Proven Concepts

I'm sure plenty of other people could do everything Stealth has done. Honestly I've been thinking about it for quite a while. Note what Tweaker said; I've been positive a C source conversion would be completely possible, and if Stealth hadn't done this, I was totally going to. My idea was more to make platform independent portable code (which might be what Stealth has done, though it's hard to tell without a source release) to enable porting the Sonic engine arbitrarily to virtually any capable platform. Obviously would require wrappers and hookup code, but if he already got it this far, all that needs to be done is to move Genesis specific function calls off into separate, easily replaceable files. I guess it's all moot to talk about it, since he'll likely "never" release it, given his standard.

I think it would be less about trying to sell it and more like "You see, guys, we do better work than a professional company. :v:"

 

Why didn't stealth port knuckles and tails to the sega cd version instead of the other way around?

 

Stealth should send a DS version of all this to SEGA, so they could hire him to make Sonic Jam DS and make a profit.

A quite big profit, since all of the old school Sonic fans would buy it.

 

(I dunno. I already have the original games, so I don't see why I would buy them again.)

 

You know, I was initially trying to make your exact point and then I kind of regressed for a bit. What I was trying to say moreso is that if someone had the skill the convert the Sonic CD levels into a Genesis format, they'd have the skill to do it all on their own, as Stealth did. All the resources are there, and anyone who chose to put in the required amount of skill and effort could achieve the same exact thing. What I was trying to imply is that, so far, it appears nobody else has been willing to step up and do so aside from Stealth.

You seem like you're capable of a lot of the things he is. Is there a specific reason you choose not to do things like this? (out of curiosity; I'm not trying to start anything here)

Don't you think he has a right to retain his own work? Not everybody chooses to, but I think we should at least respect the people who do. Some people never consider that when you do something, nobody is entitled to it—Stealth said in his post he released it because he thought some people would find it interesting and/or inspirational. If you saw something was possible through a proof of concept like this, wouldn't you want to try something like it yourself? Just a thought.

I, um, erm... because he didn't? :v:

 

This is very impressive. The method you used to port to GBA sounds similar to how the genesis games in Phantasy Star Collection were ported. (I recall an interview somewhere mentioning that they wrote a tool to convert the assembly to C macros) I had considered doing something similar to what you did with a different game, but hadn't put much effort into the idea.

I'd love to see more work like this in the future and hope you have plans for it. Also, if you've developed any tools or source that could help people do similar ports I hope that you could consider making some of that information public. Great work.

 

He'd be more extremely awesome.

 

The sounds they picked seemed to lend themselves better to not getting as affected by the static-y sound quality that I've come to associate with the music of the GBA and DS. (Note: I'm only referring to GHZ and the Special Stage music here.)

I will agree on SFX, though. They don't sound better, in addition to sounding different.

 

Well since they already released Sonic 1 on the GBA, maybe he could do a Sonic 2 port and win Sega over. Then again GBA games are few and far between. As everyone else has been saying, beautiful work!

 

Just didn't get there. :v: It wasn't that I chose not to do it, it was more that I was thinking of doing it but hadn't committed to it yet. Which is why I had Quex ask you if anyone was already doing it -- I was thinking about it, but certainly didn't want to step on anyone's toes. But then this came out which pretty much answered the question. Makes me wonder if other "secret" ports are being done and if I'd really be reinventing the wheel. (Why it's probably better to be public about such a project as it's being developed, so others don't create redundant work OR know that they won't be out of place.)

You misunderstood what I meant. I meant that it was moot to discuss how his source code could be reworked into platform independence because he keeps it to himself. I don't care that he keeps it to himself. I wish he wouldn't, because it could lead to the abolishing of folks being forced into tedious assembler as well as revolutionize those who want to make fangames. Including Sonic 2 HD or such projects.

I mean I could do so much more so much quicker with the Sonic engine if I reworked it into C. In its assembler form, I keep hitting subtle bugs related to screwing up data alignment, registers being altered in other function calls, and just lots of weird ... things happening. Usually I eventually find the cause, as any good programmer/debugger would. But it's tedious, especially having nothing better than the Gens K-Mod debugger. Writing it in C puts a lot of that alignment responsibility and other quirks at the compiler level, including standardized function calls and other wonderful things. I have no idea how well the Sonic Engine would perform on the Genesis itself if rewritten in C, but I can't imagine it'd be too much worse. I acknowledge that certain routines, particularly H-BLANK critical ones, would probably have to remain in assembler. But that's part of the wrapping game.

And, of course, much of the source remains undocumented (at least, in Sonic 2 that I'm working with, I don't know about the Sonic 1 disassembly.) So my additional responsibility includes commenting what a function even DOES. Although I am quite proud of myself for decoding some as advanced as the object placement routine, but again, I haven't looked at S1 and maybe that was done already. It did lead me to find what apparently is an age-old Sega bug that never come up due to "perfect" data layout. But imagine a full C implement of the Sonic Engine -- likely almost every necessary function would have to be understood, and then well-named and well commented! Even with Stealth's thing here, if he has no plans to release it, I think I'm up for a competing project, especially with the realization that it is, obviously, completely doable. If anyone can confirm that I wouldn't be toe-stepping, I'd be inclined to start organizing it.

Re: Point B -- isn't it actually that the original assembler source has been lost? Or that's what I remember hearing.

But for the most part it's pretty indicative of where Sega's "Sonic" priorities are these days. I think they're trying really hard to leave the past in the past and move on with their "new" Sonic.

 

You might wanna fix the SCD link. I keep getting an error on it.

 

Just tried this on an actual DS for shits and giggles and it's amazing. I ran into a few visual bugs last night on VBA and they aren't there on the DS. Awesome awesome awesome!

 

Hay guys. I am a trial member.

But seriously, press L+Start on tittle screen. NOW.
EDIT: Why did I waste a post? D:

 

Just go into the address bar, highlight it, and press enter.


It Worked For Me! (SM)

 

Good question.. were you trying to point out the debug mode?

 

It's not debug. :v:

 

It's a pity no more work is planned on the GBA port. :v:

 

Did this get sped up even beyond the original or does it feel faster because of the low resolution?

 

No it's no... holy shit stealth, you outdid yourself.