You can SKIP 3D modeling now
HTML-код
- Опубликовано: 19 сен 2024
- Full Control GCode is a new approach to getting your 3D printer to print exactly what you want! And it's not just useful for creating weird vases, it's actually a great tool to make functional parts as well.
Thanks to Private Internet Access for sponsoring the video: www.privateint...
Get Full Control Gcode fullcontrolgco...
Lampshade prints on Reddit by Flashlightpic5-3218 www.reddit.com...
Filaments used:
DAS Filament PETG B-Stock www.dasfilamen...
Prusament Galaxy Black PETG go.toms3d.org/P...
Product links are affiliate links - I may earn a commission on qualifying purchases (at no extra cost to you)
Read the article to this video here: toms3d.org/?p=...
🎥 All my video gear toms3d.org/my-...
I use Epidemic Sound, sign up for a 30-day free trial here share.epidemics...
🎧 Check out the Meltzone Podcast (with CNC Kitchen)! / @themeltzone
👐 Enjoying the videos? Support my work on Patreon! / toms3dp
Thanks again to Private Internet Access for sponsoring the video! Check them out at www.privateinternetaccess.com/Toms3d
Has this video inspired you to try Full Control GCode - and if so, what are you going to print with it?
I don't have an Office license to play with it with, but I'd love to be able to print some more workshop organizers for my modeling paints. They're functional geometric shapes, and would probably benefit a whole heck of a lot from printhead movement optimizations like the ones you mention for the battery organizers.
Private Internet Access was recently bought by an Isreal based malware distributor. So, if you actually care about your privacy, be wary about PIA, Express VPN, Zenmate, & more.
UHM... Yeah but no.... It is already very hard to get good prints rights, adding "full control" to the process will make life miserable. I will just model the part better I am pretty sure the time I save there vs doing it on full control will be much more than the saved print time.
@@dinosoarskill17 The PIA/Kape merger was a few weeks shy of two years ago. It’s no longer recent, or relevant.
PIA software is open-source. You’re free to verify that it contains no malware.
@@therealmakmillion it went through finally recently, the merger talks started 2 years ago
Print time: 30 minutes
Programming time: 8 to 52 weeks depending on experience
lol yeah this definitely ain't for beginners!
Natural Language Processing models, such as GPT-3, may make this a lot more accessible.
this exactly
Someone will simplify and create a gui that makes this more accessible to the layman. Most likely in the form of a cad/slicer, no worries.
@@Iamwolf134Shove your AI where it don't shine.
this is like writing in assembly vs writing in python
you gonna have so much more control with the tradeoff of not understanding what you're writing a few days after you write it
And then it gets released in Python: mind blown. =)
My immediate thought was "This sounds like OpenSCAD with extra steps" ... but the longer I think about it, the more I'm realizing that it's quite the opposite: OpenSCAD has a completely different model, it only cares about geometry. The steps there (add, cut, rotate, ...) are to create a final model, but they have nothing to do with actual physical production. FullControl GCODE on the other hand only really cares about nozzle movements - the geometry created is almost just a side effect. As you've shown, this can be very useful for optimizing paths.
I really like openscad but I also really like how it can work with almost raw math.
I was thinking of OpenSCAD as well. Changing it to output GCODE directly and hopefully also include G2/G3 instead of chopping up circles/curves to facets would be nice too.
@@roysigurdkarlsbakk3842 That would be a lot of feature creep for openscad. Filaments, materials, support, bed heating, fan control, all sorts of stuff that are beyond its remit.
OpenSCAD does solids, this does.. lines only! Theyre very different I'd say.
OpenSCAD does not know how to compute extrude speeds for example.
Thinnk this would with better with something like implicitcad wich only does mathematical definitions of geometry instead of the aproximation openscad does while rendering.
Using excel is a bizarre implementation, but I will give it props for creativity.
It reminds me of OpenSCAD but specifically for FDM printers, instead!
Well it stands the reason that you could do this as an add-on to open scad. Let's hope some sharp bit weenie takes up the project.
Yeah, this seems like it could be very easily implemented in a scripting language. But I guess spreadsheet interface is somehow more accessible?
a python version is on its way!
@@RonaldoMessina well yeah I figured as much.. but having it right there in scad would be pretty sweet. So you could use its display to see what you were doing. I guess you could still do that with a mix of python and scad, don't know
I bet this academic person in the course of doing their work has become very proficient at Excel and not at anything else the least bit programmatic so... here you go.
This video just makes me appriciate slicers even more.
😆
My first reaction is that this would be super useful for small manufacturing of things that are basically simple shapes - for example, Thom's battery holders. The extra time spent setting up the gcode will pay for itself after a few dozen units. But for one-offs or very complex shapes (like, say, a miniature of R2-D2 which is what's on my printer right now) the time and plastic saved from optimizing the gcode is not going to cover the time (and wages, if a business) spent creating that gcode by hand. That said, it's still very useful to be able to get under the hood from time to time! Glad to learn this exists.
Pretty much my take away as well. This could also be huge for cosplayers or anyone else looking to save on print time & post processing while trying to a create more uniform strengths.
@@hazonku I doubt it would be useful for cosplay. Generally they are using pretty complex models made by other people and even if they are making themselves using a modeler is way easier than trying to input basic shapes into a very complex structure without any GUI.
This is really only useful in extremely niche cases.
If you have the budget for large projects you should be optimizing your fill the best possible ways you can within reason.
Super cool, very much like structure synth but straight to Gcode. Maths hurts my head but might have to suck it up!
ugh but forced msoffice is cancer
@@natalieisagirlnow it's a shame, I refuse to pay for office but 30day trial lets us play a little at least.
I'd be curious to see why it didn't work in libre office, it does have vba support. In any case, looks like a Python version is coming soon. This seems like it's for people who find openscad too high level.
The idea is being around for a while. Back in 2017 wrote my own program to do similar things after got inspiration from some article.
And there is not much math in a whole thing: just a series of very simple and easy to grasp concepts. F.e. (in case you doing your own programming) - how much length to extrude when doing a line of x mm with some given layer's height.
Definitely an interesting approach - but you should think of real problems where it may shine. In most cases regular CAD is a way to go.
Drugs are bad, m'kay? Don't do drugs! :D
This should be a python package
edit: seems like the authors agree and there will be a python version soon, yay!
Totally. The python package is screaming out loud for this.
Going to be the most handy way to use this. Yes.
VBA I guess. It should have be done in Python in the first place in my opinion.
Yeah, python would be an excellent way of realising the code as people are already using it for GUI implementations. The only real difference would be the 3D aspect and the gCode translation.
@@古德曼-t1c It would be fairly trivial to implement in Python or any language you would care to use anyway. gcode is very simple and only a few commands are used for most of the printing. Most of it is just telling it where to move the nozzle and how much filament to extrude and then add some boilerplate at the top and bottom. Personally, I'd use Perl.
i cant decide if implementing this in excel is genius or an awful hack :D
its a neat idea though, if a bit niche :)
Its an awful hack. Sure it was probably quite easy in the beginning, but the level of complexity where it should've been remade in a better language/framework for the task is long overdue at this point. There is a lot more potential left on the table, even such simple things like a live preview, which just cannot be added to exel, even with macros.
Well, they seem to think it was an awful hack and are re-doing it in python, according to their website
maybe it's just... both
i think it's semi-genius. It begs to be implemented as a plugin for a cad, to allow combining the tools. Some CADs already have spreadsheets built into them.
@@victortitov1740 By the same token, many CAD programs have some scripting API that would be more efficient and flexible than trying to port it to their particular implementation of spreadsheets - which wouldn't include VBA macros anyway.
I've been playing around with non planner printing recently.. this will certainly come in handy.. thanks allot Tom
This looks like a really useful tool! I'm not sure yet what application I could use it for myself, as most of the parts I print are designed to be quite complex irregular components to larger projects. However I do think that being able to optimise print times for larger, more repetitive components would be very useful. Thank you so much Tom for making this great video about it all! 💙💜
This is fascinating! I really hope to find some time to get a trial of MSOffice and try this out.
Or better yet someone should refactor the VB to a real language so the performance isn’t so awful an you don’t have to use Excep
VBA macros work in libre office.
@@MatthewSmithx yeah i don’t really like using excel
Yeah so not only does libre office support vba (it should work, not sure why he has trouble), the website quite clearly says a Python version is forthcoming.
You can get a Microsoft dev account and use Office for free for 90 days. As long as you use the account you should be able to extend it indefinitely.
I can see this being very useful for small parts. I recently spent multiple days trying to model cable combs for my PC. These are super simple, an array of circles. But in the slicer, I got a bunch of those infill artifacts you mentioned. So I kept tweaking the 3D model, then my slicer settings, then the model again, trying to trick it into doing what I wanted. I am going to try using this software to generate the part. Thanks for the info!
prusa also has this super-annoying thing where it always, always (it is impossible to defeat afaik) tries to make near-vertical walls thicker. This is very unneccesary most of the time and it slows down prints insanely.
It's like hand optimized assembly but for 3D printers.
My thoughts exactly.
Great - a way to make 3D printing even more complicated. My only hope is that the sort of optimisations that this can provide, can be built into regular slicing software. Better model and print move analysis seems to be the key.
Great tool. I can imagine it's not at all useful for the one-off print where you just need to get something done, but for a bulk manufacture case it makes a lot of sense to try and save every single printer minute possible. Also would make it *much* easier to automate things like knocking the part off the bed for taller parts if you're doing continuous printing.
I can imagine Josef Prusa drooling over this and planing an implementation in PrusaSlic3r, Josef… your move.
Wow it requires a scary level of pain tolerance to build something like that on top of Excel :-)
Seems like feats like this is common in academia.
@@mtenkawa and in business. The increase to the maximum row count was cheered by many accounting firms.
"The possibilities are endless.. but it has limitations".. It has more limitations than possibilities. Very, very niche tool for something very specific.
Not really. It's just because we still not where additive manufacturing needs to go. Right know we are using a modeling program that is designed to show models in a virtual reality. Then we use an other program to disassemble it into slices and making 2d versions.
When 3d printers get able to use non planar printing we need something like this..
It just needs to be remade in a better programming language
When mass producing, this would save thousands of dollars
This is making me think that 3D editing software needs to have a "3D printer" mode whereby aspects of the modeled objects are tailored to being 3D printed, like you tell it you'll be using an 0.4mm nozzle with 0.2mm layer height and the snap-to-grid functionality is adjusted to work to those parameters, ending up with objects that are more "3D printer friendly", which would be especially helpful for mechanical prints.
makes me think it openSCAD could be modded to something like this? instead of using uh excel...
or etch a sketch mode, with a 3rd z dial
I don't know how long it's been there but Blender does have very rudimentary and basic 3D printing visualizations. It will color highlight overhangs or intersections that can be fixed before exporting.
I was thinking the same thing. 3D programs are WAY ahead as far as modeling and automated repetition when it comes to slicing. They SHOULD have this option to control and out put gcode right from the start. This "full control gcode" I think could possibly be written as a blender addon... maybe. Blender already has a slew of mathematical nodes for shading and geometry.
I set user parameters in Fusion 360 to optimize these things as you say. Any shells are multiple of my extrusion width parameter, etc. It takes more intentional planning up front with regards to how the model will be printed and used in order to set it up the right way, especially with more complex models, but it's powerful and functional all the same.
I feel like I am looking at the first computer generating a line of dots on a screen and within a short time the fine tuning and simplicity generally comes along. This type of tech if its developed will be a revolutionary leap for printing. Imagine doing a whole complex item such as a 3D printed plane all in 1 (or 3) easy spreadsheet tabs. Thanks for making these vids ...... love your work. stay safe and cheers from down under mate.
This looks good for simpler geometries, or geometries which can be defined with mathematical shapes, where you really want to optimise the GCode toolpaths. Its a bit of a dark art getting slicers to do what you want sometimes, having full control is cool, but far too cumbersome for complex object modelling. This could be good for production environments to produce simple objects efficiently like that battery tray. Looks like one of the only options for non-planar slicing other than GCode post processing that I know of.
Besides, that battery tray could have been done using a repeating squiggle line going in a mesh pattern, batteries dont care about uniform support.
I don't think it matters how much work it takes for a complex object, it will be worth it for some - it only matters if you are only making a tiny number of them so upfront optimisation costs won't be paid back...
So most of us probably won't use this sort of tool for a complex object, but somewhere that makes a kit using 3d printed parts for example might well - every part they have to make takes less time to print, and uses less material than the slicer generated object would. (Though I could see the folks that like OpenScad finding this a very easy method to get exactly what they want - it appears a similar thinking process)
I am surprised this didn't exist all these years already
It's like the parametric control of gocode. Super interesting
As a machinist and programmer, This is gonna be a game changer for me. I often write full programs on the machine itself as it's much quicker than modeling it in a computer and post processing it to work the 5 different machines in the shop. This is gonna be awesome for making even quicker prototypes.
It seems to me it doesn't do anything you couldn't do yourself. Gcode is crazy simple.
This would be an amazing openscad add-on
For the most part, OpenSCAD can achieve much of this with far fewer keystrokes. However, OpenSCAD probably could have some new capabilities if it could be blended with this.
@@johndoner6245 true, and export the model to a splicer, when you optimize the model properly, could work better. But examples like this are great to show its possible.
Or even cadquery
This reminds me of when I used to make cool designs on a graphing calculator during math class.
I 100% know that I've made the shape at 4:34 on my ti-84 before in class
Just like a Post Script printer. You don't send a Circle to the printer, you send a mathematical description of what a circle is. Quicker, and very accurate. I love the idea for a 3d printer.
I cannot overstate how excited this makes me. Up until now, the only way I knew to design a propeller required me to manually discretise it’s curvature (I’m not sure if there is a way to do it continuously in F360, I couldn’t find what I was looking for.) which didn’t sit well with me. I wanted a way to define an angle wrt radius, I thought openSCAD might be the way but didn’t look too far into it. Thanks for the great video.
Openscad is awesome, but it approximates curves
@@JohnDoe-rx3vn Everything approximates curves. Gcode moves in straight lines for 3d printers (unless you know of any which take the circle commands and even then you'll be limited by the steppers and the firmware).
Wow!! I have started learning it and it feels so good to be a student of Proff Andy and get supervised by him for my major project 😃
I know I'm late to the party but this looks very useful for me. I've been using the custom non-planar slic3r for a while and love the idea of this. Definitely need to try this out
Cura 5 came in rescue, except for the non-planar part.
I'm thankful I didn't went back to Windows for that.
That shot with the focus slowly going down the side of the part was an awesome shot. Very nice.
Thanks for bringing cool stuff like this to our attention!
Great video. I wonder how the speed and quality compares for you battery example if you were to use CAD and build it up the same way with thin wall circles + squares that are repeated. That would get rid of the infill weirdness that you mentioned and would be a more direct comparison.
I applaud you for trying something new and sharing. What I say next is not meant as a criticism but only as an alternate approach.
You could have gotten rid of the tiny weird wall fills with a single setting in the slicer. Minimum extrusion length. Make that sufficiently large enough and those short BS fills go away. Also, if you pick the right line width and Union overlapping volumes, you get more weirdness that just goes away when trying to do something like this.
That leads be to the tab at the bottom. That could have been a separate model that you overlap with the cylinders. Again, with the union overlapping volumes turned on and a singular isolated resize in the z axis you can make that tab at the bottom one layer thick.
I could have done everything I've seen so far in the first 7 and a half minutes of your video in about 7 and a half minutes of knowing just the very basics of how to model and how to leverage the slicer settings. In fact I'm half tempted to try making this battery holder but for different size batteries.
I'm not taking away the value of what you show.. the mathematical approach to being able to generate GCode is fascinating, wonderful, and really does open a whole new world. It's just that the example you gave really is much easier to do using a technical / tool based approach where the wavy shapes you demonstrated using functions really is where the mathematical approach shines. I get that you wanted to make the topic approachable. I just don't want people thinking that the battery holder couldn't be done much easier using a better understanding of the slicer and a better understanding of how to model.
So what rules do we want for a challenge? 1 line width cylinders, max line width percentage, tabs at the bottom that are one layer height tall. The cylinders need to be welded together so he wall overlap will need to be carefully considered.... The thing is SmartAvionics on GitHub has a custom version of Cura that does an amazing thing with variable line widths where two wall lines need to fit within a space
Yeah this is looking pretty easy to do using basic tools. I could probably use TinkerCad generated STLs and simple grouping either within TinkerCAD or within Cura to do this same thing. Both support measured movements, so you create the smallest block, group it together, duplicate it, move it by the size of that block, group that together, duplicate it, move it by the size offset again, group, duplicate.
Yeah I can get that workflow down to about 7 and a half minutes.
I noticed they are working on a python version. Could this be used to build an addon in Blender to optimize my drawings directly into a gcode export file?
Maybe but it misses the point.
@@cprogrck I guess he wants a simulator for the code, which is not that easy btw. We tried that years ago but decided the CAM software from somebody is better for this.
I appreciate how you always come up with those rather uncommon things! It makes your channel really outstanding!
Seems like something useful for making airfoils with smooth surfaces for wings and fanblades. Maybe racks and gears too.
That’s a very interesting new angle of making gcodes. Thanks !
This reminded me of OpenSCAD - where you have 3d modeling aimed at programmers, and there's that neat talk by Matt Adereth and his dactyl keyboard which he programmed and 3d printed using that approach
Openscad is pretty intriguing, if I could create code of it, that would be awesome. But as soon as you have more complex stuff, it maybe gets useless.
But he made a great point with small edges, that takes too much time, so you could have the same result when you programm the model on Openscad, export the STL and the slicer makes the rest. I would even say it could compete.
@@Spelter It's fine even for complicated stuff, though? Just modularize. The Dactyl is a pretty convincing proof of concept that you can get complicated results with it. ( though , technically, it's clojure over scad, but it's the same thing. )
@@SiriusXification True, I do the same in OpenScad and make small parts and modules with parameters for various things, but the splicer needs to make better gcode.
I found superslicer some time ago, which has various new options for the infill, which takes most of the time and it's not that bad.
I was trying to model a pretty complex mathematical shape in F360 a while ago, and it was essentially impossible for a hole host of reasons. Not only is F360 a big bugfest, it has really poor memory management. The few times I could actually get the model simple enough for F360 to be able to extrude a surface, it would start to hang and take up to 20 minutes to process a single modification.
This would be really useful for that,
Had a friend do this with machining. He saved so much time. Instead of printing 200 brackets at a time. He machined 50 brackets on an aluminum plate at a time
My jaw dropped at that Z hop feature. That's going to be huge for cosplayers to save on materials. If you can just work the z into the design that means free strength & less filling & sanding.
Super cool! "Traditional" 3d printing process workflow is 3D Modeling->STL->Slicer->GCode. FullControlGCode is basically trying to take out the tessellated/triangulated STL manifold from the equation and use the 3D Modeling information about how the model is actually designed/created into GCode.
I've been following the guy on Twitter for a while now, some really cool prints you can do using this. Glad to see it get a bit more exposure, I hope it continues to evolve and develop.
I can only imagine a conversion tool to be made from vector files can be visually drawn on autocad or sorts and it just does the rest, honesty this is super cool
I can see how this is useful for making interesting 3D printed art, optimizing something that you need to print a lot of and tinkering with non-planar printing (which is interesting and kinda mesmerizing).
That transition at 7:21 Holy god damn! Clean!😎
Here's an idea: What about combining your squiggly line formula into the battery holder so that every layer is 1 continuous line (like vase mode) instead of individual cylinders? Sort of like a semi-circular shaped sine wave.
The resulting shape may not grab the batteries on all sides, leaving small sections of the perimeter open, but that still should be more than enough for a secure hold and would result in the fastest possible print times.
For one off pieces this would be unnecessary faffing about and pointless but if you're trying to pump out tens or hundreds of these things in one go it would save so much time
The next version of FullControl Gcode is coming out soon written in Python, so a lot more versatile and efficient than VBA
I mean, it's literally "write header to file, repeatedly calculate coordinates and write to file, write footer to file". Not brain surgery.
7 months later and it is still coming soon
Thanks for this, it addresses a lot of things I want to do with my printer.
I've now subscribed to your channel on the back of this ;)
I'd like to comapre this to assembly programming and compilers. Nowadays in 99.9% of the time a compiler will generate better machine code than a human progammer (the 0.1% being really niche stuff). But for a long time in history compilers were not that smart. We're at that point with slicers, Full Control Gcode shows us a what a really smart slicers might be able to do further down the line.
Really cool stuff! Thanks for bringing attention to it.
it would be interesting to see CAD software built specifically to model stuff for additive 3D printing and be able to do some of this more easily vs having to manually create features in excel
I think that it the only way to go if this shall be a useful way to do thing. Without the support you have in CAD, making any complex model in this way would be madness or you will just go that way yourself. :-)
@@bknesheim exactly, but imagine the possibilities of an additive 3d printing oriented CAD software that could take advantage of this simple but kinda revolutionary idea!
@@byzhiphop Yes and I do think that you could get a lot stronger arcs if there are no layer on the edge.
Very cool and you still have more speed in the battery holder algorithm by alternating the start position of each row for less travel time
Good PoC. Hopefully someone will bring it to more usable environment.
Finally a step closer to Designing / Editing right in the Slicer's Project plate.
About time.
coooool! this is like animation nodes for your printer! This is super awesome!
Awesome! I love seeing innovation in the 3D printing community. Hopefully this will eventually result in some optimization with the current slicers that we use. Great video.
except it's based on the worst closed source cancer
This would be amazing with like, python or something, instead of excel.
CADquery is probably the closest but yeah python would have been nice here too.
Hopefully this leads to more optimized, dedicated software solutions that can run faster and work in conjunction with a Gcode visualizer of some kind. I don't know much about software development, but could it work as a Cura plugin, for example?
I was thinking the same thing. I despise excel and their convoluted functions... Syntax is a nightmare
python can easily interface to other software too.
In python it is relatively simple to read an STL file and slice it, or use it as an exclusion zone.
(I have written a python script once with STL slicing and gcode generation for printing a mesh like structure.)
Indeed, Excel isn't built for this (seems my original comment stating that was removed??). Anyway, the problem is that any little mistake causes big issues, especially if you're free to edit it. This is why a programming language was invented basically. Excel is for data representation mainly and should be used as such. Matt Parker has great fun in showcasing all the cases where Excell was used as a database etc. :) Having worked in science myself, this is very likely because the researcher just used something he was familiar with.
Now we bust need something between this and a traditional slicer. Something where you can generate moves automatically, but edit each single move manually
You can edit auto-generated G-code in any text editor, it's just as much fun as editing auto-generated HTML.
Get this, you could also do this in F360 by just creating a circle with one perimeter and the same square type. So you won't get the infill but you will still perhaps 10 minutes more, but that is saved depending on the sketch from going from Full Control GCode to F360.
For hobbyists, this is probably a bit complicated. But as an engineer i can see, how this could help me some day. Thanks.
I was learning how to program Heidenhein in high school. Compared to more modern CAM software (for example Alfacam back my days) it was slow and uncomfortable. So for me this is a way back. Slicer is basic CAM SW but it does not let you program tool path. That could be interesting for next gen slicers.
Awesome Video. It is amazing to see what you can do in Excel and VBA.
This seems so useful to me. I can wrap my head around it fairly easily. I usually kinda suck at achieving good precision when I use graphical cad sofware. This seems so much more logical to deal with when I am trying to achieve high precision. Only little bummer is that I have libre but I am sure that I can just get office for this purpose. It is honestly worth it.
Chefs kiss to the macro shot of layers at @8:05
You can skip 3D modeling as well as you can skip designing a PDF with your favorite drawing software: technically possible, but an edge case for projects of limited size and complexity. Also, you could have done so since forever. 😉
That said: nice prints, and an interesting approach for those cases where the slicing process would introduce problems. I wouldn’t delete the slicer, though. 🙂👍
(Reposted from a reply because I'm curious what everyone thinks.)
Do you think direct control operations could be added to slicers?
That's what makes me most excited about this. I've been dreaming that one day slicers will add "hey printer, do this" type commands to their programs. Or at least some sort of gooey system for editing the G-Code. I mean honestly.. Why can't I just click a layer and tell the printer something like "pause for 3 minutes", "add a line here", or "insert district control spreadsheet from file". I don't want to turn slicers into fully blown CAD programs but why not have some tinkercad type functions in there?
I started writing G code in 1979 for Wire EDM using a teletype. We have advanced a long way since then.
I think that the "printing" element of 3D printing is about that software abstraction that covers the conversion from logical data object into pile of melted plastic spaghetti. This is a very clever setup, but it's a bit like hand optimizing a bit of software in assembly, it's only useful if you're going to run that process a lot so the cumulative efficiencies mount up.
This is why I am subscribed to this channel!
I appreciate having my horizons expanded. Great Video.
So finally we came full circle with CAM programming ;-)
This is neat. Should have been done on the Browser, in JavaScript - not in Excell with visualbasic macros.
Gawd dayum! That"s a fresh looking hoodie.
Blender 2.79 and later updates can make shapes with math equations too(possibly earlier versions, but I'm still using 2.79)
Very cool! There's some seemingly simple shapes like these I've done many a time in CAD that take a surprisingly long time to actually model
I assume that G-Code generators will be the new alternative to slicers; although I think keeping this entire project inside of an excel sheet is one of the sillies things I've ever heard. This could easily be a double view program that allows you to script and visualize at the same time.
I can't think of any advantage of implementing this in excel vs anything else.
Accessibility for a larger audience?
Most likely simply because the person/people working on it already knew how to work well with Excel.
When you make a tool, especially when doing so primarily for a research paper, you just use what you know. No reason to make it even harder on yourself than necessary.
It might be as simple as Excel being the GUI-creation-capable tool with which the creator was most familiar. After all, *the best programming language for a project is often¹ the one you already know.*
Other languages may have all manner of advantages, but having to spend weeks or months learning another language may nullify its benefits, result in the project not being finished, or even not being _started_ if it's expected to take too long. Yes, learning a new programming language/tool is often a worthwhile investment in itself, but the time to do so may just not be available.
My understanding is that this was a university research project, so _having some results to show fairly quickly_ could have been important.
Personally, I'll always favour "tool that actually exists, but using sub-optimal platform" over "highly-optimised and perfected tool that the creator hasn't had time to finish/start" 😉
¹ "Often", not "always".
@@sagichnicht6748 excel is behind a paywall
@@RonaldoMessina Yet, many possess it and even more know how to use it, many of them not literate in any programming language. Just saying. Not claiming it is necessarily the best choice.
Interesting. Ive used shells to make those with similar results. Those vases are cool but im pretty sure i could model them as well
You are always pushing the limits man! Good stuff!
Reminds me a lot of Prof. Taekyeom Lee's work with ceramic printing.
Yea poor version of Rhino w Grasshopper
I feel like its obligatory to say, "Nerd!"
Very cool video. I have always wondered about things like this. Im constantly fighting with slicers trying to get them to stop doing certain movements and timewasting actions. This is great!
A really fun concept but as you also mentioned probabaly not that useful for most people. I would spend more time making the gcode than I would save on the print time of that part hehe. Probably fun to just play around with though and really cool!
We both know someone who likes to produce vase STLs. Might be another tool for his process.
@@OldCurmudgeon3DP Absolutely, for that it's a great tool!
really promising program. i tried the non planar cylinder and 3 more toolpaths . It has amazing potential
As always, great video Tom! Full Control is pretty amazing! Started messing with it the other day from your tweet. Being an Excel Wizard may now be a pre-requisite for 3d printing! Loved the information because I did find myself running into issues. Definitely some limitations, but that is fine! Will be curious to see if it gets integrated in some regards into the popular slicers.
This is like vector vs bitmap. Legit.
Reminds me of OpenSCAD but more direct to GCode
We need a tool like that integrating a language (can just be python with libraries) and a live gcode preview.
Surely something like OpenSCAD would be easier than re-inventing the wheel.
Exactly my thought. Similar speeding up might simply be done by optimizing sizes - f.e. using multiples of layer width for walls.
I see this being great to create some basic shapes i recently found gridfinity orginazation and this would be great for making some simple bxes to go with it.
regarding battery holder: just turn off infill and set perimeters to 1 within slicer and you'll get even better optimized result
Using the Excel add-ons that enable the use of C# or Python to define functions would make this a little easier 😄
I can't lose the feeling that this would have been so much more useful if it was simply a Javascript object which you could call the line and other functionality on from JS code that you write yourself AND if that object would not only generate gcode, but also immediately render it in your browser while allowing you to edit the JS code in the same UI ...
Sure, but it may not exist at all. Like any comment about an open source project: feel free to build your own, if someone has excel skills - it makes sense for them.
Rendering stuff is way more complicated that generating gcode, where the only thing you really need to compute is the extrude speed.
Awesome tool for mass production 3d printing!!! shortening the time = printing more = earning more at the same amount of time.
7:21 amazingly done and super smooth
I’m reminded of hand coding gcode for the CNC mills I used to run.
That's cool. We need a slicer which can convert models into full control gcode on a 5 axis 3d printer. No supports will be necessary then for many print designs.
It's like OpenSCAD with gcode generator and slower.
Fun fact: Did you know, all Prusa's printers, up until later development of Mk3S were modelled in OpenSCAD?