Quarrel Kukulcan

http://blog.machinimatrix.org/wp-content/uploads/2013/11/fitted_mesh-270.blend Make sure to enable the extra "Keep Bind Info" setting when you export. The armature in this file has extra data that the normal SL export preset doesn't send. NOTE: This is NOT full Bento. It's the original skeleton plus Fitted Mesh collision volume bones. As far as I can tell a fully Bento version of this armature is not freely available anywhere.

There is probably something wrong with your armature. I just did some test imports. I was still able to check "Include skin weights" even if my mesh contained a vertex with more than 4 weights, a vertex with 0 weights, or a vertex weighted to a nonexistant bone (although obviously the object was malformed in SL). If I misspelled a bone name in the armature, though, I lost that checkbox.

I just spent several minutes looking for something online that was brief, current, and not Avastar-related. And failed. Clearly I need to bookmark better. Generally speaking, you want to: select just the clothing go into File -> Export, choose Collada (.dae) choose the "SL + Open Sim (rigged)" settings preset (it's hidden under an options gear in the latest Blender) -- and double-check that "Selection Only" and "include Armatures" are on export The uploading half of it is covered pretty well at Oh, and you can't export clothes that work right with Fitted Mesh weighting unless you either use Avastar or you copy weights from a special base file (like the ones at http://wiki.secondlife.com/wiki/Mesh/Rigging_Fitted_Mesh) and click on the extra "Keep Bind Info" option during export.

The UV template goes above and/or below all your working layers, at whatever level of transparency works best for you to guide you as you paint your texture. You don't include it in your final export. The AO map and any other shadow map go above all your work at 100% opacity and using the Multiply image blending method. (I can't tell you where that is on your software, but I know Photoshop, GIMP, and Paint.NET all support it.) If you have something like a highlights map or layer, that goes above your work too, probably with the Screen blend mode (though you should also experiment with Lighten and Dodge).

I can find no text to this effect in the current TOS. The closest I see is that anything you create that gets seen in a public area can be freely screenshotted and video recorded by LL and all other players, who can then do anything they want with that media.

There is no database, but many creators provide a template for their own product and either bundle it as a texture with the item in SL or make it available for download from a website. Also, anything that uses Bakes on Mesh follows SL's own ages-old avatar UV template.

If SL has to enlarge your object so no single dimension is less than 1 cm long, it'll magnify any misalignment between parts.

You've uploaded all six textures to SL? Then it sounds like a simple matter of the Select Face option in the edit panel. (I'm assuming skyboxes are like any other prim.) https://www.instructables.com/id/How-to-apply-textures-on-individual-object-surface/

Animations get set to "loop" or "not-loop" when they are initially uploaded, and then they're locked in forever. If you need a non-looping animation to run repeatedly, you need to use a script. Something like default { state_entry() { llRequestPermissions(llGetOwner(), PERMISSION_TRIGGER_ANIMATION); llSetTimerEvent(X); // change "X" to the animation duration in seconds, plus a little extra } timer() { llStartAnimation("animation_name"); } } on any avatar or HUD attachment. You'll also need to put a copy of the animation into that object's contents.

What do animators do when they create toe-wiggling animations? Do they rotate the foot bone? The toe? Both? Are such animations so scarce there is no convention?

What is the purpose of the foot bone, given that legacy animations drive the whole foot by the ankle bone? It's especially confusing that the sample Bento skeleton uses the foot bone to drive the toes, and furthermore places the toe bone past the end of the foot and weights no vertices to it to it. Is there any consistency to how existing animations and common mesh bodies use these bones?

I'm pretty sure no one is going to openly help you violate copyright.

It depends how you add vertices. If you extrude or duplicate, each new vertex gets the same weights as the one it was copied from. If you subdivide (including poking and inserting faces), Blender averages new vertex weights the way you mention. If you add a whole new cube or other basic Blender primitive, it will have no weights.

If you enable the Direct and/or Indirect option from the baking panel, yes -- and you will likely have to do a little experimentation with exactly how strong to make your lights. If you do a diffuse-only bake (not Combined) and choose only Color, lighting is ignored. Lighting is also irrelevant to Normal and Ambient Occlusion bakes.

That method -- creating a 1024 square by making a 2x2 quilt out of 4 identical copies of a seamless 512 square -- is exactly what I had in mind, though obviously I used words that didn't convey it right. I'm curious -- what do you think I mean? It's not a pure advantage, but it might be a desirable tradeoff -- less of a problem than the blurriness you get from enlarging. I don't know what you're making or how important it is for your pattern to be an exact size on the object. But it's something I've run into personally. I did make a shirt texture from a 512 AO and a seamless diffuse pattern that was something weird like 340 and found it looked better if I quilted & cropped back to 512 than if I enlarged. I'd love to see a picture of this. I have no idea how that's happening.

I was going to say the first problem looks like a rotated model. If one arm is fine but the other is wild, it sounds like you mirrored your mesh but forgot to mirror the weight names.

Blend modes are a powerful tool. There are lots of ways to combine two images beyond the common, intuitive idea of "just average their colors" (which is what Normal mode does -- it averages (opacity)% of the top color with (100-opacity)% of what's underneath). (If you don't see "Multiply" mode stressed in manuals and tutorials, are they tutorials about using AOs in SL or are they tutorials on your editing program in general? The latter won't be able to tell you which mode to use because they don't know what you're trying to do.) With an AO map, you want the dark parts to darken your diffuse map and the white parts to do nothing. So you don't want Normal mode. Normal will mix in black where the AO is black but will also mix in white where the AO is white. Multiply mode uses different math. It treats the top image colors as a fractional scaling factor from 0 (black) to 1 (white). Since 0 times anything is 0 and 1 times anything is the thing, you get the effect you want: black makes things black and white does nothing (and grays are somewhere in between). Multiply mode has nothing to do with sharpening textures. That's only about making your AO map stop washing out the white areas. Enlarging an image to a higher resolution without it looking blurry is an extremely hard and very widespread problem that is at least fifty years old. It's impossible to solve perfectly, and even how to solve it "well" varies depending on what the original image looks like, how big it is, and how much bigger you're making it. You are not alone in being stymied by this. Don't be surprised if you have to experiment with different variations on enlargement method and post-enlargement filtering -- all of which depend on your image editing program -- even if you get some hints on what to try first from other creators who are familiar with your tools. Enlarging a seamless texture will have all the same blurriness problems as enlarging other textures, plus possibly also ruin the perfect seamlessness of the edges. What can work is, instead of upscaling it, putting 4 duplicates of it into the corners. There will be no blurring, but you will end up with your pattern half its original size on your finished SL object.

Paint.net does support the Multiply layer blending mode (according to https://www.getpaint.net/doc/latest/BlendModes.html). You want to use that mode, not Normal, for the AO layer, and have it above your texture. Otherwise your paint program will mix the AO's white areas with your texture and wash out your colors. (Chic points out you can put your texture on top and put that in Multiply mode, and that might be mathematically equivalent but that's a weird way to think about it.)

Posting a screenshot of your UV map in Blender would be a big help, especially with the expected diffuse texture also present. Do you only have one UV map per material? Have the object's texture scaling, rotation or offset factors been accidentally changed in SL? How many different materials are in your object? If it's more than one, are you setting each one individually to the correct diffuse texture in SL via the "Select Face" radio button?

Also, "Select Face" (near the top of that panel), followed by clicking a particular material surface, if you want to set distinct diffuse/normal/specular images (or other settings) differently on different materials.

Mirrors reverse things. In Blender's case, that includes normals. (It doesn't mirror weights as part of mirroring mesh components, either. You have to mirror the weights separately.) Just remember to check and reverse them yourself and you'll be fine.

You definitely want "up" slopes (away from the surface and toward the camera) to be green on top and red on the right. Look at this experiment. The sun is directly overhead. The top right sphere (relative to the ^ arrow) looks properly raised. The bottom left is properly sunk. The other two have the highlight 90° off from a realistic direction.

Shouldn't you have reversed the green only? Normal maps in SL should look like they're lit with green from the top and red from the right.

Changing some body shape sliders causes collision bones to relocate and scale (but not rotate -- rotating is actually irrelevant to this issue). If your custom skeleton also contains location changes on the collision bones, SL's shifts could interfere with yours, or override yours, or cause disproportionate child movement, or...

Was this a commission for Sir Mix-a-Lot?