Josh

No, I am only interested in existing projects in an advanced state of development.

You adjust your model to fit the bones. I admit the program is limited. Why don't people use that Mixamo thing? I looked at their site but couldn't make sense of their business model.

If you could have a "project" page on Leadwerks.com to show images, a video, a description, and a current demo of your game project, what would you put on there? I can think of a few that would be good, off the top of my head, including Combat Helo and Ravage. I'm not going to pursue this soon, but I like to ask about things in advance so that I have the information when I am ready to act.

Good question. I feel like it's convenient to have your videos, files, and a short explanation all on one page, as sort of a multimedia presentation. However, it should probably be broken up into multiple articles. I think more than three video embeds on a page is too much. The formatting can also be improved a bit, if you don't mind.

Thanks, "cute" is exactly the look I was going for.

The articles database makes a triumphant return with a beautiful new template: http://www.leadwerks...e/page/articles The front page shows the most recent articles. Clicking on a category on the right gives you the entire list of articles for that category: http://www.leadwerks...s/_/programming The new template I designed will help organize and present articles better. There's some great content written by the community, but I haven't provided a very clear specification for the articles section until now. Here are some things you can do to improve the appearance of your articles: You can separate your article's contents into pages with the "page" tag: If your article uses a YouTube video, you can submit up to three YouTube video IDs. These will be displayed in the upper-right hand corner of the first page of your article, as seen in this article: http://www.leadwerks...o-animation-r11 You can add a 160x120 JPG image to your article, and it will be displayed in the article lists. Do not add PNG images or your own icons, as they will make the page look inconsistent. Just use screenshots from the contents of your article: http://www.leadwerks...er_page=5&st=35 The articles section is not a place for code snippets or downloadable tutorials. These items can be posted in the Leadwerks Asset Store "Code" and "Tutorials" sections: http://www.leadwerks...werkspace/files Do not link to other articles, external images, or anything else in your tutorial. Many links eventually become broken with time. We want to minimize maintenance demands. I'm also running Google ads on the articles and video pages, mostly just as an experiment. I feel like these are sections that are okay to display an ad in, and hopefully the community won't find it intrusive: http://www.leadwerks...ine-preview-r16 If you haven't been to the videos section lately, check it out. It's looking pretty good: http://www.leadwerks...ace/page/videos Thank you in advance for your help keeping the community's content organized.

If you call registerabstractpath twice, the second file names will overwrite the first. It just stores the file name with no directory, and a link to the full path. You don't need to call it at all though, because it automatically gets called for the current directory when the DLL starts up.

The physics system does not necessarily get updated every frame. It gets updated 60 times per second, and entity matrices are interpolated between the last two frames of data calculated, to make everything smooth.

I don't think OpenGLES has depth textures in the spec. Some implementations may support it, but the mobile feature support is way ahead of their performance. The iPad 2 is the fastest thing out there, and there's no way it could do full shadow mapping. Maybe the next generation of mobile devices: http://toucharcade.c...-to-ios-devices

I apologize preemptively for the title. So I picked up a Samsung Galaxy Tab today, for testing...yeah, that's it, "testing"... I installed the USB driver for Windows, restarted my computer, and was able to start debugging Leadwerks3D on it with no trouble at all. However, the rendering showed black objects that ought not be black. At this point in the game, having such a basic issue is very scary to me. I tracked the problem down to the lighting uniforms being passed to the shader, but something didn't make sense. In GL Shading Languages (GLSL), you can have arrays of uniforms, like this: vec3 mediump lightdirection[4]; A uniform array like this should be accessible from the main program by setting individual elements in the array, or by setting the entire array at once with a pointer to eight floats. See for yourself: http://www.khronos.o...spec_2.0.25.pdf Page 35: It did not appear that the light color and direction was being sent correctly, and more detailed debugging confirmed this. After a few hours of testing, I finally determined that uniform array names on iOS have end with '[0]' (I didn't test any other element indexes, at least on the iOS simulator. My provisioning profile just expired, and that's a whole other story...), and just setting the uniform array at once with the name will not work. On the Samsung Galaxy Tab, the situation is reversed. My HTC Evo is the only device that follows the OpenGLES specification and allows either. Samsung Galaxy Tab: "lightdirection" iOS: "lightdirection[0]" Evo: "lightdirection" or "lightdirection[0]" (as it should) The solution? I'll just avoid using uniform arrays on any mobile devices. It's not hard to avoid now that I know what to look for, but it's hell figuring out which parts of the GL spec the driver authors ignored. Fortunately, you don't have to worry about any of that, so you can just make your games and be happy.

I wouldn't build this into the main editor, but a separate modeling tool would be nice.

The natives are getting restless...

I can't comment on this because I haven't seen a demo. The octree system skips top-level entities that are out of view range and doesn't even iterate through them. The number of entities out of view range, or in octree nodes outside the camera frustum, will have no effect on performance. You actually can use texture arrays, but I recommend against it because it's SM4 only and support for the feature varies. I would just build a separate surface per material you encounter in the grid.

I don't think z-sorting will help, but overlay=1 and depthmask=0 may help.

In the collision function, if the collision speed exceeds a limit, make the collision cause damage to the player. Then it will work for every impact. If the player goes below the water height in the scene, record the time he went under. Once the time past that exceeds something like 30 seconds, start subtracting health each frame.

The other language subforums are closed for new topics. They'll be merged into the main programming forum soon. Entities require manual freeing because they have too many linkages to ever be collected by the GC. I'm not a big fan of GC because of things like this.

First you need a character class with a "health" value, which by default would probably be 100. Then you need to decide what actions and events would cause the health value to decline. From there, it's easy to draw a rectangle with a width proportional to the health value.

You don't need to use an octree to on a small scale. The built-in one will work perfectly. If you were to try to get really detailed with the face rendering, you would experience slower performance on modern GPUs. Anything less than about 2000 triangles will render at about the same speed as 2000 triangles. Polygons don't matter so much as draw calls do. Hiding backfaces and trying to only draw each visible face would be much slower than just chucking a bigger surface at the GPU. Therefore I think you should look at what size grid gives you the biggest size mesh that can be rebuilt from voxels in real-time. The build stage is where your internal voxels and faces would be discarded. Your primary need is a routine that turns a 3D grid of voxel data into a single mesh. Rendering isn't a problem. Beyond that, you might then look at collapsing the eight chunks above that when any one of them changes, without rebuilding the others. Collapsing eight large surfaces into one would be pretty fast.

Please do not add tags in the thread title. Just set the variable to NULL and it will get cleaned up automagically. You can call collectgarbage() in Lua to force a cleanup after setting the variable: http://www.lua.org/manual/5.1/manual.html#pdf-collectgarbage

First you need to define exactly what you want. Make a mockup in photoshop or something. Then it's easy to figure out.

I added a C++ tag to this thread by editing the first post in the full editor, then moved it to the general programming forum. I don't recommend going through and tagging every single post, but I just wanted to make sure it worked.

If you keep increasing the numbers geometrically, you will eventually run into problems. 32*32 = 1024. 128*128 = 16384. That many entities is not possible. You will need to do something more advanced, like collapse them into a single mesh. I'm pretty sure Minecraft does something like this, probably breaking the world into 16x16 or 32x32 chunks, and reconstructing single meshes for each chunk based on what voxels are filled.

That's out of my range of expertise, and beyond the scope of a skin, but it's a good suggestion. At this point I am trying to make the site 100% consistent and clean, so I will just be focused on existing functionality. I would like to have something like you suggest, along with syntax highlighting for C# and Lua. That's actually one of my favorite aspects of the site. There are much fewer little lines and dividers everywhere, and they are subtle, so your eye kind of makes up the geometry itself. For example, in the header menu we used to have an image that looked like 3D plastic in the background. Now it's just two flat lines on the top and bottom, and it looks less "annoying" and more "real", to me. Yeah, I sort of agree with that. I am still figuring out how these work. I subscribed to some stuff so I can see what happens. Fixed. That was due to me making my own modifications, and viewing it in FireFox, which tends to hide errors.

Entities just get inserted into the smallest node that completely contains their AABB. Every node can store entities, not just the terminal ones. That way you don't have to worry about splitting up the scene, and it all works dynamically. That was one of those ah-ha moments for me when I figured that out.

It creates octree nodes down to a minimum level, as they are needed, and constantly reinserts entities into the structure any time they move. There's some pretty crazy stuff, like traversing up and down the hierarchy from a given node to find adjacent lights, and storing recursive counts of certain objects that reside in the sub-hierarchy of a node. For example, if I am looking for lights, and no lights occur in the sub-hierarchy of the current node, I can just skip it without checking all it's children. This is also where occlusion culling in Leadwerks3D takes place. What else do you want to know?