Shop Forum More Submit  Join Login
About Deviant vasilnatalieMale/United States Recent Activity
Deviant for 6 Years
Needs Core Membership
Statistics 344 Deviations 1,490 Comments 13,277 Pageviews

Newest Deviations

Guess I'm making a new one by vasilnatalie Guess I'm making a new one :iconvasilnatalie:vasilnatalie 3 1 Reworking face by vasilnatalie Reworking face :iconvasilnatalie:vasilnatalie 4 4 Spline IK with tilt for Blender by vasilnatalie Spline IK with tilt for Blender :iconvasilnatalie:vasilnatalie 3 5 Test and demo of partial physics in Blender by vasilnatalie Test and demo of partial physics in Blender :iconvasilnatalie:vasilnatalie 3 0 Remapping textures from alternate UV in Blender by vasilnatalie Remapping textures from alternate UV in Blender :iconvasilnatalie:vasilnatalie 9 3 Herald of New Gods by vasilnatalie Herald of New Gods :iconvasilnatalie:vasilnatalie 0 2 Download NPR Koi .blend for Blender by vasilnatalie Download NPR Koi .blend for Blender :iconvasilnatalie:vasilnatalie 5 0 Foolish Carp by vasilnatalie Foolish Carp :iconvasilnatalie:vasilnatalie 3 0 Koi by vasilnatalie Koi :iconvasilnatalie:vasilnatalie 5 9 Textured fish by vasilnatalie Textured fish :iconvasilnatalie:vasilnatalie 1 0 Koi by vasilnatalie Koi :iconvasilnatalie:vasilnatalie 4 10 Proof of concept: mesh-based angle limits by vasilnatalie Proof of concept: mesh-based angle limits :iconvasilnatalie:vasilnatalie 2 0 Advanced Math Nodes for Blender Cycles Materials by vasilnatalie Advanced Math Nodes for Blender Cycles Materials :iconvasilnatalie:vasilnatalie 4 2 Blue Lotus by vasilnatalie Blue Lotus :iconvasilnatalie:vasilnatalie 3 0 Wave by vasilnatalie Wave :iconvasilnatalie:vasilnatalie 5 0 Scale UV project by vasilnatalie Scale UV project :iconvasilnatalie:vasilnatalie 2 0


Bonsai by SilentLadyGTA Bonsai :iconsilentladygta:SilentLadyGTA 24 19 Dexters Laboratory by LindaRoze Dexters Laboratory :iconlindaroze:LindaRoze 508 47 .:MMD Newcomer:. Kion Tinkerwell by TheKagamineTwins1234 .:MMD Newcomer:. Kion Tinkerwell :iconthekagaminetwins1234:TheKagamineTwins1234 1 2 Ichimei by tatagiba Ichimei :icontatagiba:tatagiba 57 8 Crow by COOLZONE17500 Crow :iconcoolzone17500:COOLZONE17500 16 4 [City View] by Rainy-Blaze-74 [City View] :iconrainy-blaze-74:Rainy-Blaze-74 169 34 GTA V by SilentLadyGTA GTA V :iconsilentladygta:SilentLadyGTA 15 7 GTA V by SilentLadyGTA GTA V :iconsilentladygta:SilentLadyGTA 20 5 A bicyclette by SilentLadyGTA A bicyclette :iconsilentladygta:SilentLadyGTA 25 29 A Rare Photo of the Artist, Hard at Work. by ArtificialFleshling A Rare Photo of the Artist, Hard at Work. :iconartificialfleshling:ArtificialFleshling 4 7 [MMD]''Black Raven'': H.W (WIP) by AbyssLeo [MMD]''Black Raven'': H.W (WIP) :iconabyssleo:AbyssLeo 28 6 Ethereal Messenger by 3Dhealing Ethereal Messenger :icon3dhealing:3Dhealing 26 3 Materials: Edibles by silveralv Materials: Edibles :iconsilveralv:silveralv 1 0 [MMD DL] SOULCALIBUR : Ivy Weapon (Download) by LGMODS [MMD DL] SOULCALIBUR : Ivy Weapon (Download) :iconlgmods:LGMODS 12 2 (MMD) Metabee - DL by SchattenKitsune666 (MMD) Metabee - DL :iconschattenkitsune666:SchattenKitsune666 6 0 [OCxCoco] La Llorona + process by happy-eevee [OCxCoco] La Llorona + process :iconhappy-eevee:happy-eevee 32 1


Guess I'm making a new one
Thought I was bored enough with rigging I should make some hair (bangs are an imitation of Uzuki, because I know those are good angles for physics, but I probably ought to scale them down some.)  Then I got tired of looking at a naked body so I started playing with clothes.  I think she'd look pretty good in some elevator mary janes, finally time to make some girly shoes.

I was just playing around but stuff went faster, maybe because I didn't care as much.  Hair curves were just two point curves, mostly adjusted after conversion to mesh.  Topo on that dress is terrible, will have to remake I guess.
Reworking face
Not being very efficient here, but I'm trying to take the next step to improve the model.  This is the Azusa face although I have made some edits.  Hair is placeholder hair (imported from Uzuki.)

I've been trying to mess with bones on this face.  I like the direction I'm going with the rigging on it, but it's frustrating and I lose interest quickly, I like the potential deforms I'm getting but haven't a clue how to create good controls for them.  I decided to just rework a few materials and I liked what I saw, so time for a render.
Spline IK with tilt for Blender
Demonstration of creation of spline IK bones that respect curve tilt.  This is more cumbersome than it ought to be.  If there's a better way, I'd love to hear about it.

In searching online, I couldn't find a solution for this, so when I made this, I thought I ought to share.

Spline IK structures are nice for a lot of things.  With proper use of handles, they're can have very elegant controls.  I'm using them to define parts of models like eyelids and mouths.  Adding tilt to these structures gives an additional dimension of control.
Rigging.  Experimenting with spline IK eyes.  It's good, very smooth, very powerful, but I need to figure out how to tune some accessory structures (like the cheek needs to be a part of smile-eyes motion.)  Would like to do this with the mouth as well.
Have been a little lazy for a few days-- screwing around with things, not really getting anything done.  Today, I work on improving my Azusa base for a new character.  It's nice to have a base that's mine to work on, where I know the verts, I know how everything works.  So far: tried to improve the lips and general facial structure, cleaned up tris & >5 poles, altered proportions especially legs (long legs were *long*) and breasts, improved eyelashes; working on abdomen right now.  I've got Uzuki hair in the file so I have some reference points.  I like that hair, I ought to make my own fully original version of it.
Test and demo of partial physics in Blender
So I want to be able to do some physics stuff.  But I need continuous meshes.  You can't create an object that's half-affected by physics and half-not (not generally).  So how do you join them?

This took a while for me to figure out, and this isn't quite perfect yet (although it should be good enough.)

In the demo, the left half of the sphere and the right half of the sphere are separate objects.  The right half has physics, the left half doesn't.  I have a few hidden objects as well: a full sphere for copying normals, and two separate objects composed only of edge loops for shrinkwrapping verts.

So the first thing is that you need to create continuous normals, even though you have two different meshes.  This is done with a data transfer modifier from a version of the object that's whole (also present in this scene, but hidden).  You transfer the normals from the continuous object onto the partial objects.  I think you can go ahead and write this modifier if you want, it becomes custom normal data, but you don't have to write it.  Just make sure, if you're keeping it as a modifier, that the source of your normals is moving with the same armature, so that you're copying from the right place.  You need autosmooth enabled on your mesh to use custom normals.  Change the autosmooth threshold unless you want sharp edges at >=30 degrees.

With cloth physics, that's about all you have to do, because pinning is 100% reliable.  Create a new vertex group, assign your edge verts to it, and designate that as a pinning group for your cloth physics.  This pinned outer ring should have normal weights so it follows your armature, but your cloth group probably shouldn't (don't know if it matters though.)

Now, with rigid body physics or soft body physics, that's not enough, because rigid bodies have no mechanism for locking parts of them, and soft body goal weight is not 100% reliable.  That's why I'm shrinkwrapping the edge of the physics mesh back onto the object.  Note that I've created a new object, just the outer ring of vertices, so that even when the physics rebounds inwards, it shrinks to the right part.

I'm also shrinkingwrapping the second line of verts here (to a duplicated and static identical line).  I don't know if that's necessary.  I thought it was in order to keep good normals on that stable outer ring of faces.  Note that I've created new vertex groups just for the purpose of limiting these modifiers to only the verts I want affected.

The remaining issue here is that the shrinkwrap modifier doesn't always snap the edge to the correct vertex.  You could create alternating vertex groups and double up your shrinkwrap modifier (or divide it up even further than that) if you wanted, but that's a pain, and not often necessary.

This particular example could be improved by smoothing out the vertex group indicating the goal weights, and by using different topology for the sphere-- splitting a UV sphere through the poles is leading to weird deformation from the physics.

There are also ways to do this with bones that just reference the physics-- shrinkwrapping, damped tracking various vertex groups associated with the mesh, etc.  I would say, in general, doing it with bones tends to be smoother, but more complicated, and with less detail than using physics directly.
Remapping textures from alternate UV in Blender
This is a technique that should be useful to a lot of people, including people editing models for MMD.  It's not too complicated for people who only know a little bit about Blender.  I've found myself offering this as a solution a lot on Reddit BlenderHelp, and people seem scared of it, so I thought I ought to demo it so I have a place to link people, since I didn't learn it from any particular existing tutorial.

There are a lot of uses for this.  I started modelling in order to see if I could get better UV mappings for tiled normal map application to MMD models.  I didn't know enough to do it back then, but this is perfect for that-- you can create a whole new UV mapping without losing your existing texture.  It's also useful for combining textures, as here, where I use it to cross a seam, or for creating bilateral textures from models with only half of a UV map (mirrored across the other side.)  Or for baking textures from projection maps, which are easy to create textures for, onto unwrapped maps, which minimize distortion.  On my koi model, I used this to bake multiple bump maps onto a single texture so that I could smear the edges of the bump map to eliminate seaming artifacts.

You can get more advanced by playing with more complicated node setups, and I think this technique might be a good introduction to playing with nodes.

When baking textures, back everything up beforehand.  It's easy for Blender to think it's smarter than you and bake a whole bunch of textures-- improperly-- that you didn't want baked.  If you save all images, instead of saving just the image you baked, you might lose a lot of work.

This doesn't bake alpha.  Haven't had to do that, but if I did, I might bake a separate alpha texture and apply it as a mask in GIMP.  Maybe mixing with a transparency shader in Blender's nodes would do it, I don't know.

Don't worry about color/non-color too much.  On default settings, Blender does gamma/deganma on your textures, which on an emit texture, has no effect.  That's why I'm using "color data" in my texture node.  If I used non-color data, it would only do half of that and I'd end up with a brighter texture than my input.  If you're concerned, just compare the textures, it will be obvious enough on anything with a reasonable range of levels.  (On a near-white tex, might not be obvious.)

Hope this helps somebody.  As always, I can explain more if needed.
Herald of New Gods
Starting work on a new character.  I might not know what I want very well.  I think I'm going to have to spend some time looking at google images.
Download NPR Koi .blend for Blender…

NPR-styled koi (and wave and flowers) I created.  Rigged, incl. facial bones.  Materials and textures.  For Cycles.

Includes a text block with a bit of info.

Release CC-Sharealike 4.0.  I don't care about attribution, I'm just sick of people putting rules on things they made out of free sources.

Uploading to blendswap as well, first time it gets updated, the link gets replaced with a blendswap link so I don't have to maintain the file in multiple locations.
Foolish Carp
Antonin stopped sanding but did not look up.  He voice was loud and clear though he faced away.  "You're not incorrect."  His apron and hands were gray with the dust of pine and pumice.

Antonin did not speak and did not return to his work, but Marten did not lower his head back to his own work, a thick strip of leather and an awl ever-dull.  Marten was accustomed to long pauses from his grandfather.  Antonin spoke the same way that he ate.

With a long fingernail Antonin gouged sawdust from a crevice and coughed, quietly.  Then he continued:

"There are river carp that arrive at the great falls that mark the boundary between East and West.  It is not uncommon.  Some men say that the carp want to see the white water lilies that wilt at a touch and cannot be grown nor shipped.  But I do not myself know.

"Some carp try to climb the falls.  This is foolish.  All the waters of the West rush many hundreds of feet almost vertically over jagged rocks.  It is impossible to climb.

"Some carp see the placid locks and try to shadow a tradeship.  This is even more foolish: they enter a cage.  They will be caught.  The ugly will be eaten.  The beautiful will be captured and bred.

"Some carp see the falls and turn back.  There are some men that say these are the wisest.

"There are other men that say that these carp are the most foolish."

Antonin wiped his stone on his apron and resumed sanding.
Just about final.  Eyebrow bones are cool :)  Have to integrate the other stuff into the scene and do some better lighting.
Arrgh.  I spent all this time making these nice blurry material index maps, and then i spent all this time making these nice bump maps for the fins, and then I turned around and Blender said, "Oh, you mean you didn't want me to revert those to the empty originals?"

No matter.  I do things twice (or more) so often....  Getting a little handier with the texture painting keys that make it not quite as huge of a pain in the butt.

The koi's starting to look good, at least.  Or, he was, until I lost my textures.  I guess it's a he?  Maybe I need a name.
Was thinking I'd do scales with a curve shrinkwrapped to the body, and then an array of scales running through the curve.  I'll see though.
Making a koi is kind of slow-going.  I don't really know what a koi looks like, and can't find any orthogonal pics, just artsy pics.  But it's going, I'll get there eventually.
Proof of concept: mesh-based angle limits…

Proof-of-concept and demonstration of using meshes to limit bone angles in 1, 2, and 3 dimensions.  In the file I am use booleans on the angle-limiting meshes in order to create an improved floor constraint.

Not sure how practical this is, it is complicated and requires dividing an armature into multiple objects in order to get around dependency issues in Blender.

Although mesh-based angle limits are useful for eliminating twitch that can occur with Limit Angle constraints, in this file, it's easy to get twitch because the boolean can completely destroy the angle-limiting mesh.
Advanced Math Nodes for Blender Cycles Materials…

Collection of several "building block" math nodes that I find useful.

Contains VectorMultiply, Length, RectangularToPolay (and vice versa), ReflectVector, HalfVector, 4x4 Transform, Gamma and Degamma, Invert Power, Schlick Fresnel (by IOR or reflection-at-incident), Remap Value, arbitrary Clamp including threshold, random Dissolve, VectorTile (like tiling UV), VectorMirror (for mirroring UV).

Models provided are simply my test models to make sure it was all working, that I left in the scene for demonstration purposes.

Uploading to Blendswap as well.  Released CC0.
Realized what UV Project modifier is good for: sticking in the middle or the end of a modifier stack.  Like for an arrayed object.  Which my wave is.

Organizing my learning like this has been useful.  First day, I discovered a Blender bug in the data transfer modifier (reported.)  Today, I revisited the UV Warp modifier, which I've used before (change UV before a displacement modifier, to create something drip-like).  Didn't expect to learn anything, but I did; I realized how it could be used to create decent UV morphs for mmd_tools (reported that too.)

There are a few projects I'm knee deep in.  One is my carp tattoo, where I'm still working on the non-carp components (you've seen first draft waves + lotus.)  Second is a demonstration of an improved floor constraint, using bones with mesh-limited angles and booleans on those meshes to establish floors.  Third, started today, is a collection of math node groups based on structures I've made a dozen times (remap value, vector length, vector scale, gamma correct, clamp to, etc) and some other structures that would be too painful to create dozens of times, but created once, might be useful (rectangular-to-polar, 4x4 transform, reflection vector, Schlick Fresnel, stuff like that.)

I once read a mathematician write about what it's like at parties for them, that they always hear, "I'll get you to do my taxes then!"  Of course, no.  Being a mathematician doesn't mean you don't screw up when you add numbers.  It means that you think in mathematical terms.  I'm like that too.  Not a mathematician, not nearly enough education, but I like math a lot.  I think in terms of math.  That doesn't mean you should ever have me try to solve for x :)  So I have to do a lot of work on these math nodes, and I have to be sure to check my work, which isn't the easiest thing in the world when I don't have numeric output.  (Rectangular-to-polar is a good example of that.  I've written a rectangular-to-polar function about a hundred times, and I don't believe I ever got it right the first time.)
Blowing my Blendswap downloads on a bunch of CC0 Cycles materials.  When it comes to materials, I don't even bother looking at something unless it's CC0/public domain-- it's not worth managing license info for mats.… : Good looking metals, wide enough variety, fairly easily customizable color.  Good pack.… : A huge pack, but without any preview.  I started previewing until I saw that a material used UV coordinates when I wasn't using a UV mapped mesh and said, screw it.  Bad render+preview settings in the file makes it even harder to see if there are any good mats.  Missing textures.  The mats I saw were nothing special.  Not worth the download IMO.… : Beautiful procedural leather, incl. good-looking procedural normals.  Nodes setup looks messy, and you probably need to edit it (for things like scale), which might be a pain.  Very worthwhile though.… : A single good looking wood mat.  But it uses a colored texture, which both limits recoloring and causes problems at seams.  So-so download.… : Doh!  The first time I looked at this I did something stupid, maybe I was just looking in solid view.  This is actually a very nice collection of diverse fabrics with procedural bump.  Good download.… : A somewhat diverse group of well-made mats: ceramic, cast-iron, car paint, two curtain fabrics, brushed metal, some other boring metals/plastics.  Normals and alpha on them where appropriate.  Includes a single texture, a good leather bump, but I'm not sure what uses it, since there's no leather mat; other stuff still uses UV coords.  Doesn't have a billion recolored copies, which is nice.  Poor preview settings in the file, single object makes it more difficult to preview.  Good download.… : Variety of "coated glossy" shaders that are supposed to look like plastic, metal, or ceramic, but which all look fairly ceramic to me.  But it's a good ceramic, with a wide variety.  Comes with preview jpgs that are nice.  Nodes are intended to be edited but aren't very friendly to editing (a lot of "coated glossy group.008" kind of stuff.)  Decent download.… : Procedural wood, with good grain, no use of vertex UV.  No normal or bump.  Colors modifiable but not as easily as they should be.  No clearcoat.  Decent download.… : Shiny hexagons.  I was hoping this was from generated textures coords, but no, it's from vertex UV, so vulnerable to seam issues.  Alas.  Node group has problems with tangents and isn't anything interesting, but at least the included hexagonal normal map tiles well.  Poor download.… : Wicker chair mat using colored textures, vertex uv, and nothing special in the nodes.  Would save you a minute if you want to use this exact material, but won't be useful for anything else.  Poor download.… : Three well-organized and not overthought shader groups: a glossy with fresnel, a metal, and a glass.  All look good (but the metals at the top looked better than this metal.)  All intended for customization and well-designed to that end.  Includes some good textures, but the node groups themselves don't are about vertex UV.  Good download.… : Procedural gingham with procedural bump.  The bump isn't very good.  Good enough if this is exactly the mat you want, otherwise, probably not worth the download.… : Very nice nacre (pearl) shader.  Unfortunately doesn't preview well (needs a lot of samples before it stops looking like multichromatic noise) which makes it hard to tune, and the default settings are a little overboard.  But this is a tough material to get right, that you're probably going to need only once, that's a focal point-- good download.… : A collection of procedural camouflages that differ only in color.  It's a simple musgrave from generated coords that makes the pattern.  Overall node layout is weird and kind of dumb and difficult to expand.  But if you need a CC0 camo in a hurry, hey, this is that.  Decent download.… : Another collection of camos that differ only by color.  (Guys, we can choose colors.)  Better nodes, organized, but not with English names (German?  Not sure.)  Don't like that the colors are managed by color ramps, which make it hard to edit.  Decent download.… : Aerogel material (soft edges).  I downloaded this because I haven't been able to get similar looks from volumetric shaders without emission, which needs to be tuned to scene lighting since it's kind of a cheat, and I was hoping this didn't use emission-- but nope, it uses emission.  Still, decent material, although there are drawbacks to it.… : A well organized, better-than-default velvet shader.  Velvet shaders don't bake well, and so I tried a bake test on this.  Which took forever.  I cancelled it half way through.  I don't think it was going to bake well.  But, nice velvet shader.… : This is actually CC-BY but I wanted to check out the effect.  It's faking volume absorption through the use of ray length.  If you don't mind using CC-BY materials, these would make good shaders for candy.  Even though these don't use volume, they use glass shaders, so there's probably not much of a performance improvement when using these at defaults.


United States
Playing with MMD. Not much of an artist, but I enjoy myself.

Have written a few tutorials for LearnMMD at… ; I will probably write a few more in the future. I also have several tutorials here, although sometimes, that's for a broad definition of "tutorial."

Still tend to the R-18 side of things. Helps me maintain my interest.

Always happy to help people learn anything they can. If you're having problems with an MMD model, and you don't know why, please feel free to send me a note. It's usually a good idea to send a link to the bugged model as well. (But I don't just fix things. I don't believe in that. Instead, I'll help you understand what's wrong, and how to fix it yourself.)

Big believer that we all benefit from each others' art and tools, which is why I think it's so important to share freely. I strongly encourage all other artists to do the same. The artists' rights that I support are the rights to freely transform that which cannot be used up, stolen, or destroyed. Art is fire and fire cannot be owned.


Add a Comment:
STRRRWBY Featured By Owner Apr 4, 2018
Thanks for the fav:)
vasilnatalie Featured By Owner Apr 4, 2018
Thanks for making cool art!
STRRRWBY Featured By Owner Apr 4, 2018
Aw, you spoil me :)
annamaeblythe Featured By Owner Jan 25, 2018
I was curious if you ever thought about 3D printers. It just randomly hit me that with 3D printing, you could print out your models in whatever pose you wanted to. Not sure if you'd want to set up an Etsy store selling them, but it would be fun :P
vasilnatalie Featured By Owner Jan 25, 2018
It would definitely be fun.  I used to paint war-game miniatures (more painting than playing) and was always excited about 3D printing possibilities.  But I don't have the money.
annamaeblythe Featured By Owner Jan 26, 2018
Keep your eyes peeled - 3D printers are getting cheaper and cheaper every day :D You can already find some for under $200 dollars if you're willing to get a kit and assemble it yourself.
vasilnatalie Featured By Owner Jan 27, 2018
I will, thanks!  (Not even $200 to spare, especially with recent computer troubles.)
annamaeblythe Featured By Owner Dec 16, 2017
I know you likely wouldn't care to make use of this . . . but it's free and open source and pretty darn cool!…
vasilnatalie Featured By Owner Dec 20, 2017
I have taken a look at that, and it is cool.  One of these days I'll look more at these kind of procedurally generated characters (MakeHuman is another popular add-on with a similar purpose.)
annamaeblythe Featured By Owner Dec 11, 2017
I just found this video :…

I thought it was SUPER COOL how many features it had and how easy it made it all look!
Add a Comment: