MAME v121 released

[Demon Cleaner]

Next MAME release

MAMETesters Bugs Fixed
----------------------
dunhuang0120u3gra [Vas Crabb]

Source Changes
--------------
Updated DIP locations in the following drivers: [Vas Crabb]
- bublbobl.c: bublbobl, tokio
- cps1.c: all games
- dunhuang.c: dunhuang
- mario.c: marked all as active high (confirmed from manuals)
- segag80r.c: astrob2, astrob1, spaceod and pignewt
- segag80v.c: all games, fixed switch ordering

Made more functions static to reduce namespace pollution.
[Atari Ace]

Improvements to the Touchmaster driver, still labelled not working.
You can actually play tm and tm3k, but sometimes you have to reset
because they stop registering the touchscreen input. tm4k is
protected and hence totally unplayable. [Luca Elia]

Fixed compilation problems with altivec accelerators. [Vas Crabb]

Fixed bug in Voodoo statistics accumulation. [Atari Ace]

Reverted to old behavior where even with -switchres disabled,
MAME will use exclusive mode when rendering with Direct3D. This
means some of the switching to/from fullscreen mode problems will
come back, but should prevent confusion over poor performance by
default. [Aaron Giles]

Made it a non-fatal error if we are unable to change a joystick to
absolute mode in the Windows input code. Some drivers apparently
freak out about this. [Aaron Giles]

Defined a couple more inline functions: div_64x32_rem and the
unsigned equivalent divu_64x32_rem. Added inline implementations
for GCC x86 and MSVC. [Aaron Giles, Vas Crabb]

Cleaned up timer interfaces. Created new module attotime which manages
a new structure type attotime, equivalent to the old mame_time. All
time math functions are now implemented in attotime and have been
removed from timer.h. The following search/replace over the code has
been made:

mame_time -> attotime
subseconds_t -> attoseconds_t
time_zero -> attotime_zero
time_never -> attotime_never

make_mame_make -> attotime_make
mame_time_to_subseconds-> attotime_to_attoseconds
mame_time_to_double -> attotime_to_double
double_to_mame_time -> double_to_attotime
add_mame_times -> attotime_add
add_subseconds_to_mame_time -> attotime_add_subseconds
sub_mame_times -> attotime_sub
sub_subseconds_from_mame_time -> attotime_sub_subseconds
scale_up_mame_time -> attotime_mul
scale_down_mame_time -> attotime_div
compare_mame_times -> attotime_compare

SUBSECONDS_TO_DOUBLE -> ATTOSECONDS_TO_DOUBLE
DOUBLE_TO_SUBSECONDS -> DOUBLE_TO_ATTOSECONDS
USEC_TO_SUBSECONDS -> ATTOSECONDS_IN_USEC
SUBSECONDS_TO_HZ -> ATTOSECONDS_TO_HZ
HZ_TO_SUBSECONDS -> HZ_TO_ATTOSECONDS
MAME_TIME_IN_HZ -> ATTOTIME_IN_HZ
MAME_TIME_IN_SEC -> ATTOTIME_IN_SEC
MAME_TIME_IN_MSEC -> ATTOTIME_IN_MSEC
MAME_TIME_IN_USEC -> ATTOTIME_IN_USEC
MAME_TIME_IN_NSEC -> ATTOTIME_IN_NSEC
MAME_TIME_TO_CYCLES -> ATTOTIME_TO_CYCLES
MAME_TIME_IN_CYCLES -> ATTOTIME_IN_CYCLES

In addition, all the mame_timer_* functions have been renamed back to
their original names timer_*, so it is timer_set() instead of
mame_timer_set() once again. The mame_timer object itself has been
renamed to emu_timer.

New games added or promoted from NOT_WORKING status
---------------------------------------------------
Best Of Best [Luca Elia]

New clones added
----------------
Cluedo (prod. 2) [James Wallace]

[Harrison]

Wow! Another new version of Mame already. Not many new games but it looks like a lot of internal code updates have taken place. Time to update my set again.

[AlexJ]

I seem to spend more time verifying my set against a new torrent than seeding as they update it so regularly. Still, glad to see development is still going strong.

[Harrison]

How are you updating your set each time Alex? If you just join the latest torrent and point the download at the existing set you don't need to verify anything, just let it download the difference, and then seed away.

[AlexJ]

How are you updating your set each time Alex? If you just join the latest torrent and point the download at the existing set you don't need to verify anything, just let it download the difference, and then seed away.

I say verify, I mean check - uTorrent looks for what's already downloaded so I start >90%. For some reason (probably the size of the thing and number of files, combined with a USB External HDD) this takes ages to do.

[Harrison]

The MAME set does usually take a few minutes to check through before starting to download the differences. I think it is mainly due to how many files are in the set, and utorrent needing to check the hash and size of every one.

[Demon Cleaner]

AlexJ is right. I keep my torrents that are updated regularly on my internal disk, then do a copy with ViceVersa to an external one for backup. But if you use the external to update the set, it will take ages and sometimes freeze the PC when checking the already existing files.

[Harrison]

It is a very data intensive process. I only ever store such files on the internal HDs in my server.

I did once test using a NAS drive to store and seed files, but every time I needed to reboot the system uTorrent took forever to verify each torrent being seeded from the NAS before it then started to seed them, so I gave up on that idea.

[Demon Cleaner]

That was one of the reasons I send back my NAS.

[Harrison]

That isn't a limitation of the NAS though, as a NAS is only really an HD or collection of HD's directly attached to the network. It is a restriction of how TCP/IP works, which slows down the access of a large number of small files at once.