|
Re: MAME Opinions
10/29/20 03:22 PM
|
|
|
> Your answer is exactly why there's no interest in Mame now. > > Give me back April fools, funny threads, wip pages like 15 years ago. > > You're killing MAME by not listening users. > > It's a very sad day today for an early fan like me.
I can’t win with people like you. I have done a lot to improve MAME, in terms of emulation, usability, and architecture.
General usability improvements I’ve made kind of recently:
- Better UI for assigning inputs and configuring analog controls.
- Better performance and usability for the system and software selection menus.
- Ability to control which devices with keyboard inputs are active, to make systems with multiple keyboards usable (e.g. a computer with a serial terminal plugged in).
I’ve done a whole lot of stuff for the artwork system, a lot of which has come together in the last couple of months:
- Parameter animation for more interactive layout features.
- Circles/ellipses that actually look round.
- Support for SVG and Windows DIB images.
- Allowing arbitrary sets of elements/screens to be toggled on and off.
- Arbitrary blend modes for screens and elements, not tied to their layer.
- Support for multi-segment and dot matrix displays using external images.
- A whole lot of clipping and alignment issues fixed.
- Not immediately crashing MAME on encountering an invalid JPEG file.
I’ve posted samples of this stuff on the artwork forum here, in the shout box at Bannister, and in various reddit threads. Haze posted a video on his YouTube channel demonstrating new features in Cosmo Gang.
Emulation improvements off the top of my head:
- Proper video emulation for Laser Battle/Lazarian and Cat and Mouse. This made the games playable, previously you could get hit by stuff you couldn’t see and area effects were missing.
- Sound emulation for Cat and Mouse and Cheeky Mouse, and analog effects for Money Money and Jack Rabbit music.
- Proper microcontroller emulation for Xain'd Sleena, Gladiator, Great Swordsman, Joshi Volleyball, and several other games. This fixed several gameplay issues.
- Proper Osborne 1 memory/peripheral mapping, and support for the SCREEN-PAC and Nuevo Video high-resolution mods.
- Support for most of the original Macintosh and Macintosh Plus keyboards and keypads, and bug fixes for mouse behaviour.
- Apple II Grappler+ and Parallel Interface card printer interfaces, and bug fixes for Apple II Mouse card.
- Proper emulation for several Amiga keyboards from Cherry and Mitsumi.
- Sun workstation keyboard and mouse emulation.
If you look at MAME code today and compare it to five years ago, it’s barely recognisable. Today’s MAME code is far more modern and approachable. It’s easier to maintain, and easier for new contributors to get started with. I’m proud to say I’ve been a big contributor to the transformation, even if some of the changes weren’t my ideas to begin with. MAME has been around for over two decades now, so modernisation is a slow process this work is still ongoing.
I actually do have a plan for proper touchscreen support, but any major changes to MAME are a lot of work, so progress is slow. There are several contributing factors, including the size of MAME, the age of the some of the code, and the fact it has to be kept reasonably stable because it’s undesirable to break the release cadence or stop other people from being able to work on the project. It also competes with other demands on one’s time (real life syndrome). Although it’s not obvious right now, some of the recent changes to the artwork system will help with touchscreen support.
On top of all that, I make sure we get a release out almost every month. I do some basic quality control checks, get the release notes together, make source and Windows binaries available in multiple places for redundancy, write an executive summary, and respond to questions on social media.
But as I already said, I know I can’t win. So I’m going to go back to sipping my vodka, and thinking about what MAME stuff I can do that you definitely won’t appreciate.
|
|