Josh

The Rocks and Stones Model Pack by Pure3D is now available here: http://leadwerks.com/werkspace/index.php?/page/products/_/assets/rocks-and-stones-pack-r1 You can also still get the Industrial Cargo Model Pack for the low introductory price of just $9.95 until Monday, September 6, before the price goes up: http://leadwerks.com/werkspace/index.php?/page/products/_/assets/industrial-cargo-model-pack-r8

I am definitely interested in supporting Mac ASAP, since iMacs now come with decent GPUs. Not so sure about Linux, but I am willing to provide it if there is a demand. I'm not worried about the timing of the consoles, because if I wait for the right time to start, it will never come. Even if The PS4 comes out 6 months after LE for PS3, the process of getting it on the PS3 will still be good learning for the PS4. I think they'll try to make the current generation last as long as possible because it takes such a long time to recover the money from a new console.

There is an undocumented command: GraphicsModeExists:Int(width:Int,height:Int,depth:Int,hertz:Int)

I am very interested in supporting XBox and PS3, and the source code is being written in a way such that support for those platforms can be added in. If your company is interested in contracting me to develop a port of the engine for those platforms, I would be interested. I do not have a dev license for either and cannot get one of my own yet.

Sorry, the frag shader should actually be "mesh_diffuse_bumpmap_specular_pom.frag". Parallax occlusion mapping (POM) is the raised effect I think you want. Plain parallax mapping is just sort of an enhanced bumpmap. I honestly don't even remember exactly what it does, and it should probably be removed.

I just want entities to be synced, files to be transferred, and let the user transmit whatever raw data they want. I could spend the next year learning raknet, and still wouldn't know anything. Actually, the only things I don't know how to do myself are NAT punch-through and TCP file transfer, so I'll probably just write the commands and then let the network programmer worry about the details of implementation.

I have not written a formal spec of the format because it is pretty self-explanatory, and I don't think there is a big need for people to edit the files. They usually just save them from the editor and load them in the engine. What would you specifically like to know?

Use this material: texture0="abstract::ceramics_02.dds" texture1="abstract::ceramics_02_nrm.dds" shader="abstract::mesh_diffuse_bumpmap_parallax.vert","abstract::mesh_diffuse_bumpmap_parallax.frag" The heightmap should be packed into the alpha channel of the diffuse texture. This eliminates one texture lookup, which will make it faster.

I am using the default GCC compiler Codeblocks selects when installed. I think I have MinGW 5.1.6 installed. When I enable the "Stop compiling after the first error" option in the build settings, I get this message from the compiler: cc1.exe: error: unrecognized command line option "-Wfatal-errors" Do I need to install another version of MinGW? Which versions will support this option?

Please post your texture and material files.

Yes. Built-in pathfinding will be supported.

Is there any way to do this?

I don't think there will be a lot of new features added to 2.x, but when a bug arises it will be fixed.

The faster debug compile makes me tend to think there are some optimizations I did not turn off. Normally I would expect the debug build to be slower. Perhaps the debug build turns off all optimization, because it assumes you are just testing?

Here's some results I found. I might not have found all the MSVC optimization options, but here's what I got to the best of my knowledge. It's surprising MSVC debug mode compiles faster than release? That might indicate additional optimization settings I missed. 23,000 lines GCC, no optimization - 0:12 MSVC, no optimization, release - 2:46 MSVC, no optimization, debug - 1:01

That card should be okay. The issue you describe reminds me of an ATI driver bug from a few years ago. If you can get those drivers updated, I think you will do okay.

That seems very odd because .lib and .a files are routinely distributed, and I thought MSVC and GCC were both widely used.

The beta will go on sale at a discounted rate when I feel the time is right.

What an interesting first week. I compiled a C++ program for Android, made a programming language, learned about iPhone development, and figured out a lot of C++ stuff I did not know. It's nice to see that a port to Android will work pretty much like I was hoping. I just write an abstract driver for every system, and have specific drivers that extends that class: class GraphicsDriver {} class GL4GraphicsDriver : public GraphicsDriver {} Then when you have something like a surface that is dependent on the graphics driver, you do this: surface = GetGraphicsDriver().CreateSurface() And a GL4Surface object is returned. Of course, this is just the internal workings, and you will only have to call CreateSurface(). The GL4Surface is an extension of the Surface class, the same way the GL4 graphics driver extends the base graphics driver class. I would like to get something running on my HTC Evo, but the details of an Android, XBox, or PS3 version aren't too important right now. What is important is to get the C++ core done, in a way that makes it easy to add support for more platforms in the future. So as planned, you'll get a .lib file for C++, a .dll you can use with any language, and multiple Lua scripts can be attached to any entity. A roadmap of the development plan can be viewed here. I hope to accomplish most of stages 1 and 2 myself, and then recruit additional coders for the last leg. I don't intend to write the networking code myself, either. I figure it will be easier to develop streamed terrain from the beginning, instead of trying to tack it on two years from now, so I will see what I can do about that. I've never seen an infinite streamed world with the density and complexity I want, but Leadwerks seems to do best when we do really aggressive development: http://leadwerks.com/werkspace/index.php?/page/roadmap What I like best about this process is the code I write now is ForeverCode: It's good for any platform, and it will last for the life of the engine, which will be a very long time. I really can't imagine ever writing a version 4, because version 3 is being designed to be perfect. At this point I would like to thank Roland Stralberg, Mika Heinonen, and Ed Upton for their feedback and wisdom.

Everything appears correct, but I recommend using POM instead of plain parallax. It is quite a lot more obvious of an effect.

I'm a bit confused because it seems GCC .lib files are not the same as MS .lib files. Am I to understand there is no standard static library format for C++? Can the MS compiler load .a or .o files?

Please, try not to argue with one another.

If you are confused, I would not worry about it. You will get a .lib file written in C++, and that's all that matters. I'm working with Visual Studio atm. I think I need to learn everything I can about C++.

Here are my imperatives for strings: 1. Ability to add two strings, i.e. Print( a+", "+b ) 2. Ability to insert objects into a map by string. If a mesh is loaded, I want to insert the mesh into a map and be able to retrieve it with the file path. 3. Maximum compatibility on all platforms. The way I see it, I have three options: 1. Use std:strings. 2. Add operators to const char* (if it's possible) 3. Make my own string class. Questions: 1. Is std:string safe to use? Will it work on PS3, Android, etc? 2. Is there a way to automatically convert a string I type into my own string object? i.e.: String s = "Hello"; Thanks.

Thanks for the help, it's working now.