Just a few years ago, the project was little more than a curiosity that can make textures. Now, it has tons of nodes, funding, third-party contributors, texture painting, no license DRM or subscriptions, and a lot of power. For me anyway, it is enough where I do not feel the need to have to buy into Substance Designer as a result of Genetica becoming defunct.
The FOSS nature meanwhile means you can optionally pay for it as you go to the download page, but payment itself is not required. It is recommended though to subscribe to its Patreon page if you have something to spare (which could be your Substance money if you choose to discontinue that subscription).
After numerous failed attempts by FOSS to create a Substance alternative (remember Neo Texture Edit?), we now have what looks like a rather robust app. with a small, but healthy development scene.
So Iām starting to take a bit of a closer look at this and it seems rather promising, at least in my very early stages of getting serious about texturing and trying to create good looking materials.
So far, 2 things come to mind that seem to be holding it back.
There really arenāt a lot of tutorials for it, at least not compared to say something like Substance Designer.
While it has a fair number of nodes, on a so far quick comparison to Substance Designer, thereās a reasonable number missing.
The strange thing is, from what I can tell, itās actually possible for anyone (with knowledge of GLSL and some ācodingā skills, so that totally rules me out) to add new nodes/libraries.
As such I went looking for some, but no such luck. Iām not actually taking about complete materials or node graphs like the asset library you can get from the actual website, but actual new nodes, like wood or fur.
I had hoped to find the most obvious collection of new nodes, that being a set which basically duplicates a bunch of Substance Designer nodes that arenāt already in Material Maker. That way, while there are some UI differences, if there was a good match up of base nodes, it would be much easier to follow along a Substance Designer tutorial (of which there are a lot) and create much the same material.
I expect Material Maker needs to still go through a significant UX redesign before we start seeing itās adoption by the kind of users whoād be willing to invest their time in creating these kinds of value-adds.
Iād be interested in developing/adding some new nodes, but I refuse to use it in itās current UX state. Certainly an interesting project to keep an eye on and itās potential is immense to be sure tho.
Itās UX really could do with some work, hell, Iād love to just be able to keymap different key/mouse options for general navigation so I could match pan/zoom, etc to be the same as I have in Blender and to be able to use a Tablet. I wonder how much of that is tided to Godot tho.
Even so, in my early tests Iām getting some good results, Iām just finding it a little surprising that there seems to be so little interest in it. Very few Youtube videos, nothing much in the way of forum posts, etc, the reddit thread is half dead.
Is it a case of half just donāt need it and the other half use Substance Designer anyway even with the Adobe ātaxā?
Iām with @Felix_Kutt , I tried it and was immediately turned off by the UI/UX. I donāt have Substance (or any Adobe products scams) so I was hopeful that Material Maker would fill in Blenderās gaps for me. Maybe in a couple years Iāll give it another look and see if itās more usable. I imagine youāre not going to see a lot of general interest until that point, aside from early adopters who can power past UX annoyances. Your average consumer doesnāt have the time and energy it requires to figure out how to use MM effectively
Playing with it sinc 0.94 (?) ā¦ This is using the godot engineā¦ so some UX changes arenāt possibleā¦ for me it was my first ārealā contact with shader nodes texture generationā¦ (yeah @Ace_Dragon i remember NeoTextureā¦ programmed a real spherical node there but didnāt really understood the concept back then). MM also has SDF.'s .
It is (still) fun for me because with my i3-3220ā¦ no extra GPU back thenā¦ now a bit better, sadly on my slower computerā¦ ) because the big brother (no i will not mention itās name) wouldnāt even run on my system .
There is also another similar app : texturelabā¦ (on Electron/vue.js/three.js) looking more like the big oneā¦ but with lesser nodesā¦ and smaller community (?)
Ohh and just have a look at the assets page (Materials):
What i meant was: As far as i understood this it is using the node system of godot engine editor.
Yeppā¦ but itās a bitā¦ unusual (for me) because of the translation of the parameters to the godot engineā¦ but anyway you can look into any source of the nodesā¦ in the editor itlself not the source code of the whole thing.
And here is a quick example of the Material2022 challenge result (jpg with 80% enough to watchā¦ 400KB instead of 2.6MB png )
My substance version is old. I have it via steam and the version is designer 4 and painter 1.5. But even at such old versions I see no advantage using MM 1.0 over them plus the UX is night and day.
But make no mistake, I have stopped renewing Substance and Iām expecting great things from future releases of MM. I do genuinely think itās got the potential to flourish into something wonderful.
I donāt buy that. As a godot user myself, itās toolkit/control nodes are actually quite flexible and great. A lot can be achieved with it if one is just willing to put the work in. Not significantly different if you were to use Qt or what have you.
Theyāve just taken the easy/lazy road for 1.0 which is understandable. Just getting this thing functional as an effective proof of concept is a lot of work after all.
I noticed that, but not looked at it yet, not sure thereās much point in it. Personally I think the dev would be far better off not having painting at all and focus on the āmaterialā side. If you really need 3D painting, then outside of Blender, its likely better to just use ArmorPaint.
Yeah, I noticed that, it even applies to the final output node, so one could add extra image map saving, if you wanted too. But its all way to much like coding for me and that just does my head in.
Thatās my hope, just look how far Blender has come in the past 2-3 years (since the 2.8 release). My worry is that without some attention/love, that the developer will just treat it as a very causal side project and slowly lose interest over time.
Leaving us with few options other then Substance Designer.
Theres many ways to skin a cat, using the graphnode in godot doesnāt imply you have to do it in exactly the same way shadergraph in godot does(which is itself a rather horrid UX). But again the UI toolkit is pretty flexible and certainly more and better can be done with it.
But as I said, they took the easy road which is to do the bare minimum with the graph node resulting in something similarish to the shadergraph in godot engines editor.
And now Iām the confused one. Not sure how to read it or what to read out of the sentence. At a blind guess, you mean how you can drag and release a nodule onto empty space?
Beyond itās initial release I have not payed much attention to it. Donāt really know whats going on with it.
Yes, one of the keys to success will likely be to find other contributors to the source who would take over some of the other parts. Someone to be in charge of the UI/UX perhaps and someone to act as a dedicated developer of the painting subset.
But no, thereās no reason they should be separate applications similar to substance, Iād even go as far as to say IMO itās one of the mistakes substance made.
Ahhā¦ thanks for thisā¦ i read something on the MaterialMaker discord channel and assumed it just is this wayā¦ (even if i know actually some little bits about the blender node systemā¦ silli me). Maybe iām also used to it now (and i really have to jump into Godotā¦ tinkering for no ggod sin 3.1. whateverā¦ and i didn;t want to wait for 4.0 )
Thinking about itā¦ one horried think is the disconnection happening sometimes if the node is moved in a certain way or the absence of the blender like feature of swapping inputs (but this sometimes also isnāt what i wantedā¦)
Yes if you drag a node input or output and release the mouse MM is given you a menu of possibilities to choose a connecting nodeā¦ blender does have this since 3.1 ??
A bit off topic but I was actually one of the early backers that bought a license of Armory engine before it went open source. But alas, it hit a stagnatory state pretty fast and got stuck in essentially a perpetual alpha while in the mean while godot was actively releasing stable builds and taking off.
Which is all to say based on some other past experience Iām predicting a slow death spiral. Sad as it may be.
Fair enough. I just tend to more think that spending time on UI/UX and getting base Material Maker in good shape should be done first, rather then adding a painting system into that same UI/UX and hence creating even more that needs fixing.
Then once all the code and systems are in a good place, look at adding painting features or combining with ArmorPaint or some sort of mix.
From what Iāve read, the current code needs a good overhaul/refractoring and moving to Godot 4 at some point. Along with extra nodes to better match Substance Designer. Adding a paint system now just makes that all more work and harder to do.
.
