paramecij

I'm writing SM3.0 code in LE2,... but theoretically (not tested) you can do something like this: inside a shader if it's an OpenGL 2.0+ renderer with 1.20 or so shader model (which LE2 is) .. if supported by gfx card and driver of course according to this nvidia doc, you can, but check http://feedback.wild...EXT_gpu_shader4 on the support .. . But I would still check with Josh if this is paramount to you.

that alphatest shader just discards pxiels whose alpha is below 0.5, no transparency involved, just a cut-out, much like a color key.. but wouldn't all pure black alpha channel make your whole image totally transparent?

Yes, it should be universal, it should be int8, int16, int32, int64, float16, float64... in every programming language period. then algorithms and other code would be more easily portable. among other things.. and it's ONLY two types of number anyway, fixed point and floating point. But in real world it seems there is more types then there is known species on this planet...the type gets resolved at compile time anyway, so it's there just for our pretending...

I'm assuming you are trying to set an environment path variable... in Visual studio check the menu for tools / vs command line, then type: set AndroidNdkDir="c:\your path here" ... you can check if a variable is set by just typing: set name_of_variable, and if it exists it will print out it's value.. if you try this in a regular windows cmd the set command only works for that cmd session, global environment variables can be edited from control panel / system / advanced / environment variables

F'n yeah! And don't forget to brand them also, like LE3constfloatref... you know so they're easily distinguishable between other, lesser, more primitive types

Thanks klepto! Much appreciated!

Would any of you two fine gents be willing to share the up to date extended ,dll perhaps? I'd do this myself but don't have bmax, or would the demo version of bmax suffice ? (only 30day time limit is mentioned)

My attempt at procedural low poly trees (~500 tris each): I have a whole forest and every tree is unique (50 in this scene). The trees are breakable in 20~30 different parts. Now I need to figure out how to place the leaf textures so they look good...

Basicaly you can have only 1 material per surface (which is a group of triangles) and you can have multiple surfaces in a mesh, and multiple meshes in a model. So you can target and paint a single surface, the whole mesh(and all it's surfaces) or whole model (and all sub meshes it contains) both in code and script. So if your mesh consists of only 1 surface you can have only one material assigned to it. - you can easily check that out in the modelviewer. -edit: or you could create a shader that blends additional textures based on uv's or whatever you can think of, you can assign textures to a material and send shader values from script

Just to be clear - you are baking the normal maps from a high poly model to the low poly version, and are using SM2 only to further tweak those maps, right?

Great work as always.. how much effort would you say is needed to rig a model for facial expressions, like the one you have shown? (one that doesn't have optimal topology for facial animations) Also the spell effect is really nicely done, and the whole spell casting has a good look 'n feel to it, which is I think harder to achieve in a first person view. But i'm curious where you got those hands from, looking nice from what I could see in that dark..

I was kinda hoping you'd do that Anyway here's a quick preview mock-up (I haven't made any actual packs yet): Most of these textures are 2k or more. My texture "library" is not finished, so far I've only done ~25% from the pictures I took, some still need a little tweaking (color correction etc), while most don't have a normal map (yet).. But if they would be of use to you, I can upload them somewhere this weekend... Cheers

Looking good. If you need any additional vegetation textures or source photos, I have a huge growing library of them, that I'm willing to share.

I have the same issue with your test files (nvidia 8800, old drivers), but not with the model I converted myself! -Edit: Tried again but just replaced your .mat file and the issue was gone texture0="building1.dds" shader="abstract::mesh_diffuse.vert","abstract::mesh_diffuse.frag" shadowshader="abstract::mesh_shadow.vert"

I also recommend blender, and some cgtuts or youtube tutorials.. some additional links: http://www.makehuman.org/ - Easily generate base human models http://usa.autodesk.com/adsk/servlet/pc/item?id=13571257&siteID=123112 - Softimage Mod (free), model, rig, animate ... http://www.gimp.org/ - Make textures http://charles.hollemeersch.net/njob - Make normal maps http://www.turbosquid.com/Search/Index.cfm?keyword=game&sort_column=A5&sort_order=asc - Some free 'game friendly' models to be found http://www.thegamecreators.com/ - inexpensive animated characters and guns

The head model is rigged with bones, more on that here - http://www.leadwerks.com/werkspace/topic/4055-neutrino-game-engine/

Why not try and do both? Since you said it's for a test program, it would be good exercise. If I understand the "walk" correctly, you could get away with LE's rotation and translation commands only, if used on a pivot that you re-position when necessary. But by directly manipulating the matrix you could do it a bit faster. The pure physics approach probably wouldn't work that well in more complex scenarios (could be wrong), but for a cube on a plane you could start by manipulating the torque and velocity. Best from both worlds would be if you use a physics body for movement and collision with terrain/objects and you animate the mesh manually (that's basically how most games do their characters, just that in your case the character is a cube)

I doubt there's a tool out there that would work perfectly in your case. Better to make your own. By parsing and copying files that are referenced in the .sbx only won't work in cases where the models lua files load external assets, and if you load them in code (obviously). A better candidate I think would be the applog file, which shouldn't be hard to parse

I'd say Daimour's effort is definitely good enough. It doesn't matter if it's not 100% secure, as nothing ever can be, but as Flexman said, all that's important is that reasonable measure was taken. And Encrypting/password protecting assets is considered a reasonable measure - like locking the door to your house is considered a reasonable measure to keep unwanted people out. Just don't leave your key in the lock (hide password). Sure there are some people who can pick a lock, and the door can be forced open by just anyone who's determined enough, but then they can't use sleepwalking as an excuse for getting inside your house in court! As breaking in by picking a lock is a fully conscious act as is circumventing copy protections etc(which in itself is breaking the law), one can't simply argue they didn't know what they were doing. But more importantly, the effort and or risk is usually not worth the spoils to the majority and is enough to deter them from it (or they give up quick enough), which is probably the best you can do (with reasonable efforts).

I assume we will be able to use all that CSG goodness in real-time with code?

Sorry, we use comma for the decimal mark. I meant 8 kilometers and 192 meters.

Right. And it comes out exactly the same as 8,192 kilometers by 8,192 kilometers (8,192*8,192=67,1). ... so technically the features list could say more than 67 square kilometers, as they often like to do in marketing

I have a large collection of various rocks I made some time ago, but still haven't got around to texture them properly. I'm unsure on what's the best way to unwrap them, so there's no distortion, good use of uv space and that it's also easy to paint the texture in an app like photoshop. One of the (many) methods I tried was by marking uv seams on "bigger edges" (don't know exactly hot to put it), so the uv's were split into 3-6 chunks or islands. Then I flattened and relaxed each chunk and puzzled them together tightly into a square and made it fit the 0-1 space. From the uv standpoint I think the results look good, but the problem for me is painting the texture afterwards in gimp. I can easily texture each uv chunk, but making the seams not visible is near impossible or too time consuming. Is there something like mudbox, only cheaper, out there? I do have access to Zbrush at work, but is something like this possible in Zbrush, or maybe even in photoshop? (painting with images over the lowpoly model, seamlessly over uv chunks) Or some other, preferably more automated way of unwrapping (I have like 50 of them) and then painting these stones? A couple of the lowpoly rocks (150polys), with normal map baked from highpoly version, and badly done diffuse texture

I agree with gamecreator. Looks great, very fitting for nose-art, and also has some background to it, what more could you want ?