The only reason I can see for the savestates being implemented in bsnes is for hacking purposes. I know that when I'm testing a game and someone finds a bug 3 hours into it, I don't want to play through 3 hours to find a bug that takes all of three minutes to fix when I could just load a save state before the bug and save 3 hours in the process.

Let me just say, I really admire what byuu's doing with bsnes. If I could actually code worth a shit, bsnes has almost inspired me to *try* to make a Genesis emulator in the same vain.

And just to reiterate myself in case I wasn't clear.

Yeah, we definitely aren't competing. I wouldn't be giving away all of my new findings and helping to fix bugs on the ZSNES bug reports forum if we were competing. Likewise, my DSP core and mode7 formula is from anomie/SNES9x, additional IRQ info from anomie and TRAC/SNEeSe, color add/sub help from GIGO/SNESGT, etc etc.

All of the different emulators equate to two things:
1) Some of us can't work in teams. There are many different reasons for this.
2) Some of us have different opinions of where to draw the line between compatibility, speed and accuracy; but we're all working toward the same ultimate goal. All emulators are just different paths to get to the same place in the end.

Speaking of which, I finally convinced pagefault to remove the OAM addr stuff I mentioned a while back, leaving me with nothing to openly bitch about ;_;
You can all make threats on my life instead of his when Uniracers stops working, heh. Though I still think a game-specific hack should be added for it or something, at least. I was just bitching because it affected all games, and wasn't counted as a hack :(


<perfectionist_nazi>
Why does everyone call the S-DSP the SPC? x.x
</perfectionist_nazi>

I think we will cut down on the WIP's since we have zget now which downloads the source and compiles for people. This will stop the confusion caused by the WIP releases, do you think this would be better?

We plan to fix as much as possible and continue to work on it as long as the developers have interest in doing it, so far we expect to be working on it for a long time. I can't say if some things will never be fixed since so much has to be fixed we have no order or priority in what to fix so we are going to work on general things to begin then address specific issues. We are rewriting a lot of the code anyway so it's being done properly (at least to standards today). I have much respect for the rom hacking community if they have an issue they would like fixed I would be more than happy to escalate that problem so it is fixed so they can release whatever they are working on.

We really do not have any organization at this point, but we are getting better at it and assigning things to people to work on so hopefully this will get much better. I will talk to Nach about what he says to the end users, it's not their fault they aren't using a certain date or build because the game doesn't work. That just isn't right. I think we are going to aim for some kind of service pack for 1.42 in January then another release in June with the major changes. We have a lot of problems we have discovered in the code like hidden hacks and assembly that steps out of bounds causing random results so we have a lot on our hands to get a decent release out.

I am completely open to any input you have you think could help us improve our project and the scene in general. I just would like everyone to get along and be happy, users, rom hackers and coders alike. Lets not turn this into what has happened to the xbox scene.

I also forgot to mention _Demo_ is the only one with access to zsnes.com and he is never around so we can never update the site with info... I wish we really could.

It was like that when we got the code, it's being ported to C as we speak. But yes it was not a good decision to write it like that.

I don't know. I think I'd rather see either ALL WIP releases or more frequent official releases. I still keep getting stuck on the fact that ZSNES hasn't been updated in 2 years officially. That just doesn't sit right with me. I think that's the root of the issue. Even if you cut down the WIP releases, you still have an old outdated official version which is the only thing on the zsnes page and then you have WIP releases. The two could coexist just fine if there was just more frequent official updates. The problems from 2 years of advancements and no advancements from the last official build. The changelog is ENORMOUS since then.  I agree with creaothceann's mentality here.

I don't don't know why you think we have trouble getting along. I have no problems with you or Nach personally. I don't like  the direction ZSNES is going with it's releases and some things on the forum, and I've been a bit vocal about it, but that's about the extent  of it. This site was made to bring the ROM hacking community together. I like unity.

You want my opinion. Here are some things I think you guys should do.

1. Get a hold of demo and assume control of zsnes.com. It's really sad that nobody can even update the zsnes page. demo doesn't care about it anymore, so go set up a page and tell him to point the domain to a page with somebody who DOES care enough to update the site again. This is a major problem. THIS is where people go to look for newer versions of ZSNES. Most people who hae the last official build got it from there or another archive site that got it from there.  A website is very important if you ask me. A simple news update a few times a year can go a long way with keeping the public informed and letting people know you guys are working hard. YOu are working hard, but not many people know it.

2. I know you guys have tried a few different things with bug reports. Are you still actively using a bug tracking system? I really think you need to do this and be proactive on bug reports. Let people know you care about them. Log them maybe even comment on why something hasn't been fixed. You even convinced me that you guys DON'T care about bug reports. I reported that same Wozz bug FOUR years in a row, even provided information at one time and it seemed like it was ignored until I 'annoyed' you guys enough with it. This isn't how it should be at all. I see this happening with other bugs as well. So get things together with the bug tracker and let people know it's there and to look there first before reporting a bug. And you may want to post comments on bugs on why they haven't been addressed whether it's because you don't know how to fix it, low priority, or whatever the reason.

3. Maybe it's because only 2 of you on the ZSNES team are capable of making any real core changes, but it often seems like ZSNES cares more about the fluff and less about what counts in the emulator. Plenty of development time is spent on things that seem to have little to do with the main emulator. Again, if only one or two devs are the only people that can even make core changes, then that's life and understandable. But it seems an awful lot of focus is there for the fluffy things. Although I don't think YOU personally have been working on any of that stuff.

I guess alot of it is a perception thing. Maybe it's because you guys don't communicate that well to the public what is going on to tell the REAL story. Nach loves to say plenty, but it's usually on all the things he's doing which is usually nothing directly related to the core emulator. I think some periodic updates on the ZSNES website would really help.

Let people know what the REAL ZSNES story is. Nobody aside from some forum posts really knows what's going on and just make up their own stories based on the perception they get. Set the record straight once in awhile with official statement.

That's my two cents. All of this post was meant to be constructive. I can't judge my tone of voice well tonight.

As far as I can see, pagefault is the only one who can make serious core changes. Jonas Quinn can at least touch the memory mapping stuff and things like that. Nach has openly stated that he has no experience with writing emulators, so of course he works on features instead. It's too bad, he's really intelligent, did a great job with Cx4, etc. I think he could do really well on the core. The difficulty of writing an emulator is just like the difficulty of assembly ROM hacking: completely overrated. The only two things you need are a brain, and time.

As far as complaining about broken savestates, geez. Of course they break. The only reason ZSNES states haven't broken between recent versions from a few years back is because little to no work on the core was going on. A savestate is basically a glorified dump of all of the internal variables used by the emulator. Make any changes and now you have missing, as well as duplicate, information in your old savestates. You can get away with some of it. But eventually, there's just too many differences and they no longer work.

Save RAM? Yeah, that should work fine. No idea what's going on for you there. Sounds like an isolated problem, at least.

I have moved this thread to our board so it doesn't get hijacked any more than it has. Since this was about a debugger at one point.

http://board.zsnes.com/phpBB2/viewtopic.php?p=131049#131049