Masterxilo

FindWindow is used like: HWND WINAPI FindWindow( __in_opt LPCTSTR lpClassName, __in_opt LPCTSTR lpWindowName ); so it should be HWND hWnd = FindWindow(0, title.c_str()); in the example you linked to.

And it's still being worked on:

How can it be so sure that they won't overlap? Does it do runtime checking (which would be pretty slow...)? You can promise the C/C++ compiler that two pointers won't overlap using __restrict: int fast(int * __restrict A, int * __restrict B) { *A = 5; *B = 7; // RESTRICT promises that A != B return *A + *B; // no stall; the compiler hangs onto // 5 and 7 in the registers. } int slow(int *A, int *B) { *A = 5; *B = 7; return *A + *B; // LHS stall: the compiler doesn't // know whether A == B, so it has to // reload both before the add }

Well in the original post you're asking about a cross platform live sync&display type of functionality like CE 3, but what has this got to do with having multiple monitors?

For this you could just stretch the editor window over all you screens. Or the new editor could allow to define more views that have separate camera positions and appear in their own windows.

You don't have to create a new font if you just want to use another color. You should always leave the color to white.

I think the wiki should be replaced by something using this website design and template (like the tutorials section). This would unify things. Btw. what happened to the planned "official" documentation which was supposed to become a replacement of the wiki?

I can use every library just fine. Even a glut32.lib compiled in 2001 so I don't think these need to be recompiled.

The lighting just lights what was rendered onto the buffer on SetBuffer(buffer); RenderWorld(); , it doesn't take any changes you make to the world after rendering it into account. You can easily make an object invisible yet still cast a shadow (meaning what is below won't be affted by the light making it cast a shadow), I think Josh already showed how to in your other thread. The idea is to use a material that doesn't render anything with it's normal "shader", yet has a "shadowshader" that renders the object solid.

There's a function to copy all the keys over to some other entity, but the physics and position attributes you have to copy yourself.

You can parent objects that are not in the same world to each other. I don't agree on the gravity center entity, this sounds like a very hack way to do things. What if your objects are below y = -50000? The rest of it really lists good means to achieve things in my eyes.

You'd have to create a different version of the font image for every resolution.

Well you should be able to configure code blocks to use the visual studio compiler, so yes. People might have done this already, google for it. But why not use visual studio directly?

It doesn't work for me either. I have to resave them in photoshop. But when I render to texture, I use lossless formats with pretty high resolution only anyways. You'll be glad you did this later, believe me.

Set the weapon's material to be always drawn on top of all. Or put it in another world and clear the depth buffer before rendering. Lot's of ways to achieve this.

How do you know? Maybe LoadScene accesses and changes some wrong memory offset but no one noticed so far?

I never found myself to need it with 3ds max.

What doesn't work? Does it not show up at all? Try using another cull mode...

@ wh1sp3r: Windows Phone 7 at least won't support native c++ code anymore at all...

Hi, funnyben. I don't think you'd even need/want to use any client side prediction with an rts. Wouldn't you want to use the reliable TCP protocol and only transmit the user interactions (button clicks)? Syncing all the unit positions is way too inefficient and inaccurate...

"Install Java JDK" is this also part of the steps required for compiling a C++ app?

Blend modes are implemented by OpenGL = the graphics driver and finally by the GPU itself. This is one of the few things even shader programmers don't have access to. There's no way to find out the color of the pixel that was already painted underneath the current pixel.

Mip-mapping of the textures (exchaning them for lower res ones) is the only thing that is done automatically.

I wouldn't use the CSG solids as ingame models/assets but rather as guidelines for the artist making the asset. The crysis example I posted was made into a much nicer bridge model for the final game. This would of course require the ability to save those solids as e.g. .obj models.

Far Cry 2, Crysis, STALKER, none of the newer engines use it. Call it CSG, call it placeholders, call it "solids", but even some of "the newer engines" use this. E.G. Crysis: http://doc.crymod.com/SandboxManual/frames.html?frmname=topic&frmfile=Solids.html This simplifies collaboration between level designers and artists a lot. Will the editor have some kind of plugin mechanism that would allow anyone to make something like this if you decide not to add it? Same for anything else one could wish.. Sure it does add polygons with a polygon rendering engine like opengl. Cube marching is used to "transcript" the volume data of the voxels into a polygon model. ZBrush doesn't use voxels, that's why you have to tesselate the mesh yourself with that sculpting app. The voxel volume used for terrains in crysis has a very low resolution and voxels per se are not too useful for non-aa flat surface modeling. We do need solids/csg. I agree. Originally I thought people would find that too restricting, but the general desire seems to be to have more of these pre-made things built in, so I won't worry about it. I agree with franck22000 and Laurens there. The most important thing (for me) is to be able to replace, extend, change, disable or just not use such built-in/pre-made features. You could use the gmf sdk for the writing part. Finally!