Port Wolfenstein on Amiga 500 / 1000 OCS

[britelite]

Originally Posted by Miggy4eva

This thread is just meant to be a collection of demo's to show simlar things the Amiga has done and to document all ideas, techniques, brainstorming of how we can do it.

But here's the problem, you can't say "how we can do it" when YOU wouldn't actually be a part of the process. You aren't a part of any kind of brainstorming when you don't understand any of it.

You're already setting your expectations WAY too high, so you would only be disappointed by the end result. Trust me when I say that the ST-version is pretty much exactly what you would get on an A500.

If you want to move it out of the coders forum then fine

Considering that you are unwilling to accept facts when given to you, it would be better to move these kind of threads out of the forum completely.

If you really want to get a good piece of brainstorming done, then actually produce a proof of concept. Because in that case you would actually have acquired some of the knowledge needed to have a discussion on the feasibility of something, and more people would be willing to help you improve. But as it stands now, it only looks like you want others to do stuff for you, so that you will be able to take credit if something gets done.

[nobody]

Originally Posted by Miggy4eva

The purpose of this thread will be to collect any resources we need to make Wolfenstein happen on the Amiga. I will collect links, tutorials, interesting quotes and discussions I get out of other conversations I have read through on this topic, and videos of demo's and games which show the OCS doing everything required to run Wolfenstein, in case some thinks it's not possible.

Finally, please give this some attention and thought, and make constructive posts only to help us move forward. I have wanted Wolf on the Amiga for SO long, it simply must be done. Can we please start work on it. I am ready to any graphics work and other roles that I can.

Problem is that you cannot lead a development team by adding links and quotes or by reducing colors on images. You have to be a major part of the team, either a coder or pay the coders to do it for you. You cannot be a frontman to something that other people can do and don't need you at all.

[turrican3]

do you think it's possible to use blitter c2p routine to make a better stunt car racer on the amiga 500.
I suppose that the speed won't be good.

[britelite]

Originally Posted by turrican3

do you think it's possible to use blitter c2p routine to make a better stunt car racer on the amiga 500.

Nope

[LongLifeA1200]

Project Battlefield
Requires at least 1.5mb before it can work on an A-500/1000 (intro alone requires 1mb before playing). Texture heavy. Plays in a small window for performance. Don't know if there are options to change the window size.

Project Intercalaris (sequel to Project Battlefield)
Requires at least 1.5mb before it can work on an A-500/1000 (not sure if that is only due to the intro). It's texture heavy, but there are some graphical options to improve performance:
1 = Full Screen
2 = Medium Screen
3 = Small Screen
f = enable/disable floor texture
d = enable/disable ceiling texture
a = toggle rotation speed

Behind the Iron Gate
Requires at least 1mb before it will run on an A-500/1000. Fewer textures, but more frames per second.

[turrican3]

Originally Posted by britelite

Nope

Sorry if for you it's obvious, but why nope ?

[britelite]

Originally Posted by turrican3

Sorry if for you it's obvious, but why nope ?

Because Stunt Car Racer doesn't use a c2p. And it wouldn't benefit from being drawn in a chunky mode either.

[Miggy4eva]

Originally Posted by LongLifeA1200

Thanks for compiling that showcase list. Ambermoon gives me an idea for another project that I otherwise would have missed. And that link to the raycasting engine was an interesting read.

Although I'm not much of a fan of Wolfenstein 3D (which I always misspell "Wolfenstien"), I can't help but wonder if porting to the system you suggest is feasible, so I began with trying to get a hardware timeline:

1982 Intel 286 6mhz (up to 25mhz)
1985 Intel 386 12mhz (up to 40mhz)
1985 Amiga 1000 7mhz 256kb RAM (up to 8.5mb)
1987 Amiga 500 7mhz 512kb RAM (up to 9mb)
1987 Macintosh SE 8mhz 1mb RAM (up to 4mb)
1987 Macintosh II 16mhz 1mb RAM (up to 8mb)
1989 Intel 486 20mhz (up to 100mhz)
1992 Wolfenstien 3D

Really difficult to pin-point the average household PC specs in 1992. PCs were very expensive. Around '92, it was common to see PCs being sold with 25mhz and 4mb of ram (Apple doing something similar). I don't mention Commodore here because you're only interested in porting to a stock Amiga 500 / 1000.

A stock A500 is just a few kilobytes short of the minimum memory requirements for Wolf3D, so there is a chance it might have to lose a few things. A stock A1000 has only half the memory of the A500 so there is a high possibility of dismissing a lot more, unless you're prepared to make drastic changes to the way the map works, such as partitioning rooms so that when you activate a door, you teleport to the connected room in front of the door you just activated, after loading has finished swapping memory space for different objects, textures, animations etc.

When it comes to 3D, any Commodore-Amiga without an Akiko chip (the CD32) has no magic tricks that it can perform to outdo a PC of the same CPU speed. All the 3D calculations and placement of pixels are handled by the Motorola 68k processor.

Finding an example of Wolf3D running on a PC of 7mhz to get some idea of performance is difficult. Wolf3D was written to be compatible with a 286. While 286's started at 6mhz in 1982, by the time Wolf3D was written in 1992, it was common to find 286's running at speeds of 10mhz. Although the minimum system requirements for Wolf3D states a 286, a 386 was strongly recommended since those CPUs started at 12mhz and most 286 owners with a CPU speed of 10mhz reported having a "playable" experience of around ~15 frames per second when "full windowed".

If you search for the following videos on YouTube (sorry, links I post to YouTube don't appear on this forum) you will see the difference in performance within the range of 5 to 10mhz:

~5mhz "Wolfenstein 3D hacked to work on 8086/8088 CPU"
10mhz "Wolfenstein 3D on 10 MHz 286 computer"

So as you may see, the A-500/600/1000/2000 is only 2mhz more than the small windowed "slideshow" example but 3mhz less than the full windowed "playable" example.

Of course, to achieve a "full" port, you will need to shift the goal posts to that of the A-1200/3000/4000/CD32. But then we wouldn't see anything interesting (unless you like Wolfenstein 3D).

At the very least, this challenge provides the motivation to learn 2.5D/3D techniques to be used on the Amiga for other games.

Thanks so much for your post, I have added those 2 videos to my original post under a new title at the bottom for Wolf running on other platforms. I found quite some other examples too on Apple iiGS, C64, Gameboy colour.
The 5mhz version you showed is actually running on an 8-bit 8088! At 4.77mhz!

Originally Posted by turrican3

do you think it's possible to use blitter c2p routine to make a better stunt car racer on the amiga 500.
I suppose that the speed won't be good.

I think it would be AWESOME for someone to make Stuntcar racer faster on Amiga! It always suffered from lazyport syndrome.

Originally Posted by britelite

But here's the problem, you can't say "how we can do it" when YOU wouldn't actually be a part of the process. You aren't a part of any kind of brainstorming when you don't understand any of it.

Well I can facilitate it and encourage it. I am not a programmer but I understand more than you think and I'm a very technical person.

Originally Posted by britelite

You're already setting your expectations WAY too high, so you would only be disappointed by the end result. Trust me when I say that the ST-version is pretty much exactly what you would get on an A500.

Really?? Only as slow as that garbage version? I really hope not, I hope blitter and copper and smart progranmming and optimisation can help!!

Originally Posted by nobody

Problem is that you cannot lead a development team by adding links and quotes or by reducing colors on images. You have to be a major part of the team, either a coder or pay the coders to do it for you. You cannot be a frontman to something that other people can do and don't need you at all.

I don't need to lead it, I'm just kickstarting a conversation and gathering information to one place

Originally Posted by LongLifeA1200

Project Battlefield
Requires at least 1.5mb before it can work on an A-500/1000 (intro alone requires 1mb before playing). Texture heavy. Plays in a small window for performance. Don't know if there are options to change the window size.

Project Intercalaris (sequel to Project Battlefield)
Requires at least 1.5mb before it can work on an A-500/1000 (not sure if that is only due to the intro). It's texture heavy, but there are some graphical options to improve performance:
1 = Full Screen
2 = Medium Screen
3 = Small Screen
f = enable/disable floor texture
d = enable/disable ceiling texture
a = toggle rotation speed

Behind the Iron Gate
Requires at least 1mb before it will run on an A-500/1000. Fewer textures, but more frames per second.

Thanks, I will look at videos of these and add something to the original post

[britelite]

Originally Posted by Miggy4eva

Well I can facilitate it and encourage it.

No you can't, which is pretty obvious.

I am not a programmer but I understand more than you think and I'm a very technical person.

Based on everything you've said, I very much doubt this.

Really?? Only as slow as that garbage version? I really hope not, I hope blitter and copper and smart progranmming and optimisation can help!!

You can hope all you want, that doesn't change facts however. And I already asked you to give some REAL examples on where the blitter and copper could help, and you've failed to provide me with any. You can't claim you're a technical person if you don't even understand the fundamentals.

[Samurai_Crow]

Stunt Car Racer is a good example of what an Amiga 3D game should be. It uses the Blitter to plot polygons instead of chunky to planar routines. The Copper is used to shrink the visible display of the 3D portion of the screen. This may even allow reducing the palette depth of the non-critical portions of the display such as the scoreboard so that the blitter may use more bandwidth. Even so, it should be noted that it runs at a barely playable frame rate by modern standards.

What is needed is a way to improve the 3D capabilities of the blitter because the bandwidth simply isn't there where we need it. An accelerator card with Fast RAM would allow the CPU to plot the border bits of the scene so that the blitter could concentrate on the fill mode only. The 68000 lacks the speed to do this and run the game at the same time.

In short, software optimization alone will be a fruitless exercise without hardware innovation to also be used.

[roondar]

I always figured the main problem for any Amiga 3D game would be the pixel fill rate (i.e. bandwidth) of the OCS chipset. Even if you could do every other thing 'for free', merely redrawing all the pixels on screen will most likely make you drop to 25FPS or less - after all, the chipset can copy a maximum of ~35KB/PAL frame (assuming nothing else uses the bus at all) and that is not enough to copy a full 320x256x16 colour bitmap.

Of course, you could try to cheat somehow, but cheats tend to be of somewhat limited use in case of games - they are either too limiting or too obvious.

[Amiga1992]

Originally Posted by Miggy4eva

Well I can facilitate it and encourage it. I am not a programmer but I understand more than you think and I'm a very technical person.

Not only have you proved that you aren't a technical person, at least in Amiga terms (perhaps you are very skilled when opening up a car, or a swell wristwatch technician), "facilitate and encourage it" you do not.

First, telling people that ACTUALLY KNOW THE AMIGA INSIDE-OUT what they say cannot be true, when they are facts, and you have not the know-how to refute with a counter-argument, is not encouraging, it's the opposite (hence britelite's words that you said "do not understand").

Second, this whole thread and your attitude, does not facilitate anything. Anyone can go and look at things on the Internet. However if you were to provide working code to support your outlandish ideas, that would actually be facilitating. But you can't.

Again, you are being TOLD by those WHO DO already make things for Amiga and know the hardware very well, that if you are SO SURE this would work, you should make a proof of concept or provide any example code where you PROVE that your statements are true. Yet you do not, because, of course , you are "not a programmer". Then if you are not, how can you know the slightest bit to ascertain any of your claims? You don't. so stop acting like you do. It's irritating to the max.

[Miggy4eva]

If you don't believe it's possible then you don't have to reply to the thread. It's not a troll or abusive thread, it's not reaching out and smacking you in the face. If you truly believe I am wrong and have already told me why, then what is the point to keep telling me "no".
This thread is for people who want to explore the topic, if you don't then just ignore it.

Secondly, you and britelite seem to be arguing that nobody can talk about a wishlist unless they know how to make it themselves. You all know I can't program so basically "shut up unless you can program it" translates to "shut up", I'm not allowed to talk about potential Wolfenstein ports on the Amiga OCS.

[Samurai_Crow]

Originally Posted by Miggy4eva

If you don't believe it's possible then you don't have to reply to the thread. It's not a troll or abusive thread, it's not reaching out and smacking you in the face. If you truly believe I am wrong and have already told me why, then what is the point to keep telling me "no".
This thread is for people who want to explore the topic, if you don't then just ignore it.

Secondly, you and britelite seem to be arguing that nobody can talk about a wishlist unless they know how to make it themselves. You all know I can't program so basically "shut up unless you can program it" translates to "shut up", I'm not allowed to talk about potential Wolfenstein ports on the Amiga OCS.

What can we do to help you understand?

[Miggy4eva]

I just want to talk about Wolfenstein 3d on Amiga.
I won't say anything about "it should be better than Atari" or "30fps" for now. I feel and believe in my heart it is true but it's just causes arguments.
Let's just focus on what the best possible version of Wolfenstein on the Amiga could be.
And I KNOW that is possible. Starting point would be a bare polygon coridor engine with no shading, and big blocks for enemies. And it can only get better from there.

[britelite]

Originally Posted by Miggy4eva

Secondly, you and britelite seem to be arguing that nobody can talk about a wishlist unless they know how to make it themselves.

We have been trying to explain why your "wishlist" has no basis in reality and is just nonsense. You can't go around making random claims if you don't have the knowledge or capacity to back it up.

I'm not allowed to talk about potential Wolfenstein ports on the Amiga OCS.

You're allowed, but nobody will take you seriously unless you can show that you actually know what you're talking about. All the participants in a discussion need to have knowledge about the topic it hand, and YOU don't have the knowledge.

[DH]

Originally Posted by Miggy4eva

If you don't believe it's possible then you don't have to reply to the thread. It's not a troll or abusive thread, it's not reaching out and smacking you in the face. If you truly believe I am wrong and have already told me why, then what is the point to keep telling me "no".
This thread is for people who want to explore the topic, if you don't then just ignore it.

Secondly, you and britelite seem to be arguing that nobody can talk about a wishlist unless they know how to make it themselves. You all know I can't program so basically "shut up unless you can program it" translates to "shut up", I'm not allowed to talk about potential Wolfenstein ports on the Amiga OCS.

<sigh>

Problem with creating this thread in the coders section, is it's full of Amiga coders that KNOW what they're talking about. Why even attempt something that these, let us say EXPERTS, already know that it cannot be done, or rather, won't even make any attempt until you can supply some solid evidence that the project is worthwhile.

Unfortunately, from a coders point of view, you're asking the impossible and believe you me, they DO KNOW what they're talking about.

So please, try to understand, and try not to keep arguing that it can etc. You're not a coder, you're in no position to question these guys, but perhaps if I move the thread to a more appropriate forum, you may start to get the desired responses

[Samurai_Crow]

Originally Posted by Miggy4eva

I just want to talk about Wolfenstein 3d on Amiga.
I won't say anything about "it should be better than Atari" or "30fps" for now. I feel and believe in my heart it is true but it's just causes arguments.
Let's just focus on what the best possible version of Wolfenstein on the Amiga could be.
And I KNOW that is possible. Starting point would be a bare polygon coridor engine with no shading, and big blocks for enemies. And it can only get better from there.

It's possible but really really slow to use the MS-DOS techniques for doing it. It would be better to start over from scratch. Like the Walls demo.

[DH]

For now I have moved this to the Retrogamers Forum, as in reality it doesn't belong in the coders forum as it's basically a discussion, without hard proof, that it can be done successfully.

Perhaps now you may get some desired posts.

[britelite]

Originally Posted by Miggy4eva

I just want to talk about Wolfenstein 3d on Amiga.

Then talk about it, but don't dismiss the facts from people who have actual knowledge about the subject.

I feel and believe in my heart it is true but it's just causes arguments.

Belief and facts usually don't mix.

Starting point would be a bare polygon coridor engine with no shading, and big blocks for enemies.

Still, nobody will do the work for you. But this would be a good way for you to begin your journey and learn how to program.