What's everyone working on? (tigsource inspired)

[Davidobot]

I remember yours was based off mine. So now mine will have some of yours possibly.

That's right, you were and are a big inspiration for me. We can carry this cycle on until we have achieved the perfect raycasting engine!

One thing I had never noticed before was that you have walls that can be seen through. Like the metal grates. I could never figure out how to do them right because I didn't know how to keep the casting line going even after hitting a wall and have it return multiple strips. I'll have to see how you do it. I'll probably learn a lot from you too.

For that, you can look into raycasting.lua, starting from line 64. (well, at least that's correct for the latest version)

[Jasoco]

I just remembered the other reason I don't use textured polygons for doors. I can't individually shade each strip for fog and darkness like I can with wall strips. Well not without modifying the library even more to do gradient shading.

BTW, does anyone have that 0.10.0 OS X link? I want to try it out. Is there a change log too? How much into development is it?

[davisdude]

I can't find the 0.10.0 download, but I can answer your other question...

Is there a change log too?

https://love2d.org/wiki/0.10.0

[gomez]

a small gif that i've worked in recent days.



This is a port (using the LÖVE) of a game I made in html5 (using the construct 2) last year. I didn't have the construct license, so I couldn't publish it. Well, now, with LÖVE, I can

It still in alpha, but you can play the html5 version here: https://www.scirra.com/arcade/shooting- ... ect-me-125

[Davidobot]

So, Jasoco got me interested in my raycasting engine again and since I can't afford to fall behind, I've been working on some stuff.
Basically, that screenshot right there, shows everything, well mostly everything, my code is capable of:

• Doors
• Walls
• Transparent walls [Not shown]
• Sprites of different heights
• Floor/Ceiling
• Looking up/down
• Changing the player's height (zHeight, basically jumping/crouching) [Not shown]
• Multiple levels (up to 3, for now)
• 8-directional sprites [Not shown]
• Fog

The world is navigated with WASD and a mouse.
Now to make an actual game....

[Tst_]

Playing about with the Android port, decided to port over a mass tile concept I made to so how it handled it.

[davisdude]

Transparent walls [Not shown]

How do we KNOW they're not shown?

[Jasoco]

So, Jasoco got me interested in my raycasting engine again and since I can't afford to fall behind, I've been working on some stuff.
Basically, that screenshot right there, shows everything, well mostly everything, my code is capable of:

I played around last night and created a large map with a big open area in your engine and you get the same slowdown I do when viewing a lot of floor tiles. We really need to create a new shader for the floor that runs once per floor and ceiling and does the whole thing at once and possibly works the same way it does in older textured floor FPS games like DOOM and Blake Stone. I could bump my performance back up to 150+ FPS again. The textured polygon shader just isn't feasible when a lot of tiles are involved and you need a large open room.

[s-ol]

-snip-

I played around last night and created a large map with a big open area in your engine and you get the same slowdown I do when viewing a lot of floor tiles. We really need to create a new shader for the floor that runs once per floor and ceiling and does the whole thing at once and possibly works the same way it does in older textured floor FPS games like DOOM and Blake Stone. I could bump my performance back up to 150+ FPS again. The textured polygon shader just isn't feasible when a lot of tiles are involved and you need a large open room.

how about baking the floor into one canvas in 2d and then drawing that as one transformed large tile?

[Jasoco]

-snip-

I played around last night and created a large map with a big open area in your engine and you get the same slowdown I do when viewing a lot of floor tiles. We really need to create a new shader for the floor that runs once per floor and ceiling and does the whole thing at once and possibly works the same way it does in older textured floor FPS games like DOOM and Blake Stone. I could bump my performance back up to 150+ FPS again. The textured polygon shader just isn't feasible when a lot of tiles are involved and you need a large open room.

how about baking the floor into one canvas in 2d and then drawing that as one transformed large tile?

Because the math used to calculate the 3D gets completely broken if the point is behind the camera. (Negative distance) Which causes the calculated rear points to be inverse and up in the opposite edge of the screen. Which is why you need to break it into separate tiles. Believe me, if there was a way to calculate that all properly with all four corner points being in the right place, I'd be happy.

Until we figure something out, it's going to be a terrible bottleneck. And I'm on an i7 quad-core Haswell with two threads per core and a 2GB GPU. But each textured polygon that's drawn takes a small amount of time to render and when you have to do 30-60+ each frame, it just gets slow. Combine that with having to actually calculate the grid intersection point locations as well and it gets unwieldy.

I will say however, David does do it better than I do right now. I'll be looking into mimicking his method. (He seems to only calculate the corner locations when drawing the tile while I currently have it calculating the entire grid every frame which makes it super slow when using a 64x64 sized grid. (The maximum size of a Wolfenstein level.)

Of course, making sure to not use really large open areas would help too. The less empty visible tiles means less textured polygons.