You are not logged in.
Pages: 1
I am curious when plate info get updated.
I have a few cases where the image of the model looks like a black shadow and never refreshes even if resliced.
Also I notice that volume of material used seems unaffected by changes in the profile and reslicing as I would expect. (say solid to hollowed)
And perhaps a simple oversight but if I change the actual profile used on a plate I would expect it to at the very least to tell me the slicing
was done with an out of date profile or to simply report it is unsliced.
I imagine this has happened as features that maybe are more model centric have been added to the profiles.
Maybe we need a new category. Something like "Plate/Model Modifiers" where things like hollowing and erosion and maybe even
burnin details belong and then a plate would contain a model, modifiers and a simplified resin profile.
This might also be a way to allow compositing of more than one model per plate (although I expect that feature is intended for nanoSupport)
I struggle with this now as I have to have several almost identical Resin Profiles which are used for different models due to different requirements
of the model itself or for different plates of the same model but just need to say be infilled but are otherwise identical.
Not to get to long on this but maybe a system where we have a "Project' which contains an STL, Modifier Settings and Resin Profile...these all
combine after slicing to produce the "Plate' entry...
we kinda use the plate now as a 'project' but really a Plate, I think, is and should be a specific post processed group of image files and G-Code which is generated by the slicing system from all the above: (STL, "modifications", resin profiles and even machine details) showing some of the details used to generate the plate in a plate page is still a good idea but it seems like maybe we need a better way to put it all together. A plate is nice because it is a given that with the same resin on the same machine you should get the same result. Subsequent changes to the project or resin profiles or anything else means the 'plate' is out of date..it would still work as before but not reflect project changes.... making a new plate (or replacing one) from a modified project would be 'regeneration' and the new plate would be totally different from the old one even if only one minor change in say erosion was made.
I don't know I like the current simplicity but I also find it a little hard to work with in my environment.
Last edited by macdarren (2019-09-27 18:04:45)
Offline
I am also not happy with current way we categorize configuration, as you see there are lots of modifiers and complicated settings, for example print resolution get affected on machine, profile and also plate page. It is flexible but prone to error and difficult to know what's going on.
We can categorize things based on subject:
* Resin profile: only include settings related to resin
* Machine: Machine related settings
* Job/Print: Maybe I need it to be printed in different size this time or without AA
* Plate: Same as what we have now
But lots overlapping settings on all categorize.
For 2D printers usually we have default settings only, whenever we printing we can change settings and those settings for individual job get saved so on repeated print (not all program have this capability but on DOCX and etc there is XML for settings) we get same settings.
Another issue with current design is that resin producers could not share profile for printers. As it mostly should be formulas including light source output, build platform type and etc.
Offline
I am giving this some thought...I have seen a few partial solutions for similar issues....
Without getting into detail I think the basic idea is that there has to be a hierarchy of settings each adding to or replacing or altering settings from lower levels, like an inverted tree.
Machine is maybe the lowest level, then resin, then general curing, then motion, then project specific and finally model.
What you want is a way to totally specify each category then build to a state where all the info is known and given to the slicer.
Machine is pretty obvious and immutable.
Resin is similar but more variable, it really can only be a starting point.
Curing would blend Machine and Resin into a more defined starting point as would special Motion control (motion might be done in Machine too)
Project would then take into consideration advanced issues (like model masks and supports..see below)
Model really brings it into focus as many settings will depend on model specifics...tweaking all the previous settings which at this point
would provide a pretty decent print but there might be issues with this specific model that need addressing..these might be more dynamically generated.
In addition to this general structure more information is needed about the model and some must be generated by the slicer.
By this I mean the information generated by say nanoSupport must be passed to nanoDLP....this information like what is support and what is not, must be separated from the model so that it can be acted on differently. This would also lay the foundation for things like model masks to fine tune different parts of the model to be handled differently too....if the slicer could handle that than it could then also handle information it generates too and provide dynamic slicing parameters not based solely on layers....for instance large areas could be dimmed more than very small areas on the same plate.
This has led me to an exercise in trying to define all the parameters I can think of and try to see how to divide them up and where there are overlaps.
There may be more 'levels' or fewer but much depends on the final goals aside from 'the best print' for instance as you suggested a printer maker may want to provide several profiles for their machine using reference resins while a resin maker may want to provide resin profiles for reference machines. Both could provide higher level refinement for 'test models'. There might also be profiles for quick drafts vs ultimate quality etc.
Exchanging this tree of info would also need to be considered....if done well it could in theory be possible to import an entire tree then replace sections for the specific hardware or resin or model and still end up with a near perfect print.....the big issue is the variation in similar or different machines even with the same resin (or resins even on similar machines) and some machines even vary individually from unit to unit...so ultimately each user has to tune or the manufactures have to control it better or tune before shipment.
probably this is all stuff you can considered before....
Thanks for all the efforts and continually improving the products.
Offline
Thank you, I like it very much.
1st Phase (3rd party ideally should provide these parameters)
1. Machine settings
2. Resin Properties (Viscosite, How much light need cure time, Max / Min thickness, How strong is it and interaction with tank floor)
2nd Phase
1. Use choose quality expectation (layer thickness and etc) and other small details (it want print to be centralized) with constraints decided on phase before
2. Slicer crunch Model and extract information
This phase a bit vague as slicer info could be used on phase 2nd itself and help user decide on settings
3rd Phase
1. Print Profile (it automatically get generated for cure time / motions)
2. User can modify all settings offered by nanodlp it is similar but limited compare to what options we have now
Offline
Pages: 1