Port Wolfenstein on Amiga 500 / 1000 OCS

[LongLifeA1200]

So I don't have permission?

Enthusiasm is tiring? Guess what, people at EAB are enthusiasts. To an outsider's perspective, they'd be labelled as overenthusiastic. Your argument fails to make sense.

You've made it clear from the beginning that you are totally against this project. It's ironic that you speak of others making demands, when you keep asking for concrete. You've missed the point entirely.

No one is demanding that you or anyone else make the port. Stop taking the open invitation of the OP personally. Do you really believe you're saving people time and disappointment? It's their time to waste if they choose. It's their choice to risk disappointment. The result can be that they still enjoy trying.

But now that the OP has wisely decided to stay away from negative rhetoric it suddenly becomes all my fault for you? I see this sort strange social phenomena too frequently. An optimist encounters a pessimist that doesn't understand that sometimes it's about the journey and not the destination yet since misery likes company the pessimist will persist in convincing others to lower their energy.

Now, enough of this meta-discussion. Time to address your lies.

• You were not being "negative", but instead were merely providing facts? Honestly, what were they? Read the thread again and at least point to one.
• Despite no one providing you "concrete" and stating that to even discuss such a project is a waste of time you simultaneously were seriously "considering" porting this game? With that attitude? I don't think so.
• You want to pretend that I discourage/demotivate others from making games for Amiga? What kind of opposite-world logic is that?
• You want to believe that 2x2 pixel definition on a 160x128 screen doesn't induce headaches? That has not been my experience.
• I suppose you want to skip over making quality mods even though sound/music is what made the Amiga stand out in its prime.
• There is a Ray Casting engine demo that runs smoothly on 1MB OCS. A BSP (Binary Space Partitioning) demo for OCS has yet to be made. When I was thinking out loud about how many tri-polygons OCS can do a frame, I wasn't expecting an immediate answer, and yet you replied that I should look at OCS games and I'll know that way - as if I hadn't already been doing so. Obviously I'll have to write a test-program myself. But it just goes to show how little invested people are who have an agenda to shut down a thread just because someone else dared to even talk about development on the OCS.

[lilalurl]

Back on-topic please.

If need be, we can split the thread between "Port Wolfenstein on Amiga 500 / 1000 OCS: POSSIBLE" and "Port Wolfenstein on Amiga 500 / 1000 OCS: IMPOSSIBLE"

but that would be silly.

You guys are all grown-ups and therefore should be able to ignore any post that you find over-enthusiastic / over-negative (circle your choice).
If not, keep out of the thread and go and play one of the recently released Amiga games .

[NorthWay]

Has anyone done any back-of-the-envelope calculation for a setting with byte-per-pixel but only using 16 values and a blitter c2p in 4 bitplanes?
Running time for the c2p in different sizes, and matching running time for the cpu rendering? Or are we only hand-waving so far?

[Galahad/FLT]

Certainly doing it in 16colours shouldnt be seen as any sign of giving up, I doubt the original Wolfenstein was much more colourful.

[Foebane]

What would cause the main slow-down with an A500 port? The walls? The scaling of the sprites? If I remember, the walls, on average, only made up about a third of the playview, as the floor and ceilings are single colours. If routines could be written that render the walls at a low resolution, and scale at the same resolution with the help of the Blitter, then it should be easily possible. I've seen A500 demos that achieve amazing full-screen effects, but at a fraction of the resolution which is disguised by stippling: AttentionWhore's Serotonin and demos by the group Unique, all on A500.

[Samurai_Crow]

The texture mapping works better on chunky modes so using stencil mapping instead would take better advantage of the blitter. Secondly, the original version just used 8 textures to represent the direction faced by the characters. If we used polygons with coordinates in polar notation on the XZ plane, rotation around the Y axis would be easier and yield smoother modeling overall. Finally, if we XOR the border bits between the polygons, we could combine the fill operations into one big one.

[saimon69]

I cannot remember for the sake of me the title of one of the early wolfenstein clones for Amiga where there was no texture mapping used beside in some single places at one bitplane. I have the idea that we could alternate stencil+lowpoly to one of those one bitplane maps in some places (like the face of hitler) or even use the one bitplane in combination with different background colors, however am not sure of the technical details.

[Samurai_Crow]

@Saimon69
The Mac Classic version could, when using QuickDraw, use all non-textured polygons with its port of Wolfenstein3D. It did require 8 Megabytes to run, however.

[lilalurl]

Two posts which didn't bring anything to the thread moved to the Off-topic section. Keep similar posts there (as long as you respect the board rules).


We want the thread to move forward. Anything that goes another way will be moved or removed at any moderators' discretion. If in doubt, ask a mod before posting.

Potential culprits will consider this post as a last warning before a sanction is taken.

[britelite]

As I don't really enjoy being called a liar, I thought I'd share a little sneak peak of what could have been.



Sprites were never impemented and the wall rendering could still be optimized further. The wall rendering is done with vertical stripes of unrolled code for each zoomlevel and textures are 64x64 in size. Rendering is done in 2x2 resolution, but the blitter c2p routine used allows for two colors within each 2x2, giving free dithering. The ceiling/floor is currently being cleared with the blitter, and possible game logic would probably be performed during the blitter operation.

Last but not least, an non-interactive executable, runnable on a standard A500 512k+512k configuration:

http://www.dekadence64.org/wolf3d.lha

This project has now been shelved for my part, but if anyone else is working on a wolf3d-port, then I'll be glad to help.

If anyone is interested in discussing the technical part, I can start a new thread for that.

[Foebane]

Originally Posted by britelite

Last but not least, an non-interactive executable, runnable on a standard A500 512k+512k configuration:

http://www.dekadence64.org/wolf3d.lha

Very nice, Britelite!

[jizmo]

Love it when someone puts their money where their mouth is. Well done.

[Retro1234]

Originally Posted by britelite

As I don't really enjoy being called a liar, I thought I'd share a little sneak peak of what could have been.

Sprites were never impemented and the wall rendering could still be optimized further. The wall rendering is done with vertical stripes of unrolled code for each zoomlevel and textures are 64x64 in size. Rendering is done in 2x2 resolution, but the blitter c2p routine used allows for two colors within each 2x2, giving free dithering. The ceiling/floor is currently being cleared with the blitter, and possible game logic would probably be performed during the blitter operation.

Last but not least, an non-interactive executable, runnable on a standard A500 512k+512k configuration:


http://www.dekadence64.org/wolf3d.lha

This project has now been shelved for my part, but if anyone else is working on a wolf3d-port, then I'll be glad to help.

If anyone is interested in discussing the technical part, I can start a new thread for that.

Brilliant Demo - Seems fast but I guess after the game logic if needed the speed could be increased by making the view smaller.

could I get a copy of your textures? Did you store them in some kind of array?

I know I should look at something like 3D monster maze or something but how do you calculate angles?

[waldiamiga]

WOW! Could finally be a Wolf 3D port on the A500

[britelite]

Originally Posted by Retro1234

could I get a copy of your textures? Did you store them in some kind of array?

The textures are just the original Wolf3d textures color reduced to 16 colors, and I ever only converted the two textures visible in the demo. The textures are stored in a chunky format (one byte per texel), although rotated 90', so that they can be read more easily when drawing the walls.

I know I should look at something like 3D monster maze or something but how do you calculate angles?

It's a simple raycaster, based on this tutorial http://lodev.org/cgtutor/raycasting.html

In this demo the raycasting is a 512 frame precalculated array, as I was still mainly optimizing the wall rendering and wanted the raycasting part to have constant speed.

I also now noticed that I had forgot to enable blitter nasty in the c2p, so the demo runs slower than it should

[Reido]

Originally Posted by saimon69

I cannot remember for the sake of me the title of one of the early wolfenstein clones for Amiga where there was no texture mapping used beside in some single places at one bitplane. e bitplane maps in some places

Behind The Iron Gate?

[Samurai_Crow]

Originally Posted by Reido

Behind The Iron Gate?

Looking at the HOL database it looks like a good start. What I'm thinking about stencil maps may require a demo to accurately describe the algorithm. It will amount to drawing bricks using an optimized polygon renderer.

[Retro1234]

Originally Posted by britelite

The textures are just the original Wolf3d textures color reduced to 16 colors, and I ever only converted the two textures visible in the demo. The textures are stored in a chunky format (one byte per texel), although rotated 90', so that they can be read more easily when drawing the walls.

It's a simple raycaster, based on this tutorial http://lodev.org/cgtutor/raycasting.html

In this demo the raycasting is a 512 frame precalculated array, as I was still mainly optimizing the wall rendering and wanted the raycasting part to have constant speed.

I also now noticed that I had forgot to enable blitter nasty in the c2p, so the demo runs slower than it should

Thanks for the reply - what are the dimensions of the original textures 64x64?

in a game with free movement could all angles be precalculated? if not how long does it take to calculate in real time?

bit lame I know but I could record your demo split it into frames then try and recreate to get the angles correct - im just curious, im not going to try what youve done.

Exellent demo

[britelite]

Originally Posted by Retro1234

Thanks for the reply - what are the dimensions of the original textures 64x64?

Yes, the original textures are also 64x64

in a game with free movement could all angles be precalculated? if not how long does it take to calculate in real time?

Never timed the raycaster, but in it's current form it's way too slow and would have to be rewritten

bit lame I know but I could record your demo split it into frames then try and recreate to get the angles correct - im just curious, im not going to try what youve done.

If you want test data for trying out wall rendering, I can send you a dump of the data used in this demo (around 160kB). The horizontal resolution is 160 pixels, so for each vertical column there's one byte for wall height and one byte for texture offset, so there's 512 frames of 320 bytes per frame.

[robinsonb5]

Originally Posted by Retro1234

in a game with free movement could all angles be precalculated?

Yes, it's easy to build a precalculated array of angles, one for each column on the screen, then these can be rotated just by adding the current viewing angle.

Each column of pixels to be rendered, the centre of the screen and the viewer's eye form a right-angled triangle; the angle for each ray will be atan2(column-(screen_width/2) , distance_to_viewer) where distance_to_viewer is scaled so that it's measured in units about the size of a pixel. The field of vision can be changed by adjusting this value, conceptually moving the viewer's eye closer to or further way from the window-into-the-world which is the screen.

@Britelite: just tried your Wolf3dpreview. Nice work - very cool! Would be great to have a thread for the technical aspects.