MAMEWorld Forums - EmuChat - Re: The Sky is Falling! Or not.

> > The latest incident with Mamesick just further shows that, we highlight what
> > developers of builds need to do, John IV quietly does what is needed to clear as
> much
> > of the UI code as possible, Mamesick on the other hand starts hurling abuse at us.
>
> You Devs never respected me, I never respected you Devs. Some of you are still
> relying on that "NeoGeo accurate emulation" diatribe happened in I believe 2003 or
> so.
>
> Without annoying you Devs that NeoGeo in MAME is far from perfect now yet even if you
> are convinced of the opposite but MAMETesters opened bugs are speaking for
> theirselves.
>
> I submitted a lot of 1 line patches and so where's the deal? Maybe you Devs are
> thinking that I didn't read the game driver source code carefully and understood
> every line before submitting the patch? If not, you're more ignorant (which doesn't
> mean stupid but people who don't know the truth of a fact) than I'm thinking.
>

Some were valid, some were not. In general your track record is better than MASH's, but a fair number of your fixes still didn't make a huge amount of sense from a hardware point of view. The net result of this was that people were looking at your build and seeing 'fixes' that were actually hacks, then complaining that Mamedev were terrible for not including them, when really we just had different standards. I do appreciate that you started marking your fixes as hacks to clarify that in the end tho.

The whole incident over at github a few months back where you closed a pull request just because nobody had got around to reviewing it was rather childish too, especially considering I'd got home, gone to look for it in order to do exactly that, only to find you'd closed it and started claiming we were all against you there too.

Is NeoGeo perfect in MAME? No. A lot of the glitches people complain about in current versions DO happen on hardware tho, and then there are a handful of rather messy cases which have MT reports, the CORRECT solution to them isn't simple.

> I quit because I haven't any contact/data for who for example wrote hiscore.dat
> support in the first place and so on. So, since I don't want to make a new version of
> MAMEUIFX without features I cannot include anymore, better stop. It's my choice.
>
> You Devs also are lying when you say you warned mantainers of derivative builds about
> the new re-licensing. I was never contacted for this (which was expectable) or never
> seen an announcement here on in another place about it.
>

it should have been obvious shouldn't it? we had spent 10 months going over every line of code, assigning an owner and their choice of license until we were left with only a handful we couldn't attribute. What would ever make you think the same process wouldn't need to be done with the files that are extra in your project? It's common sense, the reminder on mamedev.org about it was courtesy more than anything else.

You say you don't have contact details for some of the people, it's the very same issue we faced with some code. In the case of the code in your build that is your responsibility to track down the authors tho, we don't even have that code, so we were NEVER in contact with the authors, and it was NEVER our job to track down the authors of it. We're not being unfair about this in any way, we're just highlighting the need for people to go through the same steps we did, albeit on a smaller scale.

The correct way to do something like highscores today would probably be to rewrite it using lua anyway, and if additional lua features hook points were needed, add them. a lua script could easily parse highscore.dat, and actually be a LOT more powerful than the existing system. chances are work done in that field would also be accepted by mainline MAME, so even if we didn't distribute the highscore script ourselves, it could be distributed with the .dat file and work in any MAME build.

> So, game over. Your plan is accomplished.

You might think that all of Mamedev are against you, but there was actually a suggestion of bringing you on to the team at one point, as one of the options going forward to give us an official UI (this discussion happened about a year ago FWIW) but in the end the option of going the MEWUI route was chosen instead because it offers better cross-platform support, and targets like Android are likely to become more important over time.