Next game to port?

[d4rk3lf]

Let's make a compromise.

- Final Fight - Aga only (for now ).
- Streets of Rage - A500

[daxb]

Originally Posted by mcgeezer

There does not seem to be a well defined hardware standard at a reasonable price that warrants creating games for expanded Amigas.

Of course there exists a well defined hardware standard. Here we go: OCS/ECS/AGA/RTG, 68000 to 68060 (some might add ppc), 0,5MB to 2MB chipram, HD/floppy and as much or low fastram do you wish.

high powered CPU with lots of RAM and high quantities of storage. After all, with enough brute force anything can be done.

I would call a 68k CPU very low powered or do you mean power consumption? Some MB RAM I wouldn't call lost of. Storage... really? I didn't knew that an Amiga is so capable that it can do anything.

[roondar]

Originally Posted by daxb

I would call a 68k CPU very low powered or do you mean power consumption? Some MB RAM I wouldn't call lost of. Storage... really? I didn't knew that an Amiga is so capable that it can do anything.

Needing several MB of fast memory, a HDD and a CPU that's well over 10x the speed of the base model just to do something that a 16 bit console can do fairly easily doesn't seem like anything other than using brute force to overcome chipset limits to me. Which was my point.

[mcgeezer]

Originally Posted by daxb

Of course there exists a well defined hardware standard. Here we go: OCS/ECS/AGA/RTG, 68000 to 68060 (some might add ppc), 0,5MB to 2MB chipram, HD/floppy and as much or low fastram do you wish.

Well there's about 10 standards you have right there over and above say.... A Mega Drive, SNES or < INSERT YOUR CONSOLE OR 8 BIT MACHINE HERE >

[VincentGR]

Megadrive is a sprite/tile beast.
Saw some Amiga ports on the Archie that were close enough and almost never a pure 68030 game.
4 mips to 9 is a huge difference but dunno if it is enough to make games like on the MD.
A friend who coded games on the Amiga back then told me that even with an 040 he could achieve 2-3 more cpu sprites.
Don't really know if this was a true limitation of the hardware or the coder.

I'll go with mcgeez here.
No prob with aga and accelerators but A500 is the target for me too.

[dlfrsilver]

Originally Posted by d4rk3lf

Partially selfishness (because I am proud owner of A500 ), partially curiosity what can be done with A500 1MB.
Back in the day I owned A1200, though.

Either way, I am cheering, if anyone makes port for whatever Aga or OCS, just think that Aga with 4MB wouldn't be that much of a challenge, therefore less interesting.

I have also 2x A500. But at the same time, Final Fight has been built for Strong 68K machine (read in my mind : X68000/CPS1).
For me Richard Aplin has gone and done the best he could squeeze out of an A500 with 1mb of ram.

When Roondar says that going for AGA+030 is useless because we have mame, i note that when Capcom ported Final Fight on X68000 from the Arcade source code, they didn't told to themselve "my baddd, we have the CPS1 version (or mame ), so what's the point ?". They did because it was cool to try proposing a great version.

The A500 has been juiced up to the maximum, the A1200 (AGA) is mostly "untouched".

[Daedalus]

And that's the thing - the Amiga was designed as an expandable machine, rather than a one-shot platform like the SNES or Megadrive. If a game is complex, it will either need to be compromised in quality or content, or it will need a more powerful computer. Optimisation is always good of course, but it's not magic and there's only so far you can go before the law of diminishing returns make further gains hardly seem worthwhile.

For what it's worth, my limited forays into games development have included working hard to get a game to run well on a 1MB A500. It was a fun challenge, but also a realistic one.

[saimon69]

Well i personally think that A1200 should be used for those projects deemed impossible or too big for the A500 base config, but i still point that failed A500 projects should be revamped; wish to see more hard disk based games for the 500 [mostly aimed at original content but still fit here], also to boost the purchase of hard disk substitutes for those machines; an A500 with an hard disk is immensely more capable of one with floppies only - for the A1200 i would envision projects that can use the copper chunky capabilities, and even voxels, rotations and similar - wonder why there is no Cameltry clone that uses copperchunky in example

[dlfrsilver]

Originally Posted by saimon69

Well i personally think that A1200 should be used for those projects deemed impossible or too big for the A500 base config, but i still point that failed A500 projects should be revamped; wish to see more hard disk based games for the 500 [mostly aimed at original content but still fit here], also to boost the purchase of hard disk substitutes for those machines; an A500 with an hard disk is immensely more capable of one with floppies only - for the A1200 i would envision projects that can use the copper chunky capabilities, and even voxels, rotations and similar - wonder why there is no Cameltry clone that uses copperchunky in example

Final Fight is too big for the A500 if you want something more than what Richard Aplin achieved in 1991.

I have access to Final Fight Arcade program logic disassembled and commented.

Capcom Engineers really went in town : Each enemy is managed by the program as a seperate entity each with its specific.

All in all, the 68000 in the CPS1 board and the X68000 is used mostly to handle the Entities/enemies, Hitboxes, AI, and the tilemetadata upload in the dedicated RAM for automatic sprites/graphic display.

The 68000 on those systems can do that because all the graphic burden is releaved by the powerful chipset on each machine.

Add up the music and sounds, and you can see that there is no way to get a proper version of Final Fight on something else than an A1200+030+FastRAM+Hard Drive.

The X68000 allows Final Fight to load from disks in 2mb of ram because it handles the graphics in the same format as the CPS1 (bitmap chunky), taking less ram space than planar graphics on the Amiga, never mind the use of an YM2151 for the music (=takes almost no room in ram), and use the japanese arcade program rev900613 (so same logic ran as in the CPS1 version).

[skyzoo73]

Seeing ports done well on Amiga ocs is undoubtedly amazing, and I hope many will continue to do so. That said, I find it absurd to try to bring games that had incredibly superior hardware in the original arcade to the Amiga. In my opinion, we should focus on achievable goals and in line with the target machine. Let me explain, some simple and not very recent arcades can easily be done on Amiga500, for everything else I think it is right to take advantage of the Amiga 1200 (and I speak as an owner of Amiga OCS / ECS). I would assume both a minimum of fast ram and a hard drive, definitively abandoning the anachronistic fetish of standard machines and the floppy as the only medium. A must have, for a modern port on Amiga, MUST be the cd32 joypad support. You can't make the eternal mistake of crippling a game's playability with a one-button joystick.

[Dunny]

Originally Posted by Daedalus

For what it's worth, my limited forays into games development have included working hard to get a game to run well on a 1MB A500. It was a fun challenge, but also a realistic one.

Having a massively expanded A500 was not an option for a great many people. There's a good reason that the vast majority of OCS games were for 512Kb Amigas - the majority of A500 owners didn't have an enormous 1MB upgrade.

[saimon69]

Originally Posted by dlfrsilver

Technical detail goes here

You see that unfeasible even with a metro siege approach? [read separated parts for arms/legs/body]

And by the way yes, it means it has to be recoded from scratch - again - together with assets cut 'ad-hoc' for it

[dlfrsilver]

Originally Posted by saimon69

You see that unfeasible even with a metro siege approach? [read separated parts for arms/legs/body]

And by the way yes, it means it has to be recoded from scratch - again - together with assets cut 'ad-hoc' for it

Metro Siege has been made within the limits of the A500.

Final Fight has been recoded from scratch, and its assets cuts more than anything to run on A500. The enemies animations looks like tex avery cartoon when they are running on screen.

The A500 can move big sprites, FF shows it. But it lacks in all the other fields : colors, scrolls, music&sfx, ram.

Look i have participated to Final Fight AGA. I have extracted the music and sfx, and some levels, and it needs a 030 to get multiple enemies on screen and with 32-64 colors on screen, it requires 19mb of fast ram !

[d4rk3lf]

Originally Posted by Dunny

the majority of A500 owners didn't have an enormous 1MB upgrade.

Not from my experience.
90% of people I know, had ram upgrade of 512kb.
I only knew one guy that didn't had it, and he also planned to buy it.
Talking about 1990...1991.. maybe before that it wasn't that common.

There's so many great games that requires 1MB, that I personally consider A500 with 512kb upgrade something like basic A500.

Btw, I agree with above comment, that nobody should aim floppies, even for A500. Port should be done for HD.
Well, if someone's like to swapping floppies, he can have version of the game on 8-14 floppies.

[saimon69]

Originally Posted by dlfrsilver

Look i have participated to Final Fight AGA. I have extracted the music and sfx, and some levels, and it needs a 030 to get multiple enemies on screen and with 32-64 colors on screen, it requires 19mb of fast ram !

AND that is why am asking if subdividing enemies makes it less heavy - for the music i got you covered: i made all the in-game ones for two channels with an average size of 80k each [was for another port attempt], but most use same samples so more optimization can be done

By the way, i think would make more sense to have it Hard Disk only - if we were in the 90s this would sell amiga hard disk like hotcakes, now... guess not :/

Final Fight AGA Playlist

[Tigerskunk]

Originally Posted by Lord Aga

Final Fight is a legendary game. The port should not be half-assed.
It's a case of "go big or go home". So AGA with tons of fast RAM and HDD only. Maybe even high clocked 030.
No more "oh this 256 colour game is kind of OK in 8 colours" nonsense.

Cool. Go for it.
Tell us when you will be finished so we can play it.

[Tigerskunk]

On that topic of non baseline configs:

1) for most people the Amiga as a games machine was the A500 512k/512K.
Everything that was released after its prime time was rather non relevant. Personally, I didn't even know about AGA before 2006.

2) as a coder I am interested in achieving something for a baseline config. Bit of the demo scene idea in action here. Also, when I code a game I want to compare to the big hitters that have been done before on that config, and if I can do better than that. Since there were barely any relevant games released for a 68030 config with 4mb fastRAM, this config is super uninteresting for me as a coder.

3) If you just start expanding, why not just expand till you can port 1:1 without problems. At which point you can also just use MAME and be done with it.

4) Nobody keeps you from creating games for these higher powered configs. Be the change you wanna see. But then be ready for getting a lot of criticism for your game not being up to the standards people expect from such a fast (compared to a plain A500) system. Anything looking below NeoGeo will be a failure in people's POV.

[S0ulA55a551n]

Originally Posted by Steril707

On that topic of non baseline configs:

1) for most people the Amiga as a games machine was the A500 512k/512K.
Everything that was released after its prime time was rather non relevant. Personally, I didn't even know about AGA before 2006.

2) as a coder I am interested in achieving something for a baseline config. Bit of the demo scene idea in action here. Also, when I code a game I want to compare to the big hitters that have been done before on that config, and if I can do better than that. Since there were barely any relevant games released for a 68030 config with 4mb fastRAM, this config is super uninteresting for me as a coder.

3) If you just start expanding, why not just expand till you can port 1:1 without problems. At which point you can also just use MAME and be done with it.

4) Nobody keeps you from creating games for these higher powered configs. Be the change you wanna see. But then be ready for getting a lot of criticism for your game not being up to the standards people expect from such a fast (compared to a plain A500) system. Anything looking below NeoGeo will be a failure in people's POV.

I know where you are coming from, but there were a fair amount of people who upgraded to an a1200. So I think base config holds water for a1200, most people had at least some form of fast ram, same as a500 owners had the 512k upgrade. So i think A1200 2Mb/8Mb is a fair base config as well.

[Tigerskunk]

Originally Posted by S0ulA55a551n

I know where you are coming from, but there were a fair amount of people who upgraded to an a1200. So I think base config holds water for a1200, most people had at least some form of fast ram, same as a500 owners had the 512k upgrade. So i think A1200 2Mb/8Mb is a fair base config as well.

I have to admit I never owned an A1200, so not much of an idea there.
How many relevant games were released for it asking for additional fast RAM?

On the A500 the 512k of slow was almost kind of standard after 1989.

[roondar]

Originally Posted by Daedalus

And that's the thing - the Amiga was designed as an expandable machine, rather than a one-shot platform like the SNES or Megadrive.

This is somewhat debatable, the vast majority of Amiga's sold (i.e. the 70 odd percent of total Amiga sales which were A500's) were not all that expandable. Sure, you could expand an A500, but doing so was usually quite expensive. So yes, the Amiga did offer expandability, but most people only bought systems that didn't really suit it.

Originally Posted by dlfrsilver

When Roondar says that going for AGA+030 is useless because we have mame, i note that when Capcom ported Final Fight on X68000 from the Arcade source code, they didn't told to themselve "my baddd, we have the CPS1 version (or mame ), so what's the point ?". They did because it was cool to try proposing a great version.

Note here that this is actually supporting what I'm saying: Capcom did not make X68000 Final Fight for the highest spec X68000 available when they released it in 1992, but rather aimed it at the most common specs in use at the time - i.e. for the lowest CPU speed and only a memory expansion needed

Also, I never intended to say that it's useless, but that I personally find it useless to do these things today, which really is a different thing.