How to approach this texture scaling issue?

[tokumaru]

I love it, seeing texture on the ground and the head tilting is really awesome. I thought it was already as good as it gets, but no.

Haha! Thanks!

Do you remember a long time ago when I asked about using your raycasting engine in Garage Cart #2 for something fun?

Of course I remember! We had some interesting ideas.

I'd still like to do that, I could really use some inspiration to get that long-languished project moving again.

I'm up for it, but the new version of the engine uses way more look-up tables than before, so I can't guarantee it will fit in just 16KB anymore, it might need 32KB. On the plus side, using objects we can set up rooms with better ambiance.

[tepples]

If Garage Cart 2 uses something like the Action 53 mapper, 32K shouldn't be a problem.

[tokumaru]

I just wouldn't want it to steal space from other things featured in the cartridge.

[Pokun]

I don't think a dungeon crawler RPG is necessarily the best way to go, although it would be an obvious choice - something inbetween that's still somewhat slow and tactical would also be a pretty good idea. When working with limitations such as these, it's important to design something that makes proper use of the unique 3D perspective for gameplay purposes (something Doom actually does better than most newer FPS games), rather than simply working in spite of it.

Like Bergalad I'm not a huge fan of FPS but I am a fan of RPGs so a dungeon crawler is the first I'd think about to make, so I had to mention it.
Yeah a dungeon crawler might not really do the engine justice (unless you allow free roaming maybe), but I think it could improve dungeon crawlers as it at least makes smooth animations easy. Choppily animated dungeon crawlers (or no animation at all like in Portopia and Megami Tensei) makes it easy to loose your orientation.

[Alp]

Looks very good indeed!
Besides an FPS, I imagine it could make a cool dungeon crawler. In games like Megami Tensei it's very easy to get lost due to choppy 3D-mazes and limited access to a map, but with a fully animated 3D-maze like this it would be easier to navigate. Head tilting could be used automatically in scenes like when you discover stairs or so rather than during gameplay.

I was going to suggest this, myself, only because the recently released "Tales of Popolon" is damn impressive, for the MSX!
https://www.youtube.com/watch?v=u9HiZfxoLmk
This game also used smaller characters, eliminating the object-size problems.

But I like Tokumaru's idea more! I'd rather "teach those alien bastards a lesson" than just generically shooting Nazis in the face.

Locomalito's "8-bit killer" took a more Contra-like approach to a Wolfenstein-styled shooter.
https://www.locomalito.com/8bit_killer.php

[Memblers]

I just wouldn't want it to steal space from other things featured in the cartridge.

Oh, that's definitely not a problem. Looking at it now, it has at least one fully empty 32kB bank, and several more that are mostly empty. It went from being 128kB, to 256kB, to a 512kB project and it shows. GC#2 will still be a product of it's time (10 years ago), so it will mostly be stuff from back then, and it will be on plain old 32kB banking on the GTROM board.

[Myask]

How many rays are you casting? 32?
How much time is the render taking compared to the raycasting?

[tokumaru]

The new design isn't running on the NES yet (I'm getting to it!), so I don't have that information yet.

The old raycaster used 28 rays, and rendering the walls was slightly slower than casting rays, IIRC, so the frame rate would drop a bit when you were facing a wall up-close. Thankfully, when actually playing a first person shooter you don't spend much time facing walls, so that shouldn't be a problem.

For the new design I added 2 more columns, for a total of 30 rays, and that's indeed the resolution for wall slices, but each slice contains 2 columns of texture, improving the perceived horizontal resolution.

I did a lot to improve the ray casting compared to the old design, but not so much the rendering. If anything, rendering is probably slower, since now I have floors, ceilings, and 2 textures per slice. I'll probably unroll some of the texturing code, though.