Back to Voxlap

[Scott_AW]

After Ken found this -> http://amsterdam1.plunder.com/x/$h2EYHiiK2_kqxOXrr8CkJP64gthkijTT/21997/?/vs6sp5.exe

I was able to compile voxlap with VC++ 6.0 with the processor pack.

Which is great because I've wanted to work with voxlap for awhile now, but didn't quite have the means to compile it properly.

So, in what time I can muster, I've begun to play around with the code and commands.  That and grab a few docs on C++ to get a better grasp.  Luckily a lot of what I've learned over the years in programming seems to be helping me understand a bit more.

I'll be working on to little projects regarding voxlap, one based of the game.c just to see what I can do with it, and another which will be based on my dungeon crawler I've been working on in Evaldraw.

One nice thing is I'll be able to have real lights and fog.

However one thing I'm a little torn on is my current method of creating the map.

Currently I have a pretty nice tile system, it allows for huge maps but currently only one story and none of the fancy stuff Game.c has.  One idea I've been playing around with but have yet to commit to is looking into the VXL format and possibly making my map maker create VXL maps out of individual pieces.  I'll have to convert the pieces to VOX or KVX format though, as I'm currently using KV6's to build everything.

Using a VXL map could make things easier to handle and improve at the cost of map size.  I can always work at making the evaldraw version support multiple floors, allowing for even larger maps.

I'll still be working on the Evaldraw version, right now I'm just testing/researching what I can do with voxlap.

Any thoughts on this?  Should I just stick with speed and large maps or smaller maps that can be altered to the voxel level?

[Redeemer]

I don't know why you're all torn up over your map formats. You can keep the tile system you use in Evaldraw. What you should do is use a tool like SLAB6 to create some voxel tiles (each of which would be something like 128x128x128 voxels in size). Then, create a little tile editor that writes some tile data to a 3D array. Design the editor so that it can output these 3D tilemaps to a file. Then, when your Voxlap game starts, just read one of these level files off your hard drive. You can then run through the loaded tile map with some for loops and "draw" these tiles onto the voxel game field. It's really not much different from drawing 2D tiles on a 2D display.

[Scott_AW]

I'm already using that method in Evaldraw, I've created a map maker and map format.  I'm considering VXL format due to it already having collision detection and I could do a little more with it.

Although I could try higher resolutions in the tests, right now 64x64x64 is the standard object/tile size.

[Redeemer]

Well, if you're goal is to make a simple rogue-like game, then I don't think you really need anything Voxlap offers. Besides, you've made significant progress in Evaldraw. Switching to voxlap and real C would require almost a complete rewrite of everything you've got.

[Scott_AW]

Okay maybe I'm over complicating it and should stick to what I'm doing.  Actually converting to using voxlap, or at least parts of it, isn't too difficult.  A lot of evaldraw is very close to C++.  The only thing I really needed to move away from evaldraw was a way of efficiently rendering voxel objects, which voxlap can do.

A lot of the current code is mostly logic and math, so translating it shouldn't be too difficult.  I'll see if I can't get a very basic demo out soon.

[Atomic]

I'm with Redeemer on this one, it would be cooler to see support for a 3D tiled voxel map than a 1024x1024x256 map. And it'd probably be easier as well.

[Scott_AW]

Think I'll stick with it then, I'll save the VXL concept for something else.  Maybe if I revisit the spaceship thing I did awhile back.