sn4k3
commented
3 years ago That will invalidate the test. Cure a layer for 6s + 0.5s is not the same as cure in once 6.5s layer. That would only be possible if LCD could render the image almost instant. Right now with obligatory lifts, firmware restrictions, light offs etc, the layer will cooldown on that time and them recure for other time, if you do that you will find the test will not match to a real print where layer cure for a continous time. Even if you have a gcode printer = no lifts, no lightoffs, change layer image takes time and layer will cooldown. Also its bad ideia to keep LED ON almost all the time, lifts, lightoff provide a cooldown window to the LCD which extend your hardware lifetime.
This test brings the commodity of testing multiple exposures in one print = less repetition and less post-processing. Even if it takes the same time of print them individual, in fact you win a lot of time by not process - clean and repeat steps
Also look at: https://github.com/sn4k3/UVtools/discussions/categories/ideas
! Before continue, please look/search on closed topics if your case was already been discussed, if yes and related, comment there instead.
Is your feature request related to a problem? Please describe
The exposure time finder with multiple exposure times is a fantastic idea, but currently the way UVTools generates slices means that print time for 6 exposure times is effectively the same as running six separate prints. It seems like a slight change to logic could make that process much, much faster.
Describe the solution you'd like
For the moment, assume I'm generating an exposure tester with everything set the same except normal exposure time. My exposure table has entries for a 6, 6.5, and 7 second normal exposure variants (all other things equal). Let's call the 6 second exposure "A", the 6.5 "B", and the 7.0 "C." The current logic generates as follows:
Layer 1: A, 6.0 seconds Layer 2: B, 6.5 seconds Layer 3: C, 7.0 seconds
So (again ignoring everything else) the printer will take 6+6.5+7=19.5 seconds to print one layer of each test. Each slice in the output file includes pixels for only one of those tests, but of course the pixels never overlap.
Would it therefore be possible to generate the slices as follows:
Layer 1: A + B + C, 6.0 seconds Layer 2: B + C, 0.5 seconds Layer 3: C, 0.5 seconds
The total print time in that case would only be 7 seconds per layer, while preserving the correct exposure times.
Describe alternatives you've considered
I'm still fairly new to UVtools (which, btw, thanks for such a useful and versatile suite of tools!), so I have a hunch that merging layers and retiming them in this way might be possible via scripting. At present, though, it's outside of my ability.
Additional context
I'm far from an expert in the mechanics of resin printing, so if there's some technical reason why this is impossible, please forgive my ignorance. :)