new amiga 500 doom like !

[no9]

Originally Posted by KK/Altair

I'm not sure how exactly switching sample number works, but isn't is supposed to swap the sample immediately (introducing an audible click)?

Fortunately samples are switched at the end of the loop so it spared us many unwanted clicks. On the other hand volume value switches to new value immediately. Not sure it it was bug or feature.

First channel doesn't click, does it? [ Show youtube player ]

Also, to make full use of wavetables, they could be swapped every VSync - and I don't expect too many MODs with 1 tick per step.

Right.

Another issue with Protracker is it's limit of number of instruments. Stuffing it with lot's of different samples easily fills all slots. And honestly it is a bit of unnecessary hassle. Generating sample variants on the fly (or precalcing it), or finding another way to source them would at least alleviate it.

[KK/Altair]

Another issue with Protracker is it's limit of number of instruments. Stuffing it with lot's of different samples easily fills all slots.

That's exactly my issue with ProTracker MODs - you get tons of limitations just by using this format. E.g. Paula would be perfectly fine with you making a 256-sample buffer with a square wave, and sliding 128-sample window over it to produce PWM effect. But there is no way to do this in a MOD - you can neither move the loop start/end, nor automate it with a LFO with 50/60Hz update rate.

[rothers]

I just wanted to say it's amazing and I really hope it becomes a full game. I will support your patreon.

Absolutely love the idea of a game showing the peak power of the Amiga after 35 years. Would have changed the gaming world as we know it had this been realised in the 80s, fascinating watching your video on how you did this.

For sound I'd just go with atmospheric samples and maybe a heartbeat (which increases with lower energy?).

[Gorf]

There is a nice synthesizer with full source code on Aminet:
http://aminet.net/package/mus/edit/harmonics

maybe you can get some inspiration from that one...

also:
http://aminet.net/package/mus/edit/AproSYS
(without source)

[b0lt-thrower]

Hey, KK/Altair, can you put a pure gameplay video up on Youtube showing WIP?

This just makes me so giddy with happiness.

[d4rk3lf]

Please consider also a "donate" option, because I can't go patreon route ( I am a freelancer), and can't pay monthly.... but from time to time.. yes.
And of course, purchase of the game, when it's done.
-------------

As for the demo release... it's a tough one..
Maybe many people would be satisfied only with a demo (as a proof of point, and a show of), and would never buy a full release. That would be unfortunate.
But if you make a demo really short, and then with few pictures on how awesome future level will be (or short gameplay vids), maybe that would do the trick.
But it will also require time spending on proper demo, rather then a full game.
It's hard to be smart here.

I'd personally perhaps do something like this:
Demo would be a small level (no enemies, no guns), you can walk, and that's it.
But also, a few scripted videos with enemies, and explosions, fast paced gameplay... something like you showed us already.

[Overmann]

I don't think an Amiga game needs a demo these days. It's better to just focus on making a good game, make some videos and the game (if it's a good game) will sell like hotcakes! I think that something like this, even if it's a mediocre game, will sell quite a few copies. It's just such a cool achievement!

[KK/Altair]

Hey, KK/Altair, can you put a pure gameplay video up on Youtube showing WIP?

That will be done for sure once I have something to download up.

Please consider also a "donate" option, because I can't go patreon route ( I am a freelancer), and can't pay monthly.... but from time to time.. yes.

I'm already considering using itch.io for major releases because of that.

As for the demo release... it's a tough one..
Maybe many people would be satisfied only with a demo (as a proof of point, and a show of), and would never buy a full release. That would be unfortunate.
But if you make a demo really short, and then with few pictures on how awesome future level will be (or short gameplay vids), maybe that would do the trick.
But it will also require time spending on proper demo, rather then a full game.
It's hard to be smart here.

You know, guys at Id released a whole EPISODE for free, and how did it harm releases? If anyone is satisfied with a demo, I'm fine with that. My job is to make game so good, that people after enjoying the demo want more of the experience.

Demo would be a small level (no enemies, no guns), you can walk, and that's it.

As I'd like to release something as soon as possible, I think I should go with the demo level. But I still want it to be a complete experience, so enemies and guns will definitely be in (just maybe not all, because of the level size).

It's better to just focus on making a good game, make some videos and the game (if it's a good game) will sell like hotcakes!

The scope of the project is simply to big for me to make it all in one go - something for which I'd have to lock myself in a garage for two years to complete. So I think that making the minimum version (close to something I already have) and enabling modding would be reasonable approach. This way people could already have a lot of fun making maps, sprites, sounds, music and scripts for the game (I certainly would myself, considering how much fun I had with mapping and modding Doom and Quake), while at the same time contributing to the project (if they decide to allow me including their contributions in the game). This would allow me to focus on the core engine, because - mind you - I don't even have variable floor heights yet.

[4pLaY]

This looks quite interesting so far, keep up the good work KK and ignore those few who had a negative vibe in this thread.

[Estrayk]

About the limitations of protracker, remember that you can use complete loop-waveforms in PCM format simulating (c64) waves. The uncompressed module can be very big but when you crunch the samples with any compressor, the .mod goes very little.

For example, one of my tunes:

[ Show youtube player ]

As you can see, it is 57k long, but this mod crunched it is JUST 4k in disk!!!!!!
Why? because I use line-based waveforms samples.

I use an old program from 1994 by Pink/Abyss called Pretracker for make the samples.

[Glen M]

Really excited to see this.

Just a thought on the demo thing perhaps it would be an idea to release a rolling demo of the game rather than playable to show off the engine.

[DanScott]

Originally Posted by Estrayk

I use an old program from 1994 by Pink/Abyss called Pretracker for make the samples.

There is a new one...

https://www.pouet.net/prod.php?which=80999

[Overmann]

@Estrayk

That's really cool! I had no idea this was possible. :P And your tune is fantastic! Really creative. You've inspired me, sir! Off to the AMIGA

EDIT: Well, I can't find a download for the Amiga-version of the PreTracker ANYWHERE, so I guess it's back to bingewatching Jeeves and Wooster on youtube..

[Tigerskunk]

This is such an amazing new venture in Amiga games.

[KK/Altair]

As you can see, it is 57k long, but this mod crunched it is JUST 4k in disk!!!!!!
Why? because I use line-based waveforms samples.

Well, I need solution for exactly the opposite. While 57k on disk is still good, using just 4k chip memory for samples would be awesome.

Just a thought on the demo thing perhaps it would be an idea to release a rolling demo of the game rather than playable to show off the engine.

No way. I've already decided and will get to making the playable demo level when I'll have some more time again (probably in 2 weeks). The demo will be fully playable, just short and without all weapons/enemies. A good demo should give a taste of full gameplay.

[Glen M]

Originally Posted by KK/Altair

No way. I've already decided and will get to making the playable demo level when I'll have some more time again (probably in 2 weeks). The demo will be fully playable, just short and without all weapons/enemies. A good demo should give a taste of full gameplay.

Cool, can't wait to see it and try it for myself.

[waldiamiga]

Originally Posted by KK/Altair

No way. I've already decided and will get to making the playable demo level when I'll have some more time again (probably in 2 weeks). The demo will be fully playable, just short and without all weapons/enemies. A good demo should give a taste of full gameplay.

We'll wait, let Gdansk not be hit by a flood, an earthquake or a fall of the comet

[Mathesar]

Originally Posted by KK/Altair

Yes, but not triangles. It uses BSP partitioning, so each sector can have any possible convex shape. Raycasting is not suitable for anything more than Wolf3D-like engines, and even in this case rendering walls directly was faster (Wildcat wolf was not a raycaster, too).

The technical details of your engine fascinate me a lot. Some final questions to check if I really uderstand your "wad" format :

1) Your map is divided up into subsectors of "ideal visibility", So, each wall is "raycasted" into the map and when you have done that for all the walls you have created a bunch of subsectors. Each subsector now corresponds to a number of walls visible from that subsector.

2) If you have too many subsectors, you merge similair subsectors (in terms of visible walls, eg, maybe they differ by only 1 wall) into bigger subsectors.

3) The subsectors are organized into a BSP (binary space partitioning) tree like Doom. However, I assume that unlike doom which uses somewhat arbitrary cuts, you always cut along the lines of the existing ideal visible subsectors?

4) The leafs of your BSP tree are the actual subsectors. I assume that each leaf contains a list with indices to all the visible walls right? Also, this list is presorted front to back. If so, your "wad" can be quite large right? I mean, if a typical subsector has 10 walls visible and you limit the map to 10000 subsectors than each BSP tree is easily 200...300 kbytes (assuming 2 bytes per wall indice)

5) During gameplay you traverse the BSP tree until you find in which leaf/subsector the player is. You can then directly render (after culling) the list of walls listed in that leaf.

A lot of questions but hopefully the answer is a simple yes

Also, I had an idea for the line-doubling and maybe better dithering:
If the fifth bitplane is enabled during the doubled line (using the copper) and that bitplane is pointed to an "all 1's" plane, color registers 16..31 are selected instead of color registers 0..15 (if I'm not mistaken). Color register 16..31 could contain the same colors except for the dither colors which are swapped. This way you could have checkerboard like dithering albeit at the expense of some chipram bandwidth which could slow down the 68000 a bit.

EDIT: now that I think about it, you could also just swap the dither color registers using the copper. Doesn't cost any bandwidth. Only drawback is that the palette is rather limited then…

[KK/Altair]

Originally Posted by Mathesar

The technical details of your engine fascinate me a lot. Some final questions to check if I really uderstand your "wad" format ...

Simple "yes", but...


1) No raycasting involved. Everything is computed geometrically on the map.


3) Doom only cuts sectors along existing lines. Both in Doom and my case, some cuts end up somehow arbitrary, because each BSP node splits entire map. E.g. first split will always split entire map along a line, which may not make any sense for areas distant from the cutting line/subsector.


4) It's not a simple list, because it contains conditions (disabled on closed doors) and sector numbers for thing rendering, but in general - yes. Currently IIRC it's about 64kB. But it doesn't have to be so. A lot of lines will usually be rendered together, so I could make line groups and use that as a simple form of compression. I just didn't get around to implementing it yet.

EDIT: now that I think about it, you could also just swap the dither color registers using the copper. Doesn't cost any bandwidth. Only drawback is that the palette is rather limited then…

In other words, the result would be a 16-entry palette, with separate colors for top and bottom halves. Nice idea, but I don't think it would be useful in this case.

[Mathesar]

Originally Posted by KK/Altair

Simple "yes", but...


1) No raycasting involved. Everything is computed geometrically on the map.


3) Doom only cuts sectors along existing lines. Both in Doom and my case, some cuts end up somehow arbitrary, because each BSP node splits entire map. E.g. first split will always split entire map along a line, which may not make any sense for areas distant from the cutting line/subsector.


4) It's not a simple list, because it contains conditions (disabled on closed doors) and sector numbers for thing rendering, but in general - yes. Currently IIRC it's about 64kB. But it doesn't have to be so. A lot of lines will usually be rendered together, so I could make line groups and use that as a simple form of compression. I just didn't get around to implementing it yet.




In other words, the result would be a 16-entry palette, with separate colors for top and bottom halves. Nice idea, but I don't think it would be useful in this case.

Thanks for your reply. I *think* I fully understand it. Clever programming
And yeah, effectively you have a different palette for the top and bottom halves of each 2x2 pixel.

Looking forward to the demo!