search

gaugo87 / PIXEstL

A program for creating color lithophanies and pixel images
MIT License
55 stars 10 forks source link

Single AMS with 7 Color print wrong colors #8

Open miesbe72 opened 6 months ago

miesbe72 commented 6 months ago

I generated the STL files with following skript: java -jar PIXEstL.jar -p filament-palette-0.10mm.json -H 150 -c 4 -l 4 -i IMG-20181225-WA0007.jpg The preview image was displayed with the correct colors. Ich folgte den Anweisungen in der instruction.txt Datei. The wrong colors were used in the finished printout. There were also 36 layers created where hardly any light was shining through.

ibgregr commented 4 months ago

IMHO the preview is just a best guess in a perfect world. If the filament calibrations are not right then the printed results will not match the preview. I did fine the the default "filament-palette-0.10mm.json" was not accurate in regards to my Bambu filaments and the printed results were way off. I ended up doing a total recalibration with my filaments and my results are greatly improved now. The printed lithophane still does not match the preview 100% but it is alot closer. Bottom line is, the closer your calibrations are the close the result will look like the preview. I have found that with my 2 AMS units, I enable the filament colors that closely match the palette of the picture being printed as that seems to leave less mixing of colors to get the desired result. There have been times where I disable Magenta because I just don't need it for the picture and that gives me a slot for a more specific color to be used. So, this definitely is not a science. I typically end up having to tweak the image I want to print a bit (brighten colors, adjust brightness and contrast, etc) so that I can get results that I want.

gaugo87 commented 4 months ago

I generated the STL files with following skript: java -jar PIXEstL.jar -p filament-palette-0.10mm.json -H 150 -c 4 -l 4 -i IMG-20181225-WA0007.jpg The preview image was displayed with the correct colors. Ich folgte den Anweisungen in der instruction.txt Datei. The wrong colors were used in the finished printout. There were also 36 layers created where hardly any light was shining through.

Hi, you're right, this mode isn't perfect.

As I explain in my video, the issue stems from the transmission distance, which isn't managed by the program. So, the more layers there are, the less reliable the calculation becomes, especially for the deeper layers. And since this mode uses 9 layers, there's some inaccuracy.

I've tried developing another approach to achieve reliable results, but my initial attempts/tests haven't been conclusive.

gaugo87 commented 3 weeks ago

Hello @miesbe72,

For some time now, Bambu Studio has added an option to add the possibility to force the sequence order of filaments at a given layer (in addition to the first layer). I find this to be a "game changer" for the "7 color with 1 AMS" option because it is no longer necessary for PIXEstL to add a white layer for color changes to occur. This white layer had a significant negative impact on the final result.

Therefore, I have revised the procedure and released a new version that significantly improves the final result.

miesbe72 commented 3 weeks ago

Hello @gaugo87,

thank you for your reply. I will try it next weekend :D