Beats of Fire

[RichAplin]

Originally Posted by Shatterhand

Even if I didn't like any of your Amiga games and used some harsh words to talk about Amiga Final Fight, I have all the respect for people working on games at that time, and I should've used better words to talk about your work.

Heh it's fine, some of the Youtube comments are a hundred times nastier. I certainly don't take personal offense, I find it somewhat amusing that people get so worked up about something that really doesn't matter in the grand scheme of things; it's really juvenile, yet (I assume) that the people commenting were around when the Amiga was so are in their 30s/40s now.

I totally won't pretend I enjoy playing any of those games; actually I found that working in the industry tends to kill your enjoyment of playing videogames; it's like working in a candy factory ... actually worse because during development you spend a lot of time playing the same (incomplete) game levels over and over and over again and it stops being fun very quickly.

There's a few games (other people's) I did enjoy (e.g. the Mario games are amazingly creative and fun) but nowadays I make a point of not working professionally on anything that has much of a user-interface; be it websites or videogames; if it's "customer facing" that means you get a ton of bug reports saying things like "Make the font bigger" (then a day later "Make it smaller again"!) or "Move it left 10 pixels" or "doesn't work on this one Android device" and after months on a project those things get really irritating.

I like the under-the-hood technical challenges myself :-)

I tend to do either backend server-side or embedded systems work nowadays; both opposite ends of the geekin' spectrum, but in both cases clients are able to define quite precisely what they want upfront and don't tend to change their mind much, also you don't have to deal with lots of of slightly incompatible web-browsers/phones/etc.

On one embedded project (a hearing aid with two buttons and an LED for a user interface) I did get a bug report saying "Make the LED more yellow!" but generally it's a lot less frustrating to work on.

On the server stuff the only things people tend to care about are "is it done yet?" and "will it still work with X million users?" both of which are fairly easy to answer.

So yeah, "Amiga Final Fight is a shitty conversion".. sure, whatever. I agree it's a shame it didn't have in-game music, but that was out of my control. I'm still fairly happy with FF from a purely technical perspective - much moreso than say Double Dragon 1, DD2, Line Of Fire, etc.

It's nice to come to this thread and see someone having another go at converting it, on hardware that's only moderately more capable than I had to target. I'm personally a bit mystified why anyone would bother, in 2015, but... that's the joy of hobbies, you don't have to justify them.

Actually I just made https://www.google.com/search?q=haig...ry+hippie+trap so I clearly spend time on frivolous geeky things :-)

[gazj82]

Well the first version of Final fight I saw was on my friends ST and considering I only had a C64 at the time the graphics and gameplay looked amazing. The original Amiga version is much much better. If I had an Amiga back then I would have been very pleased with it. Unfortunately I didn't get my A1200 until everyone had moved onto PC's.

[edd_jedi]

Originally Posted by RichAplin

if it's "customer facing" that means you get a ton of bug reports saying things like "Make the font bigger" (then a day later "Make it smaller again"!) or "Move it left 10 pixels" or "doesn't work on this one Android device" and after months on a project those things get really irritating.

You've basically just described my job :

[leathered]

Originally Posted by RichAplin

Heh it's fine, some of the Youtube comments are a hundred times nastier. I certainly don't take personal offense, I find it somewhat amusing that people get so worked up about something that really doesn't matter in the grand scheme of things; it's really juvenile, yet (I assume) that the people commenting were around when the Amiga was so are in their 30s/40s now.

Perhaps your post here veils a kind of irony – but I think it highly likely that a lot of the worst comments on you tube are from people just hearing about the Amiga and when they come across older stuff just slandering it. I'm thinking mostly from gamers born into the 'playstation age' for whom even the Snes was already an old console.

Originally Posted by RichAplin

I totally won't pretend I enjoy playing any of those games; actually I found that working in the industry tends to kill your enjoyment of playing videogames; There's a few games (other people's) I did enjoy (e.g. the Mario games are amazingly creative and fun) but nowadays I make a point of not working professionally on anything that has much of a user-interface; be it websites or videogames; if it's "customer facing" that means you get a ton of bug reports saying things like "Make the font bigger" (then a day later "Make it smaller again"!) or "Move it left 10 pixels" or "doesn't work on this one Android device" and after months on a project those things get really irritating.
I like the under-the-hood technical challenges myself :-)
So yeah, "Amiga Final Fight is a shitty conversion".. sure, whatever. I agree it's a shame it didn't have in-game music, but that was out of my control. I'm still fairly happy with FF from a purely technical perspective - much moreso than say Double Dragon 1, DD2, Line Of Fire, etc.

Now I really don't want to open a can of worms here...
I think that your conversion of Final Fight was impressive technically, and as you say you never found it enjoyable as a game anyway. Without going into too much it plays a different game from the arcade machine – I can appreciate how unlikely that seems given that it's just a couple of bobs fending off a few more by beating them with whatever is at hand, but that brings me to the other group of critics – the 'mature' Amiga owning gamers.

Not all said Amiga gamers by any means, check out the reviews on Lemon and they are by and large positive – but it is a game that divides people.
Casual beat em up fans will love it, I bought the Atari ST version and it swapped several hands at my school (did I ever see it again?), never expecting the arcade, just fingers crossed that 'US Gold' had come up with the goods.
I enjoyed it, especially in 2 player, but as my rose tinted spectacles faded over time (which came in very useful for us STE owners still dreaming that we had just bought an Amiga!) I began to miss the intensity of the arcade somewhat. But even the Snes version (I ended up with the famicom version) was lacking when it comes down to it - and that was made by Capcom!
Years later I can see that what was missing is that frenetic 'controlled burst' button mashing combined with the technical depth of the gameplay – you'll miss it if you're just looking for a button masher, but there's a lot under the hood that makes the arcade the game it is. It has a core that is similar to Street Fighter 2 – timed combos that are totally simplified for Final Fight, frames of invulnerability built in to certain moves and a great move set. But 6 months just isn't enough time to get all of that in, especially if you were not used to the game in the first place.

But I can also appreciate the original Amiga release for what it is, and I don't think it's nearly as bad as all that. For a home computer the difficulty level is about perfect – you'll need to develop a strategy to get through and work out the best moves for a situation – which is generally when the goons gang up on you. I never had a problem with the gfx, and the variety of goons on screen and throughout every stage is impressive. Boss fights could have done with better scripting but meh – you had 6 months... For the the hardcore final fight fans it was never going to be quite the same. I think the same is true now. My attempting of it must surely be down to as you say – being a hobbyist and perhaps re-kindling a fire from my youth which all started with little one on the horizon. I'm not the coder you are and we'll see if I can even adequately finish a version that will run well without some hideous game breaking bugs!

Originally Posted by RichAplin

It's nice to come to this thread and see someone having another go at converting it, on hardware that's only moderately more capable than I had to target. I'm personally a bit mystified why anyone would bother, in 2015, but... that's the joy of hobbies, you don't have to justify them.

And why Final Fight? I think Gibs has the right idea with converting Double Dragon (again!), it's much better suited for a conversion to the Amiga than this is!

I too enjoy the technical challenge, and can relate in some part to your experience of 'customer facing' – mostly within this very thread!

[Shatterhand]

Years later I can see that what was missing is that frenetic 'controlled burst' button mashing combined with the technical depth of the gameplay – you'll miss it if you're just looking for a button masher, but there's a lot under the hood that makes the arcade the game it is. It has a core that is similar to Street Fighter 2 – timed combos that are totally simplified for Final Fight, frames of invulnerability built in to certain moves and a great move set. But 6 months just isn't enough time to get all of that in, especially if you were not used to the game in the first place.

That really sums it up. I always think its absoultely messy how you guys had to port games without any support from the original developers. The more I read about making games at those times, at least on Europe, the more I understand how everything was borderline amateurish. Not that the programmers weren't talented, but "the guys on suits" surely had no idea of what they were doing. And it's amazing that even being like that, Europe had a huge flux of awesome games back at that time.

[dlfrsilver]

Yes. By working in the IT industry (not the video-game side that is), buying a licence without asking and getting the original assets is a joke, and in this regard, i can't believe how amateurish most companies were.

Look what elite did with Ghost'n'goblins, they good the arcade graphics from Capcom, hence the great port we had. I guess they asked for it to Capcom, and they got them.

US Gold never asked for the original material to Capcom, this up to SSF2, where Capcom gave them the all assets, as well as the coin-op source code ! Because most capcom programs have an incredibly complex game logic, that can't be reproduced since it's in fact overly-engineered from the start, you need the source material, otherwise you're doomed to make a "bad" game. In this regard, Rich is not faulty at all, since he was just executing the job they gave him.

[turrican3]

Mr Chaplin,
how much time did you have to make the game ??? Do you think it could have been possible to make a 32 colours amiga version or it would have been too slow ?? Or was it just a way to make a faster port from the st ??
Could you explain us what was the most difficult part in coding this game ??
Like said leathered, this is true that technically finalfight looks good (sprites sizes, animations,gameplay, and scrolling), but visually with just 16 colours it doesn't look so good.
In short : why did you use 16 colours for the amiga version ??

[RichAplin]

If most people are feeling it's the gameplay and lack of music in Amiga FF that's lacking; actually they'd both be relatively straightforward to modify just by hacking it; adding in-game music (assuming extra chip RAM) would be really easy (just have to sort out the priority of music notes vs in-game sfx).
Changing the gameplay; I really don't think it'd be that hard to just chop out the player and AI control and replace it; I'd imagine I did the character control in a reasonably generic way (i.e. it'd be easy IIRC to switch your character to play as a baddie, albeit with differences in available moves/sprite frames), and stuff like the timing+positioning of the animation sequences, collision boxes, how long it takes to recover from a hit, etc would have all been driven with simple tables...
There's zero chance I still have the source code, but it was all written in assembler and by that time I couldn't be bothered with putting in complex copy protection or anything (BTW I really enjoyed reading the "how to crack Double Dragon 2" article that someone wrote) so you could probably run it through IDA or something and get something fairly useful...

[RichAplin]

Originally Posted by turrican3

Mr Chaplin,
how much time did you have to make the game ???

Roughly six months, including time spent hacking the original PCB to get the gfx out and converting the sprites (the backgrounds were done by someone else using a frame-grabber I think), plus writing the whole thing in assembler from scratch, doing the intro, the "hidden bit", etc,etc. Obviously quite a lot of late nights were involved :-)

Originally Posted by turrican3

Do you think it could have been possible to make a 32 colours amiga version or it would have been too slow ?? Or was it just a way to make a faster port from the st ??

Errr I can't even remember if it's 4 or 5 bitplanes on the Amiga - you'd have to fire up an emulator and see :-) One compromise that would have made sense would have been to use one set of 16 colors for all the sprites and another for the backgrounds (and then change the BG palette per level)... but given that we all knew an ST version was an absolute requirement (and to be also coded by me and released at the same time!) I expect that stuff got ruled out. It would obviously have used more RAM and been a bit slower also.
Someone want to check if FF is 4 or 5 bitplanes? :-) [Edit; yeah 4 planes for the whole thing; must've been the ST requirement]

Originally Posted by turrican3

Could you explain us what was the most difficult part in coding this game ??

Putting up with the abuse, decades later :-)

Errr I don't know it was actually probably my fave 16-bit game to work on; I was exceedingly happy to have successfully ripped the arcade GFX. Stuff like the dynamic-defragging memory management, background loading+decompression, things like that were all pleasing to get working. Quite a lot of geek enjoyment when doing it.

It's worth pointing out that (as was typical) I was pretty much just shipped an arcade PCB (...that's all, I hooked it up to a 1084 monitor+homemade joystick board) and left to get on with it. My parent company (Creative Materials) had people doing the graphics in their own unsophisticated way (which IIRC was using a frame grabber on the arcade PCB and then recoloring/chopping it all to remake the maps and sprites).
Due to the large amount of labor involved in that there was zero chance of them doing separate ST and Amiga graphics sets.

They were really surprised when I told them I'd successfully ripped all the sprites from the FF PCB. I'd have loved to use the original backgrounds too but merging the parallax layers and the color issues were too much to solve; especially as I wasn't supposed to be spending any time on ripping stuff anyway....

So after sorting out all the ripping and gfx converting, and deciding which sprite frames to drop and so on, there was the simple matter of rewriting all the game logic and gameplay data from scratch...
It would have been really awesome to have reused code/program data from the PCB, and I looked at it for a day or two before realizing I was running out of time rapidly and it was a ridiculously risky thing to pursue...

I still think that was the right decision; having much more sophisticated hardware tools (logic analyzers, ROM emulators, etc, which I did have later at Codemasters) - or of course MAME(!!) would have made it more within the realms of possibility, although still a very risky option.
Really I'd have had to fully disassemble the entire game code and data blocks, figure out all the h/w (actually that wasn't too hard; did a lot of it) and then isolate the gameplay logic, rebuilt it to use a lot less sprite frames.. etc..etc.. etc...
MAME would be the ideal tool for the job of course.


..in fact that would make rather a nice project for someone now; rip the original gameplay code+data out of the arcade roms and splice it into AGA FF.... Go on! I dare you!

[RichAplin]

Heh just fired up MAME and the "TEST MENU" for FFIGHT (hit F3, it'll reboot, then hit CTRL and ALT) has a menu option "Character test" where you can scroll through all the sprite tiles... there are 0x2200 16x16 tiles, each in 16 colors = 1MB data; you can also switch palettes through the 32 palettes used (it appears to use all 32 palettes).
Also pressing 'A' lets you cycle through the "Stage" (i.e. game level) and a couple of the palettes change with that also.

You can also see the other stuff like the 8x8 "Scroll 1" top layer map (character sets plus a bunch of the intro gfx) there's 2048 of those, plus "Scroll 2" (foreground, 4096 16x16 tiles = 512k data) plus "Scroll 3" (640 32x32 tiles = 327k data)

It's pretty fun; you can also see (in Scroll 2 and 3) at the end some level-editing tiles (for showing collision maps I assume) and joke things they put in

It doesn't let you scroll around the level maps but it's interesting nonetheless.

The "Output Test" menu lets you scroll through the sound fx (handy for sampling if you need to)

Heh cute..

Also fired up FF on WinUAE - reminded me it's 4-bitplane (16 colors). Yeah, it's not terribly fun is it. I didn't much enjoy it on MAME either but that's IMHO :-)

[dlfrsilver]

i got a look today on Mega Twins from Capcom. Well this one has also MANY linked sprites, which are really complex.

The CPS-1 hardware has the ability to superpose tiles, this way some sprites are made of multiple tiles managed by priorities....

About CPS-1, do you remember how the palettes were stored for the whole game ? Was it in a table ?

[Retro1234]

So from the test menu you can get everything cool

@ dlfrsilver can you upload the tile set and map data for level 1 I want to have a look.

[dlfrsilver]

i have not ripped the tile set and map data for level 1. Only did sprites, and i ripped perfectly the background from mame tilemaps

[Retro1234]

Thank you for your reply and good luck with the project

[turrican3]

Originally Posted by RichAplin

Due to the large amount of labor involved in that there was zero chance of them doing separate ST and Amiga graphics sets.

I think this is why some arcade ports had bad publicity in the amiga community, but societies had to win money, the rights for a so much big arcade hit was surely high and they had to make profits first, doing separate ST and Amiga graphics sets was surely financially risky. But it's felt like a betrayal to have an amiga version just a little better than the st version when you compare them to the best amiga 500 natives games. This is of course not the fault of the artists who worked on the conversions they had to follow the decisions of the publishers and these societies was there to make profits and like the atari st was the 1st market at this time, it leads to not fully optimized amiga ports. But i often ask myself what could have been theses games if they were ported 1st for the amiga in mind ??? I think, this is why, it's felt like a mess. But this how the world is done,the profits first, and it's always the case today and perhaps even more.
But you know that when you are fanatic about a sytem, you don't think about that and you just feel bad about some atari st -> amiga ports and i don't think first about your games Mr Aplin when i told this, like i said i like double dragon 2 or line of fire, at this time it was good games, i always had good feelings about thses games... It's just of course my pov, some will agree, others won't.
ps: what happens to the double dragon 1 gfx !!! the game is good, but the gfx ??? what happened to them ??? Thanks for your answers.

[dlfrsilver]

In fact, if any one wants to convert these arcade games, you need an editor, to simply and gain speed, because there are too many tiles to deal with. That's the problem they met....

[leathered]

Originally Posted by Boo Boo

So from the test menu you can get everything cool

@ dlfrsilver can you upload the tile set and map data for level 1 I want to have a look.

I don't think it was really useful in such a way - certainly I couldn't get the tiles. It had some use for checking animations, and noting the hit boxes. I cannot actually say what the Capcom engineers used it for exactly, but it seems likely that it was purely for auditing purposes. As I recall there are 2 test screens - I think accessed by holding either of the start buttons, and maybe even by selecting audit from the mame menu first.
As Rich says it could be used to record the sounds - there are a few left to do, Guy has at least 1 voice sample I need to get - his kiai when doing a forward flying kick.

We got the tiles from a variety of sources, including mame ripping from Turri and Denis, but by and large these backgrounds have now been ripped on spritedatabase.net - which was not the case several months ago.

I converted these back into tiles after re-colouring and mapped them again.
I say that as if the job is finished - it isn't.

Originally Posted by RichAplin

So after sorting out all the ripping and gfx converting, and deciding which sprite frames to drop and so on, there was the simple matter of rewriting all the game logic and gameplay data from scratch...
It would have been really awesome to have reused code/program data from the PCB, and I looked at it for a day or two before realizing I was running out of time rapidly and it was a ridiculously risky thing to pursue...

I still think that was the right decision; having much more sophisticated hardware tools (logic analyzers, ROM emulators, etc, which I did have later at Codemasters) - or of course MAME(!!) would have made it more within the realms of possibility, although still a very risky option.
Really I'd have had to fully disassemble the entire game code and data blocks, figure out all the h/w (actually that wasn't too hard; did a lot of it) and then isolate the gameplay logic, rebuilt it to use a lot less sprite frames.. etc..etc.. etc...
MAME would be the ideal tool for the job of course.


..in fact that would make rather a nice project for someone now; rip the original gameplay code+data out of the arcade roms and splice it into AGA FF.... Go on! I dare you!

Lol, it seems that the way you used the tile data is already much more like the arcade - maybe you could splice the data into the original conversion!
I'd love to get hold of the data, and perhaps I could implement some of it in Blitz. But then I'm some way in now - I'd say the game is over 50% complete, with most of the remaining work being in putting all the remaining gfx in and scripting the boss fights, sorting out the individual AI.

If someone beat me to it and managed to do a completely new version of the game using the data and gfx from the original Amiga conversion then that would be awesome! I'm sure I'd feel relieved, and I've still got my engine - written from scratch. Free, I tell ya... free on the wind! (Torn polythene bag) =).

[tomcat666]

Rich: Back in the day my brother was a real beat-em-up fan and he ABSOLUTELY LOVED your Final Fight. The game was one of the most played games on our A500... So anyone talking dirt about it must be out of their minds... it was one of the best games on the Amiga, period. Huge sprites, loading in the background, everything was just great really. It is logical that some of the Arcade features had to go together with a lot of the colours, but what mattered was that the game was really playable and it still looked great.

[RichAplin]

Originally Posted by turrican3

But it's felt like a betrayal to have an amiga version just a little better than the st version when you compare them to the best amiga 500 natives games.

Sure, it'd have been better for everone if the ST had never existed. :-) The ST was unfortunately far too similar to the Amiga in terms of h/w for most companies not to use the same GFX (and often 99% the same code)

It'd slightly disagree (in terms of my games) with "not fully optimized Amiga ports" - I loved the Amiga and always did that version first - apart from the GFX color depth Amiga FF took full advantage of the Amiga (OCS) hardware. Did I miss anything? Dual playfield was useless as a gfx mode, and hardware sprites equally useless for this kind of game. Everything else got used to the maximum extent possible (incl 1MB ram and even a 2nd disk drive). Shame about the in-game music; with hindsight I'd have really pushed harder for that.

Originally Posted by turrican3

what happens to the double dragon 1 gfx !!! the game is good, but the gfx ??? what happened to them ???

The artists used a (monochrome) frame grabber on the arcade board, and manually tidied them up and recolored them. Lame - but it was all business.

Me ripping the FF graphics was something I did entirely because I wanted to (being frustrated with the results of artists doing it by hand).

I think that's the thing to remember; arcade conversions were 100% business for everyone in the hierarchy - except most of the artists and programmers, who basically did it for the love of working on games (and a pretty meager salary).

With the original PCBs the games had to be fun to get you to play+pay more, sales of the arcade PCBs were directly related to their weekly earnings in the arcades (stats were published in a trade magazine and very well known by arcade owners - all machines have built in accounting for example) but with conversions the customer paid just once so... that's capitalism for ya :-(

It would have been nice to have Amiga FF run in at least 32 color mode and use separate sets of 16 colors for the sprites and backgrounds - in fact I vaguely recall I looked at doing that but... didn't happen.

Another example; certainly nobody ever even asked me to implement the attract sequence for FF - in fact if I'd said I was doing it they'd have said "no, no, just get on with the game", I just implemented it (perfectly, including the "Winners Don't Use Drugs"!) because I wanted to. Similarly, it was me that 'borrowed' the (awesome) title music from a demo - Jolyon Myers didn't even get paid for it (!) although I tried my best, and obviously gave him a credit in the game. I believe there's no hard feelings from him on that.

But yeah, original games were certainly much better tweaked for the Amiga (e.g. Shadow Of The Beast etc having practically every aspect of the game custom-designed around the hardware limitations)

Of course if I did Final Fight for modern hardware I'd simply emulate the original board; Arcade Emulators are (were) tremendous fun to write.

[Adrian Browne]

I seem to remember, all those years ago, when I got the 1 megabyte upgrade for my a500, the intro music magically appeared.
I tried the whdload version last night and oddly there was no music.