| Main > Main Forum |
| the state of mame |
| << < (83/112) > >> |
| Haze:
--- Quote from: ark_ader on January 14, 2011, 04:54:55 pm ---No need to reply I know the answer. But it is fascinating considering how much technology has progressed from a z80 (or emulated z80) today. --- End quote --- Well the old emulated z80 might have been good enough to emulate game X, ---fudgesicle--- timings, they weren't important, just emulate the opcodes, it works. Then we found game Y, which required some undocumented behavior of the z80 to be emulated, and more accurate timings. Then we found game Z, which required further opcodes to be perfected, and strange behavior related the the actual microcode of the chip under extreme conditions in order to work properly. Then we found game XYZ which required not only perfect timings, but manufacturer differences between each revision of the Z80 in order to work. Do you seriously expect us to keep around 3 imperfect versions of the core just to keep you happy? It's stupid, MAME will emulate things, improve in accuracy, and as a result, yes, things get slower. New technology advances at such a rate that even a budget PC will run 95% of the supported systems at full speed. It's progress. It has a price. Live with it. |
| VanillaGorilla:
Thats 11 pages of my life i'll never get back. Wow ladies, you all got your panties in a knot! Personally, I would like to offer my left nut, to nicola and aaron, and all of the mamedevs for making my childhood dreams a reality. I ---smurfing--- love you guys. You are an inspiration. :notworthy: :notworthy: :notworthy: The rest of you sound like greedy babies. :angry: |
| MKChamp:
Bottom line: It's open source. Change what you don't like/want added/removed. We're all receiving a free meal and then bitching that we were handed a steak dinner when we wanted a lobster. |
| sjbaines:
--- Quote from: ark_ader on January 14, 2011, 04:54:55 pm ---How is that progress? --- End quote --- Seems pretty obvious - it's progress because the emulation is becoming more detailed (hence accurate) over time, which of course has a performance hit. Yes, MAME could have a whole bunch of simplified variants of all the CPUs, and use the simplest one that seems to work for any given game, but that's a lot more code to maintain over just having the full models and using them for everything. Maintaining all those simplified models would take a whole load of extra effort, it would be a breeding ground for bugs, and the only benefit would be faster emulation of certain games. No benefit whatsoever for accuracy of emulation. So, MAME gets slower over time as a side-effect of getting closer to its stated objective of accuracy. But all the old releases are still available if you prefer them. What exactly is the problem? |
| Grasshopper:
Wow, what a depressing train wreck this thread has become. I think it's been clear for some time that MAME has lost its way. I can't remember the last time I got excited about a new release. To be fair, I haven't really noticed the slowdowns that others have experienced (possibly because I mostly play older games) but I haven't really seen clear improvements for some time either. I assumed the reason for this was simply because most of the important work had already been done and therefore developers had got bored and left. No doubt that's partly true. However, thanks to this thread, I now realise that there is also a lot of toxic politics going on behind the scenes. It's really sad to see a project that has given me hours of pleasure, and which I still have a lot of affection for, descend into this state. If things continue as they are, then I can only really see two possible outcomes: The first scenario is that the remaining MAME developers become increasingly introverted and continue to cut themselves off from the user community who they perceive (mostly unfairly IMHO) as being ungrateful, demanding and entitled. If that happens then we're likely to see few if any new developers, fewer bug reporters, fewer ROM dumpers etc., as most of those people are recruited from the game playing community. The official MAME project will slowly but surely grind to a halt. The second scenario is that a new group of developers create a fork from a faster old version of MAME and add the features that people are asking for. Over time the forked version diverges from the official build and eventually eclipses it. It certainly won't happen overnight. But given the large numbers of people who appear to be sticking with ancient versions of MAME I see this as a fairly likely scenario in the longer term. At first glance it appear to be a fairly good outcome for the end user. However, in reality, it leads to duplication of effort and wastes resources. The end users will effectively be forced to choose between having access to the latest games and bugfixes, or having the best possible emulation experience. To me that seems a terrible shame. I've seen other open source projects fork in this way, and it takes years for them to get back on track. This split between (some) developers and (some) users is not benefiting anyone. And I see it as being entirely unnecessary. I don't see why MAME can't "document the hardware" and simultaneously provide the best possible emulation experience on the widest possible range of hardware. The two aims are not mutually incompatible. |
| Navigation |
| Message Index |
| Next page |
| Previous page |