Can The Amiga CD32 Run Perfect Versions of Capcom CPS1 Games?

[lmimmfn]

Its a shame the amiga blitter doesnt support image flipping, vertical and horizontal, maybe it does but i dont think so.

The chipset is designed with bitplanes to save memory just a shame memory/processing speed couldnt be saved flipping images.

[DanScott]

blitter can flip vertically quite easily, using negative modulo. Don't think it's so easy when blitting interleaved bitplanes though (which was commonly done to reduced CPU setup time for blitter operations)

[Gilbert]

As well as SF2, I think CD32 could also do a great version of Final Fight. Maybe 3 times as good as the A500 one because of it's bigger sprites and extra memory. The music could all play from CD too and it can stream in new data form the CD easily (if it needs it) - because the scrolling stops a lot in that game. Lots of natural pauses.

A500 version :
[ Show youtube player ]

[roondar]

It should be pointed out here that Final Fight doesn't actually use sprites on the Amiga because Amiga sprites have limited control over their priorities with respect to other objects (i.e. bobs). This makes it very hard to use sprites & bobs at the same time. Even just using sprites would be tough as again the Amiga has only limited control over sprite-to-sprite priorities (they're essentially limited purely by which sprite channel you use).

Games like this usually rely on the Blitter on the Amiga as a result.

[sokolovic]

Apparently some people thinks that the A1200 (and the CD32 would have been perfectly capable to make a nearly perfect Street Fighter 2 version (and they prove it) :

[ Show youtube player ]

[roondar]

That video is extremely cool, but I'm not too sure about the whole "nearly perfect" thing. Don't get me wrong, I am very much impressed by it. But at the same time, it's a big step back from the Arcade version in key areas (even though I do really like the extra parallax he added in).

What's hard about judging these kind of demonstrations is that each level would need to be evaluated separately to best fit the hardware. It may be that this particular level fits better than others, or vice versa that this level is actually harder than most.

[sokolovic]

Apparently they are planning to release a full working graphic engine. They won't do the whole game for copyright reasons but maybe some mysterious team will do it, Rygar-style

[mcgeezer]

The hardest thing about game development is getting the bloody things over the line and finished. I’d be the first to say Rygar wasn’t actually finished because it’s littered with bugs in places - but you have to set yourself deadlines other wise they just turn into vapourware and waste time.

Someone posted recently that porting easy games such as Rygar compared to others blah blah blah. After my next game i’m moving on to something else.

My point being that doing something like Arcade perfect ports of street fighter is unrealistic as a full game port.

[roondar]

Originally Posted by mcgeezer

The hardest thing about game development is getting the bloody things over the line and finished. I’d be the first to say Rygar wasn’t actually finished because it’s littered with bugs in places - but you have to set yourself deadlines other wise they just turn into vapourware and waste time.

Someone posted recently that porting easy games such as Rygar compared to others blah blah blah. After my next game i’m moving on to something else.

My point being that doing something like Arcade perfect ports of street fighter is unrealistic as a full game port.

Something else as in no more ports, or something else as in no more games?

Anyway, you're correct of course. Development is always about finding the balance between time and how much gets done. And there's absolutely nothing "easy" about porting arcade games to the Amiga. In Assembly. In your spare time. Usually coding alone, or with perhaps one other person also working in his spare time. Not simple.

[dreadnought]

So perhaps it'd be better to take this engine and try to develop a completely new AGA game? This way we'd get something new, and perhaps without the yoke of "it...must...be.... arcade perfect....arrrgh" it'd allow the devs to be more creative.

[S0ulA55a551n]

surely as well , Chip ram is big consideration not discussed in that video ( which I throughly enjoyed) . Surely you need a lot of frames in memory , or its still going to look jerky ?

[Havie]

Originally Posted by mcgeezer

The hardest thing about game development is getting the bloody things over the line and finished. I’d be the first to say Rygar wasn’t actually finished because it’s littered with bugs in places - but you have to set yourself deadlines other wise they just turn into vapourware and waste time.

Someone posted recently that porting easy games such as Rygar compared to others blah blah blah. After my next game i’m moving on to something else.

My point being that doing something like Arcade perfect ports of street fighter is unrealistic as a full game port.

Absolutely, it's fun starting a game and getting it up and running but then can be tedious finishing stuff and finding all those damn bugs, On top of that, you are constantly thinking about how you can improve things but sometimes you just have to say enough is enough and publish it! Problem is, you then get lots of helpful suggestions to make it better...and mostly they are good suggestions but if you listen to everyone, you'd never actually finish the game!

[Havie]

Originally Posted by sokolovic

Apparently some people thinks that the A1200 (and the CD32 would have been perfectly capable to make a nearly perfect Street Fighter 2 version (and they prove it) :

[ Show youtube player ]

Watched this and it's very interesting - obviously not the only approach as there has been debates about using sprites for one of the characters which gives you 16 colours for each character,

Also, probably proves that you can't do arcade prefect conversions of CPS1 games as despite 'surpassing' the arcade hardware, compromises have been made! The biggest being shrinking the graphics by 25%.

Got me thinking though...

[roondar]

One thing that should be pointed out about the video is that it seems from the discussion about it on YouTube that it's not actually running on AGA (emulated or real), but rather it seems to be a prototype running on modern hardware. It's designed to show what they feel is possible, so they have tried to keep it as accurate as possible.

As for how accurate it is, it seems to me that they did a really good job in taking into account the many, many intricacies, of the AGA/Amiga chipset though there are some minor details that may not have come across exactly and there is the question of combining HW scrolling and sprites on AGA.

Edit: correction, only the last part of the video is not on Amiga HW it seems.

In that version we still uses the 16 color boat from the beginning of the video. while the last part of the video I am showcasing the graphics through a webgl engine mirroring what we have done on the Amiga.