_TPL.zip
Hi! I got a little time, I got the following result
Here is plugin for GIMP to export colored image for MIB2STD boot animation. mib2image.zip (outdated version. see the last one in the next posts)
To install just extract to plug-ins folder of GIMP (in my case it’s c:\Program Files\GIMP 2\lib\gimp\2.0\plug-ins) and restart GIMP
How to use: create image with resolution 800x480
In menu - select File\Export As… (Shift+Ctrl+E) Set name\folder to export In “Select File Type (By Extension)” (bottom left) choose “MIB2STD BOOT Image”
And press Export system will prompt you to set color limit. it was found that a file larger than 1 MB is not installed in my system, so I have to limit the color Default - 128, I used 64
Result: 2 files 1 - Main picture (800x480) with an empty window for animated label (for size reduce) - “selected file name”.png 2 - Label picture (480x100) - “selected file name”_lbl.png
Not sure if this is going to be any help to anyone involved, but here’s a bunch of texture finders that help easily switch between different data interpretation modes: https://forum.xentax.com/viewtopic.php?t=15540
Hello!
As far as I was able to figure out, the time here is set in tenths of a second, so about 0.05 is doubtfulsorry, 0.05 s is Ok. Usually, an animation contains 57 pictures, so it’s possible to fit in your 47, but on my device I encountered a problem that an animation larger than 1,000,000 bytes simply is not recorded in the device. Therefore, I limited myself to 17 screens. And according to the procedure - you draw slides, convert and assemble the required animation file. Then you open the resulting animation in a hex editor and manually edit COMAND_LIST (an array of blocks of 8 32-bit words. One block - one command). I described my experience here https://www.drive2.ru/l/628474250440751976/ p.s. Before recording to the device, I recommend unpacking the resulting file and viewing the resulting animation.csv for errors.@jorrecob The solution is simple enough. It is necessary to correct the file names 0-9 to two digits 00-09. It’s strange that the original script worked. I was getting the same error with it. Why is this happening, I did not understand. Both import and export use the same code for generating filenames (‘img_’ + str(j).zfill(2) + ‘.png’). But in one case, one digit, and in the other two. (and in my version this code has not changed). I’m not familiar with python (this is my first meeting), maybe there is some peculiarity here… maybe someone will come in handy: by changing the file types in import / export scripts, you can immediately get * .mib files and pack them. I don’t think I need to move this to the main branch, but I made such a solution for myself mib2image.zip
@jtomtos and everybody involved: Before i explain the issue i have, i want to thank you already very much for all the effort you put into this topic…so THANK YOU.
Editing and using the new plugin works as intented. But i’m having issues compressing the file again.
Your alternative compress script does not work for me:
When using the original compress file by Jille, a new startup_8.boot is created:

But, when i try to extract the file again, to check if everything is fine, i get an error:
The result is a startup_8 file in extract, but with only 1 png: img_0.png
Now, just be sure i tried to compress and extract 2 folders with the original scripts. Results:
compressing an extracted startup_x folder and then re-extracting gives the error mentioned above (Error -5…incomplete or truncated stream) AND no different outcome if i use the original folder or a folder with edited png’s (.mib gimp method).
extracting an original dumped .boot file works properly and all png files are extracted in the correct folder.
So the error lies somewhere in compressing any folder…i think.
@jtomtos @jilleb ,could you have a look (or somebody else of course), as what i’m missing or doing wrong.
zip file with original dump, .boot and file with new png’s: Startanim.zip
Many thanks.
Just a heads up: it’s much easier and faster to drag&drop files and directories into Terminal window than to type in the path to file/directory.
@stupots: that’s right. My tutorial is using some Linux/macOS specific paths. Will try to make Windows version soon as well. But can’t tell when.
@mattcabb thanks, I’ll submit the script to repository as soon as I figure out how (I’m new one here 😃). combine extract and convet in more complexity task for me now. I will look for tools to do it
@jorrecob screen resolution is not a tenet. it is only important that the width is even. I’ll turn off this check. Label extraction will work only on base resolution 800x480.
Paths in the tutorial are just examples. They highly depend on the locations of the toolbox and .boot files. You need to point to the locations which are correctly for your PC.
exactly… …custom/startanim/startup_… and code unit for r-line “06” byte 17 or 18 can’t remember
… use “toolbox” there is an option for it 😃
… and here the sample, (I like it 55%) I will find time, it will adjust… anyway it is possible, thanks to man, now we are counting on creative invention 😃)) startup_6.zip
https://www.youtube.com/watch?v=kZWAs5eZuc4
There somethig to think about. It’s looks like this picture is combilation of two channels. Alpha channel is picture itself in grayscale, and GrayScale channel is a color mask. By default it has 128,128,128 color - gray (from white to black), and here at least two other colors - blue (blue-motion) and red (beatsaudio) difined as interlacing dark and light pixels. For Example blue color grayscale channel looks like this:
how it works i can’t understand. Sugget, that even -dark, odd -light is red, odd -light, even -dark is blue but i have not found some kind of logic in this… only reproducing known colors
Thank you Jille for working on this! Much appreciated!