MAME Git (2016/08/21) is compiled. MAME stands for Multiple Arcade Machine Emulator. When used in conjunction with images of the original arcade game's ROM and disk data, MAME attempts to reproduce that game as faithfully as possible on a more modern general-purpose computer. MAME can currently emulate several thousand different classic arcade video games from the late 1970s through the modern era.

MAME Git Changelog:
* Play_2: small correction (nw)
* add NFS Underground System Recovery CD dump (nw)
* Jolly Card (Austrian, set 2): Added specs... [Roberto Fresca, Team Europe]
* Jolly Card (Austrian, set 2): Added GAL16V8S dump (Team Europe, caius)
* Added support for MPU4 layouts for games beginning with 'a', based on MFME2MAME conversion. CHR protection data is still not hooked up for many of these sets, so lamping behaviour is incorrect.
* Fixed rom size in s80btest.
* update comments (nw)
* dotrikun: update prev commit
* Merge branch 'master' of https://github.com/mamedev/mame
* Merge pull request #1277 from superctr/technodr
Make Techno Drive working, fix H8 ADC inputs.
* Make Techno Drive working
The game previously displayed a 'COIN ERROR' at bootup. Fixed this by
patching the sub CPU program (not the best fix maybe but I hope it will
work for now).
The contents of 'th1fl3l.12' and 'th1fl3u.13' appeared to have been
swapped. Swapped the filenames and placed them at the proper memory
location (0x1800000) and the game no longer crashes at the Namco logo.
Played through the game a few times and it seems to be fully working
now.
* Change endianness of H8 ADC register
This was broken in a rewrite a long time ago, breaking the controls in
some Namco System 12 (and probably 23) games.
This commit improves controls in golgo13, g13knd and technodr.
* Merge pull request #1270 from PressStartandSelect/master
Document 3 diagnostic menu bugs in the Vii's BIOS
* document 3 diagnostic menu bugs
from https://tcrf.net/Vii
* Reverted last commit.
* Added vgmplay to mess build.
* Pinball fix from my previous commit (nw)
* alto2: Refactor videotiming + use attoseconds_t
Use MAME's type attoseconds_t and handle the timing accus in
attoseconds instead of nanoseconds.
Try to solve the video timing issues by using an emu_timer running
at the field rate of 60Hz, which the video state machine synchronizes to.
* Merge pull request #1275 from johnparker007/master
Update regenerated Maygay layouts - all now initialise ok (nw)
* Update regenerated Maygay layouts - all now initialise ok
* dotrikun: try another method for video timing
* Continued synching with PinMAME. Still not finished (nw)
* -vgmplay.cpp: Added POKEY support, and rebalanced audio. [Ryan Holtz]
* aica: add ADPCM diff value clamp based on encoder research [kode54, MetalliC]
if there is known issues with YMZ280B it might be worth try to same decoder there
* luaengine: remove now unnecessary stuff since a copy of searchpath is made and it won't be garbage collected from underneath us (nw)
* Re-apply comment changes
* Revert problematic change (nw)
* dotrikun: eat_cycles clamps to 0, adjust_icount does do what i intended
* Merge branch 'master' of https://github.com/mamedev/mame
* Playmatic Pinball: Finished for now (nw)
* typo + tech notes...
* Added technical notes to Inter Games poker. [Roberto Fresca]
* Merge branch 'master' of https://github.com/mamedev/mame
* compis: Fixed RTC to use 24 hour mode. [Curt Coder]
* Funworld driver: Added Inter Games poker specs and some cleanups.
* Merge branch 'master' of https://github.com/mamedev/mame
* compis: Promoted driver to working. [Curt Coder]
* Cleanup....
* Merge branch 'master' of https://github.com/mamedev/mame
* Merge pull request #1273 from npwoods/more_tafoid_bugs
fix more softlist regressions
* Changing device_image_interface::determine_open_plan() to not support creating an image unless device_image_interface::create() is used
I'm somewhat surprised that this has not bit us until now. The implication of how it used to work is that device_image_interface::load() could actually create images; it is hard to eliminate the possibility that something might be incorrectly relying on this behavior. I've determined that this isn't the case for imgcntrl.cpp and floppycntrl.cpp
* Now reporting Windows ERROR_INVALID_NAME as osd_file::error::NOT_FOUND
* Cleanup. (nw)
* compis: WIP. (nw)
* -vgmplay.cpp: Added YM3812 support. [Ryan Holtz]
* Merge pull request #1272 from mamehaze/setadditions
add the PLD dumps to the nzeroteam sets (nw)
* add the PLD dumps to the nzeroteam sets (nw)
* Merge pull request #1271 from mamehaze/mpu4test
more mpu4 descriptions (nw)
* more mpu4 descriptions (nw)
* Merge pull request #1266 from mamehaze/mpu4test
some MPU4 descriptions (nw)
* some MPU4 descriptions (nw)
* Save a few lines (nw)
* make path_iterator hold onto its search path
* Magic Card II specs typo....
* New machines marked as NOT_WORKING
----------------------------------
Unknown Inter Games poker [Roberto Fresca, Team Europe]
* Merge pull request #1260 from mamehaze/setadditions
new clones (NOT WORKING) Jungler (Subelectro, bootleg on Scramble hardware) [Andrew Welburn]
* mew clones New Zero Team (V33 SYSTEM TYPE_B hardware) [caius]
existing set was licensed, now marked as 'China?'
* new clones (NOT WORKING) Jungler (Subelectro, bootleg on Scramble hardware) [Andrew Welburn]
unique / interesting bootleg, CPU code is encrypted, and unfortunately one gfx rom is bad and can't easily be reconstructed (see note)
if anybody out there has the same bootleg then it would be helpful to dump it.
* Merge pull request #1262 from rzero9/patch-3
fmtowns_cd.xml: new dumps
* fmtowns_cd.xml: new dumps
Air Management - Oozora ni Kakeru
Hiouden II
Windows 3.1 L12
* Merge pull request #1268 from Dagarman/master
kim1: RS & ST keys, input ports update
* kim1
* Merge pull request #1267 from mamehaze/extrafruity
fix maygay m1a lamping code to work with layouts (nw)
* outdated comment (nw)
* typo (nw)
* fix maygay m1a lamping code to work with layouts (nw)
* Merge pull request #1269 from npwoods/fix_a800_brucelee
Fixed Tafoid report: 'mame a800 -flop1 brucelee' failed
* a800: Reintroduced hack that causes a800 to always use the ATR format for softlists
'mame a800 -flop1 brucelee' now works again
* Fixed an issue with legacy floppy loading caused by filetype() now being a std::string
This caused 'mame a800 -flop1 brucelee' to fail on startup, but after this fix the image mounts but incorrectly.

Download:MAME Git (2016/08/21) x86
Download:MAME Git (2016/08/21) x64
Source: Here