Are you working on a better Polymost?

I am not actively working on any 3D engines at the moment. A few years ago, I did some stuff which I later embedded into EvalDraw, however I never released a demo or documented the functions.

Are you working on a better game engine as good as Id5?

No. Contrary to popular belief, I cannot compete with the state of the art in my spare time. That ship sailed in 1996.

I've been using Windows 7 since November and haven't noticed any problems with Evaldraw and Alt+Tab. I do not use Aero. Recently, I've been looking at a different freeze, which occurs only in the 1D & 2D modes while editing, and the mouse cursor is in the render area. I have found the following settings: "timeoutonkey=0", "showgrid=0", and "usenumcpu=1" reduce (or possibly eliminate) the chances of this freeze, but I'm still not sure why it's happening. Hopefully I will figure it out soon. I couldn't say whether this is related to the Alt+Tab thing or not.

To fix the flicker in XP, set "drawprimary=1" in EVALDRAW.INI. I chose a misleading name since drawing directly to the primary buffer is the reason it flickers, yet you must provide a value of 1 to disable the flicker! I no longer care about a negligible speedup on older machines. As a result, I will change the name of the setting to "drawoffscreen" in the next version, and have it default to no flicker.

When the compiler crashes, you should still be able to edit the text. This has been true since the 03/21/2010 version, when I added custom exception handling (__try&__except). Are you sure you're using the latest version, or did you get it to crash so bad that my exception handling is of little use?

Enough.

English teacher? No way. That was my worst subject in high school.

I don't have a real job if that's what you're asking. The Duke has fed me well. I spend a few days a week working on projects at my dad's lab. Projects include: 3d mice, microphone arrays, and other fun gadgetry. I also do some mentoring at my old high school, where I show the kids how to write simple games in Evaldraw. For the rest of my time, I work on my own stuff, such as research on 3D engines, or maintaining old projects.

What if it's a human-assisted bot? Imagine an app that sends snapshots of the challenge screen to a sweat shop in some 3rd world country. I'm not aware of such an app, but if one exists, the author would surely qualify as a first class douche bag.

Hi Maren,

Some of your suggestions are quite good - thanks for writing them down and sharing it. Please keep in mind that I wasn't planning on doing much more support for SLAB6 - that's why I released the source code. Before I can address any of your ideas, I would need to fix some issues with the mouse cursor, which is not working right under Windows 7.

Storing colors for interior voxels would make high resolution models impractical due to memory requirements. That's a bad tradeoff IMO.

This is the first time I've seen Toowoomba on the USA national news. They were showing videos of cars being overtaken by the newly formed rivers. After the commercial break, they had a cute little story about a three-legged dog who saved a baby.

This thread is only for members with over 500 posts

The Pentium was the last Intel processor where you could estimate clock cycles like this. I had Intel books, but I would also measure each instruction in a test program just to be sure. Measuring clock cycles became a lot more difficult with the Pentium Pro and beyond. These days, my favorite reference is http://www.agner.org/optimize/ when I need inspiration for assembly optimization.

I attached my original test code from 1995. I wrote it in QuickBasic so you may have difficulties getting it to run. To run it, you'll need:
1. To put a TILES000.ART file in the directory. Any version will do.
2. Microsoft QuickBasic 4.5.
3. WinXP or DosBox to run QuickBasic.
Good luck!

Translucence might work well in something like Scott_AW's Voxel Dungeon Crawler. The first step is to find a list of potentially visible wall blocks, using an algorithm similar to a Wolf3D raycaster. For translucence, you would then need to sort these blocks in back to front order before rendering. VDC does not sort blocks because Evaldraw uses a Z-buffer to resolve depth issues. You can't rely on the Z-buffer for translucense though. You would have to write your own splat-style (i.e. rendering back-to-front with overdraw) voxel sprite renderer, since I have not written one that handles translucence. If you don't mind limiting your engine to 4dof, you could gain some speed and write a renderer similar to my old SLABSPRI.

I should probably mention on my website that there are improved versions of ADLIBEMU floating around the net. Adplug uses an old version of adlibemu - very close to my original. DOSBox uses a heavily modified version of adlibemu - I hardly even recognize the code anymore : ) Both should be platform independent.

Unfortunately, that's not a bug. During the raycasting pass, the depth (parallel to the floor) is stored along with each pixel color. During the 2nd pass (projecting the raycasted data to the screen), this depth is converted to a standard (screen parallel) Z-buffer depth. The problem with this method is there is a singularity near the zenith point when you are looking straight up or down. I decided it wasn't worth the speed decrease to fix this artifact. It should be noted that I hacked in the Z-buffer after I had already written the raycasting code in assembly. Obviously, a new engine written from scratch would not be designed this way.

You can use the 3rd angle (actually the first parameter passed to (d)orthorotate) for tilt/horizon correction. The idea is to keep the height component (z direction in Voxlap) of the player's right vector stay near 0. To accomplish this, you can plug in "irig.z*-.02" as your tilt angle to keep the horizon stabilized.

1. It's for mip-mapping. 1 + 1/4 + 1/16 + .. = ~4/3
2. Search for "vbuf format" in VOXLAP5.C.