Snes9X 1.42 wip

[Nach]

Let's take a simple, made-up problem for a moment. Game A is really popular, and requires register 34 to return a constant 7. Game B was only released in Japan, and expects many different values, but never 7, and indeed, the source is register 34. On the other hand, you have the GB42 chip that runs some random game. Right now, the GB42 would be focused on. That's the state of emulation. For the majority of both users and developers, register 34 just is not as interesting.

This is just as true now as ever. Only we're really reaching deep into the end of the barrel. What's the focal chip being worked on now? ST-011. Which is only used in one game, "Quick-move Shogi Match with Nidan Rank-holder Morita". In the mean time, not a single SNES emulator even gets basic SNES IRQs emulated correctly yet, so scores of games break every time any change is made in core timing.
Don't get me wrong: I absolutely appreciate the very hard work that's being put into these chips by everyone, but it's definitely the wrong priority to focus on one specific game when so many others have such serious problems in my opinion. But then, everyone's free to work on whatever they like, and some people just happen to prefer emulating new chips. Yeah, it's a lot more fun and visibly rewarding, exactly as MKendora was saying.
The good news is, there's only maybe 1-3 chips left. ST-018 being the only one I can even think of off the top of my head. I guess we'll find out how right or wrong MKendora was when the last few chips are finally emulated here in the near future.

This is now completely different than it was before. Back when MK said it, people like Overload (among others) were working on RE'ing some logic in a special chip. Half of these people were able to run any tests they wanted and RE anything they wanted. For them to work on RE'ing a chip as opposed to based hardware is somewhat a waste.

However we have logs online people captured just letting special chip games run (and we should have more of these tests posted), and people who are good with math and other things decided to RE them. The Seta 11 as you pointed out, who RE'd it, you? Overload? anomie? no, merely someone who is good at RE'ing and looked at logs. Did you provide any alternative logs to be RE'd from the base system? Not that I know of. What do you expect our friendly contributors to work on? So I got from a nice person some code and spent some time with Jonas merging it in (which also did not take any time away from other parts, was simply a matter of binding some memory addresses). If you think it's wrong that people who have no connection to any of the emulators spend all their time time RE'ing the wrong stuff, why don't you post logs of something else? You have only yourself to blame here.

[bobthebuilder]

While I don't know Gary, I'll agree that anomie was easily my superior. Both in intelligence and SNES hardware knowledge. And now his time with us is very, very limited. And now I'm burning out quickly, but still trying my best to hang on. Who's going to replace me? Nobody at the moment seems to be interested in the insane kinds of tests myself and anomie were running. But someone will come along, most definitely. Who probably won't make it as far as I have before leaving.

Why? You have written a fairly well organized/clean emulator, this alone will help others go farther. Perhaps, you could comment the hell out of your code or document as many of your findings as possible (like anomie has), if the loss of info is a major concern for you. Anyways how much farther can emulation go, that is going to be worthwhile. You have a fairly high compatibility rate as it is, and once all the officially released games can be played, only rom hackers and snes programmers will care about further improvements on accuracy that are transparent to the average user.

Sound familiar? If I were paid only a dollar for all the people who discredit my work as inferior and won't even touch my emulator because it lacks (choose one or more): savestates, SA-1 support, SFX support, DSP-4 support, 60fps on a 386 support, ... I could retire from work and have the extra time needed to add all of that by myself.

I wonder who are the crack addicts that discredit your work. Your emu is the only one I use on my primary computer. I was one of the people who fully agreed with MK, but now that your emu is around we can have the best of both worlds (ZSNES that specializes in speed, Snes9x that specializes in portability, and bsnes that specializes in accuracy ). Although I have to give bsnes props on its portability and speed, considering that its primary focus is accuracy.

I really don't see why he's so hated; but again, I wasn't there. So whatever. Hopefully there's a lot I don't know about, otherwise I'll be just as despised when I eventually leave and everyone realizes how similar our viewpoints were.

I couldn't have gotten as far as I did if people were responding to my rants with as much hatred as eg Evan threw toward him.

I liked MK alot and agreed with his points, unfortunately he began behaving like an ass towards people at the end. That thread is only the tip of the iceberg. I have never seen you behave like MK and hope you never do act like him. When you leave I hope it is on good terms like Gary and Zsknight.

[byuu]

However we have logs online people captured just letting special chip games run (and we should have more of these tests posted), and people who are good with math and other things decided to RE them. The Seta 11 as you pointed out, who RE'd it, you? Overload? anomie?

I know who RE'd it, but I wasn't aware it was done entirely from logs. Holy hell that's insane, that guy is a genius. I wonder if he could crack the SPC7110 decompression using the available logs for that ...
Well, that definitely changes my opinion on that, I was under the impression he has special testing hardware like one of the SWC DX2s or whatever that allow you to probe cart addresses using test programs.

As far as my own logs, I've posted NMI timing logs. I ended up passing those tests on my own, however. I also posted OAM mid-frame address logs, but the results were mostly apparent anyway. There isn't much that can be tested with logs that I can't decipher myself. One notable exception was DMA sync timing logs, which anomie figured out.

I wonder who are the crack addicts that discredit your work.

Mostly end users, a lot of times non-English speakers who probably can't read what the emulator is about. But quite a few people I semi-respect as well. No big deal, everyone's entitled to their opinion. I wouldn't use an emu that didn't get 60fps on my PC, either. Wouldn't make sense to, but that doesn't mean it's bad. Eh, whatever.

I liked MK alot and agreed with his points, unfortunately he began behaving like an ass towards people at the end. That thread is only the tip of the iceberg. I have never seen you behave like MK and hope you never do act like him. When you leave I hope it is on good terms like Gary and Zsknight.

Hmm, well I do publically chastise other emu authors for using hacks, but I always offer help and test ROMs to remove said hacks. Though I really shouldn't make my latest findings such a big deal, but generally if I don't they tend to slip under the carpet and never get fixed. Which is okay too, not everyone has the same priorities as me.

I've no intentions on leaving on a sour note. I really appreciate speaking with everyone in the SNES dev scene, they're all really great people. It's really cool being able to work together as one big team to fix problems, despite technically being involved in separate projects.

But that's getting off topic again. The only thing I noticed was at the end MK was pretty adamant about people not asking him to keep working on SNES9x, but then everyone else kept telling him that he secretly wanted to work on it still, so who knows. I really shouldn't speculate since I wasn't there.

[funkyass]

MK, at the end, was sick and tired of it; not of working on snes9x, but of working with others, and leading them.

[Snark]

Sound familiar? If I were paid only a dollar for all the people who discredit my work as inferior and won't even touch my emulator because it lacks (choose one or more): savestates, SA-1 support, SFX support, DSP-4 support, 60fps on a 386 support, ... I could retire from work and have the extra time needed to add all of that by myself.

As long as you don't forget all those who are approve of your emu.

And all features you mentionned are all eventually possible (save for the 60fps on a low-end PC of course)

Unfortunately, I've never spoken directly with MKendora. He left a few months before I decided to step in. In the only indirect communication I've had with him, he was slightly condescending, but no more than I am toward people now. I really don't see why he's so hated; but again, I wasn't there. So whatever. Hopefully there's a lot I don't know about, otherwise I'll be just as despised when I eventually leave and everyone realizes how similar our viewpoints were.

Doubt that could actually happen. Key difference is that you actually went out there and code the emulator.

And finally, I'm absolutely certain not even bsnes lives up to MKendora's expectations.

How so? Maybe your accuracy standards are actually even higher than his was. Though it's a good question, but in the end who cares if MK would approve or not?

I'm just as disappointed by it as well. But that's another story. I'll probably write up an article on that on my website here in a few days/weeks/decades, though I kind of touched on it already re: S-CPU hardware multiplication cycles.

The only thing I truly find eerie about that rant, is the fact I knew absolutely nothing about MKendora or that rant until after I wrote up my own rant (which is now lost after the2d.com died), and made significant progress on doing exactly what he was talking about before realizing someone out there even shared my viewpoint. In fact, I probably don't give him nearly enough credit for being the first to get a little public support behind the idea of a hardware accurate emulator. I couldn't have gotten as far as I did if people were responding to my rants with as much hatred as eg Evan threw toward him.

Again, goes back to what we were saying: You actually made the thing, not just talked about it. So I can't see how that could have happened.

I wonder who are the crack addicts that discredit your work.

¿Por qué el Kombat Mortal no juega VELOCIDAD completa en bsnes con mi PC 386?? dígame ahora o le secuestraré y pediré un rescate!!! (translated from babelfish so it probably doesn't make much sense)

[Nightcrawler]

And now I'm burning out quickly, but still trying my best to hang on.

Why is this happening to you? And for that matter, why does this happen to anyone?

Mmm, the craziness of something else broken on a change or fix... it gets tiresome quickly. You can't believe the kinds of stress that must be released when the obvious stares in your face, but mocks you because something else is preventing it to "just work nicely".

I think it's just because people don't take enough breaks. I've been working on some projects for many yeas. I've been burnt out on them several times. However, you just kick them aside for however long it takes for you to have renewed interest and pick it back up again.

Byuu has been working on BSNES nonstop for a long time now. He's probably overdue for a lengthy break.

In fact, by NOT taking breaks, you could potentially worsen the situation. You're inner attitude and burnout will show through and you'll make mistakes, cut some corners and potentially reduce the quality of what you're making.


Byuu:

It might be worthwhile to release all of your little probe/test ROMs to the public. That way other people can use and run them freely who may be interesting in improving other emulators or something along those lines.

Nach made an interesting point there. How are other people supposed to fix things up, contribute to more important goals etc, when they don't have the same information, logs, or tools that you do?

You now have ALOT of information on SNES hardware test ROMs, and resutl sets that not many do. If you don't make it public, how can you expect anybody to take your place or do as good or better of a job later?

The way we progress as humans is combined knowledge. Somebody builds on top of someone else's ideas, knowledge, and creations to make it better. You can't keep it all to yourself and expect progression. And you can't just help out the authors who request it. They have a term for that. It's called 'tribal knowledge'. That doesn't work in the long run.

It's obviously a huge unrewarding task, but this stuff needs to be made public if you expect such things.

[adventure_of_link]

¿Por qué el Kombat Mortal no juega VELOCIDAD completa en bsnes con mi PC 386?? dígame ahora o le secuestraré y pediré un rescate!!! (translated from babelfish so it probably doesn't make much sense)

what

[Silensho]

¿Por qué el Kombat Mortal no juega VELOCIDAD completa en bsnes con mi PC 386?? dígame ahora o le secuestraré y pediré un rescate!!!

Roughly translated as "Why doesn't Mortal Kombat run at full speed using bsnes with my 386 PC?? Tell me now or I'll kidnap you and demand a ransom!!!"

[byuu]

Well, thanks everyone. I'm not really worried about it or anything, just ranting a little. And to think, I take criticism a million times better now than I used to with my fan translations in the past :)

I do need to take a break already, though. Maybe I will. Or maybe I can just focus on the GUI again like I did for v0.016 or whatever. A few months of getting a very nice cross-platform GUI up and running would be a good thing, and help with the possibility of a PS3 Linux port.

As far as sharing my test ROMs, I've shared every single one on numerous occasions. I don't publically host them, but I post them frequently enough, and any devs know how to reach me. I'll be sure to post them publically eventually, but they're rather disorganized. It's either I spend my time documenting, or coding. And coding always wins out on my priority list, partly because I suck at documentation :/

[adventure_of_link]

Thanks Sho

[Deathlike2]

It's either I spend my time documenting, or coding. And coding always wins out on my priority list, partly because I suck at documentation :/

Well, any documentation is better than nothing.

Well, thanks everyone. I'm not really worried about it or anything, just ranting a little. And to think, I take criticism a million times better now than I used to with my fan translations in the past

At least you are keeping it real... I don't know MKendora, but obviously something got to him.

[Joe Camacho]

It's either I spend my time documenting, or coding. And coding always wins out on my priority list, partly because I suck at documentation :/

Well, any documentation is better than nothing.

Is it really hard to study the code? I mean, I know jack-shit of programming and stuff. But Can't you use open the program in an editor and see how it was written as a way to learn?

[Nach]

It's either I spend my time documenting, or coding. And coding always wins out on my priority list, partly because I suck at documentation :/

Well, any documentation is better than nothing.

Is it really hard to study the code? I mean, I know jack-shit of programming and stuff. But Can't you use open the program in an editor and see how it was written as a way to learn?

Sometimes one can study code for an hour an have no clue what it does. Furthermore sometimes one can look at code and not be certain if what they're seeing is a bug, or if it's supposed to be that way for some reason.

[Joe Camacho]

It's either I spend my time documenting, or coding. And coding always wins out on my priority list, partly because I suck at documentation :/

Well, any documentation is better than nothing.

Is it really hard to study the code? I mean, I know jack-shit of programming and stuff. But Can't you use open the program in an editor and see how it was written as a way to learn?

Sometimes one can study code for an hour an have no clue what it does. Furthermore sometimes one can look at code and not be certain if what they're seeing is a bug, or if it's supposed to be that way for some reason.

Well, that sounds hard.

[badinsults]

Mkendora was a whiner who wanted other people to work on the core of snes9x, because he wasn't skilled enough to do it himself, and he was unwilling to learn. There was this guy who worked on the xbox port of snes9x who added a couple of hacks to get some games working (TG3000?), and it set him off. He basically called the snes emulation scene dead, and that there was nothing to revive it. About a month later, Super Sleuth was released. I still have the IRC logs somewhere if anyone wants to have a closer look. While I agree that work into the core emulation is important, I really have no problems with some temporary hacks to get games working because that is what the general end user cares about. In essence, I take the opposite stance of MKendora. But yeah, Byuu takes the same sort of stance as Mkendora, but instead of complaining about it, he did something about it. And that commands respect. Keep up the good work.

[Deathlike2]

It's either I spend my time documenting, or coding. And coding always wins out on my priority list, partly because I suck at documentation :/

Well, any documentation is better than nothing.

Is it really hard to study the code? I mean, I know jack-shit of programming and stuff. But Can't you use open the program in an editor and see how it was written as a way to learn?

Sometimes one can study code for an hour an have no clue what it does. Furthermore sometimes one can look at code and not be certain if what they're seeing is a bug, or if it's supposed to be that way for some reason.

Well, that sounds hard.

It depends on the person... based on their experience. Some people will never get it and others can. There's no simple method of just learning the code. Take the ZSNES code for instance. Certain parts make you wonder what the hell was smoked when it was written (mind you, we are trying to correct that)... some parts are much more complex for the average person... but there are some sections that most programmers should be able to pick up. Something like the core would require you to know something about it. So when it comes down to it, it depends.

[Tallgeese]

Evan, could I see these logs? I'm interested in seeing the 'history', if you will.

[bobthebuilder]

There was this guy who worked on the xbox port of snes9x who added a couple of hacks to get some games working (TG3000?), and it set him off.

I wasn't aware of any core changes that made the xbox port different from the others. I know anomie fixed the OAM and sprite priority in the asm core for lantus to get Uniracers fixed, but I thought that was to keep it in line the c++ core. If Lantus was adding a bunch of hacks not found in the main emulation source code, then that means it was more than a simple port. Any core changes done to improve snes9x, could negatively affect the xbox port . I disagree with ports making changes to emulation code, as it almost always causes problems when updates are done to the main source.

I think it was MK's rudeness to people during his last few weeks in the snes emulation community, that caused several people to welcome his leaving. I still believe it was a big loss for the community and had he behaved with more etiquette nobody would have welcomed his leaving.

[ShadowFX]

Funny that I always kept the following log of one of the longer and interesting posts from MK:
--removed---

how could I have missed that

[bobthebuilder]

That was already posted on the first page .

I think he should have left things as he did with this post and not comeback and add fuel to the fire. http://www.snes9x.com/forum/topic.asp?TOPIC_ID=9808

Honestly, I had a feeling he was getting burnt out prior to his announcement, due to how short he was getting with people. He needed to take a break, but didn't and ended up having a "meltdown".