I think the middle ground here is the STEP format. CAD software and slicing software are two different worlds, and it is nearly impossible to see a complete package that excels at both. Slicing is always a complex subject that still leaves room for improvement on the path to getting the best print results possible with the least effort for the printer operator and minimized material loss due to failed prints. On the other hand there are many CAD solutions that are great for designing (and maybe traditional manufacturing methods), but they offer only basic slicing features when it comes to additive manufacturing. So for any designer, be it industrial, art or home use; use one program for designing and/or simulation, and a dedicated slicing program to print the objects. So far STEP is the only solution that translates the object with its complete geometry between those environments, and gives the option to make proper modifications. If there is a better solution, I am ready to be educated.
> Imagine working in Word, Writer, or LibreOffice today, where you have to save your text as a bitmap file, then open it in a separate print program, manually orient it on an A4 sheet, reconfigure fonts, sizes, and line spacing — and only then send it to the printer.
I mean, is that not how it currently works for anything more complex than a term paper? There are specific layout and printing programs, like LaTex and inDesign.
For my normal print situation, I hit "print", which opens a dialog box with a few useful options like color and scaling, and then I had print in a machine starts going "brrr". For my more simple CAD models that's pretty much the same workflow. I hit a button in the CAD program, open up the slicer, pick from a list of profiles, hit a button, and the machine goes brrr.
I feel like the situations are a lot closer than your analogy makes them seem
My problem with CAD formats is the potential commercialisation which will require end users to pay a license fee or cover some other cost to be able to 3D print, in addition to anything they've paid upfront for the file access rights. Yes, better formats are essential but let's not assume they'll be freely available open source as they should 🤷♂️
Let's not forget the wet dream of many commercial 3D file publishers, a format that would protect their rights and limit the ability of the end users to process the files on a platform of their choice. Consider a scenario where a giant like Games Workshop released files with a shelf life of no more than a few years that only allow use on a specific certified device. I know fantasy, but corporations are almost always greedy and never do something for free...
I think that perhaps the argument is here that you should be able to (3D) print from CAD by clicking on something within the program that sends you to a dialogue where you select some options from the 3D printer driver dialogue popup and then let the OS and driver take over in the background until your object is now a material product instead of a digital file.
Of course this ignores some facts of 3D printing that make it more involved than most paper printing in the last quarter century at least. It also ignores that, when more involved paper printing (like poster printing, or high volume printing, or card stock printing, laser etching...) does happen, it actually does usually use a process similar to what happens importing standardized files into a slicer, running several somewhat device specific further setup steps, and THEN "printing". It's just that in those cases, the standard file is a PDF, or vector, or bitmap, or some combination. But in all these more complicated printing scenarios, where the choices go beyond "stretch to fit", and the possible negative outcomes go well beyond fixing a paper jam, there will be some specialized software to go between the output of some text and graphics and the actual output for the specialized device. And, often the "go between" format will also not be lossless for something more complicated than text with a common font and basic formatting.
This article has convinced me that I should start looking for 3mf files when possible, but I don't think we'll have a simple general dialogue box for SLA any time soon (and I am not convinced I would want it if we had it).
Sooo are you saying people who use things like blender shouldn't be able to 3D print? Blender doesn't output to cad formats natively.
Also a lot of Parametric modeling software have their own proprietary file format. You'd be doing the same this just limited to parametric modeling software.
What I tried to say was, STL is and always was very limited, and we need to find better solutions. STEP is a good alternative, 3MF is also progress, but native files seem better.
"How much better? Like 100%?"
No, of course not. But we shouldn't rely on STL anymore either.
This article is basically right; a standard CAD file format would be much better than the triangle-surface approximations we have now. However, as noted by other comments, there is no established open standard for CAD files.
The CAD software manufacturers have no interest in creating one because proprietary formats is part of their customer lock-in strategy, and they don't want to be limited by a common standard that might not be able to do everything they want.
So we are left with what we have, because even a crappy standard is better than no standard.
Are CAM formats universal enough for slicers to be able to work with all of them?
I always saw stl as sort of the pdf of 3d printing....
I think the middle ground here is the STEP format. CAD software and slicing software are two different worlds, and it is nearly impossible to see a complete package that excels at both. Slicing is always a complex subject that still leaves room for improvement on the path to getting the best print results possible with the least effort for the printer operator and minimized material loss due to failed prints. On the other hand there are many CAD solutions that are great for designing (and maybe traditional manufacturing methods), but they offer only basic slicing features when it comes to additive manufacturing. So for any designer, be it industrial, art or home use; use one program for designing and/or simulation, and a dedicated slicing program to print the objects. So far STEP is the only solution that translates the object with its complete geometry between those environments, and gives the option to make proper modifications. If there is a better solution, I am ready to be educated.
> Imagine working in Word, Writer, or LibreOffice today, where you have to save your text as a bitmap file, then open it in a separate print program, manually orient it on an A4 sheet, reconfigure fonts, sizes, and line spacing — and only then send it to the printer.
I mean, is that not how it currently works for anything more complex than a term paper? There are specific layout and printing programs, like LaTex and inDesign.
For my normal print situation, I hit "print", which opens a dialog box with a few useful options like color and scaling, and then I had print in a machine starts going "brrr". For my more simple CAD models that's pretty much the same workflow. I hit a button in the CAD program, open up the slicer, pick from a list of profiles, hit a button, and the machine goes brrr.
I feel like the situations are a lot closer than your analogy makes them seem
My problem with CAD formats is the potential commercialisation which will require end users to pay a license fee or cover some other cost to be able to 3D print, in addition to anything they've paid upfront for the file access rights. Yes, better formats are essential but let's not assume they'll be freely available open source as they should 🤷♂️
True, but my focus was more on professional users and industrial grade designs.
Let's not forget the wet dream of many commercial 3D file publishers, a format that would protect their rights and limit the ability of the end users to process the files on a platform of their choice. Consider a scenario where a giant like Games Workshop released files with a shelf life of no more than a few years that only allow use on a specific certified device. I know fantasy, but corporations are almost always greedy and never do something for free...
I think that perhaps the argument is here that you should be able to (3D) print from CAD by clicking on something within the program that sends you to a dialogue where you select some options from the 3D printer driver dialogue popup and then let the OS and driver take over in the background until your object is now a material product instead of a digital file.
Of course this ignores some facts of 3D printing that make it more involved than most paper printing in the last quarter century at least. It also ignores that, when more involved paper printing (like poster printing, or high volume printing, or card stock printing, laser etching...) does happen, it actually does usually use a process similar to what happens importing standardized files into a slicer, running several somewhat device specific further setup steps, and THEN "printing". It's just that in those cases, the standard file is a PDF, or vector, or bitmap, or some combination. But in all these more complicated printing scenarios, where the choices go beyond "stretch to fit", and the possible negative outcomes go well beyond fixing a paper jam, there will be some specialized software to go between the output of some text and graphics and the actual output for the specialized device. And, often the "go between" format will also not be lossless for something more complicated than text with a common font and basic formatting.
This article has convinced me that I should start looking for 3mf files when possible, but I don't think we'll have a simple general dialogue box for SLA any time soon (and I am not convinced I would want it if we had it).
Sooo are you saying people who use things like blender shouldn't be able to 3D print? Blender doesn't output to cad formats natively.
Also a lot of Parametric modeling software have their own proprietary file format. You'd be doing the same this just limited to parametric modeling software.
What I tried to say was, STL is and always was very limited, and we need to find better solutions. STEP is a good alternative, 3MF is also progress, but native files seem better.
"How much better? Like 100%?"
No, of course not. But we shouldn't rely on STL anymore either.
This article is basically right; a standard CAD file format would be much better than the triangle-surface approximations we have now. However, as noted by other comments, there is no established open standard for CAD files.
The CAD software manufacturers have no interest in creating one because proprietary formats is part of their customer lock-in strategy, and they don't want to be limited by a common standard that might not be able to do everything they want.
So we are left with what we have, because even a crappy standard is better than no standard.