Added explicit control handler for the console. Ctrl+C/Ctrl+Break now
explicitly terminate the process forcefully, rather than unwinding
through the system in an unexpected state. Other console events
(exit, shutdown, logoff) request a graceful exit.
This needs an unfortunate hack with a dummy driver added to seed references to
the devices that are only otherwise referenced by libemu.a.
Also removed SOUNDS += CUSTOM since such a sound device no longer exists.
manual synchronization of VBLANK start against an external timing source.
Updated the MC6845 device to call reset_origin() on its screen at the
start of each frame if a screen is present.
The practical upshot is that now the screen timing and MC6845 timing is
once against synchronized, but by tying the screen timing to the MC6845
and not the other way around.
-----------------------
- 04067 [Misc.] blktiger: Test mode doesn't work. (MooglyGuy)
Non-Whatsnew note: In my opinion, I feel that it's a reasonable assumption that whatever video ASIC or other hardware is used defaults to having at *least* the text layer turned on. The game doesn't even attempt to write any video configuration registers apart from the bank reigster when booting with test mode turned on, which strongly implies certain default states of certain registers.
Moved -effect implementation out of OSD code and into core since
the implementations were identical across Windows/SDL and implemented
in the core itself.
New games added or promoted from NOT_WORKING status
---------------------------------------------------
Devil Island (2 sets) [Brian Troha, The Dumping Union]
their state like other devices. [Atari Ace]
--
From: Atari Ace <atari_ace@frontier.com>
Date: Mon, Oct 11, 2010 at 12:32 AM
Subject: [patch] Audio device conversions
To: submit@mamedev.org
Cc: atariace@hotmail.com
Hi mamedev,
I've been working on auto-converting the majority of the remaining
drivers to use driver_data (by finding global and static state and
moving it to a state class programmatically), but a number of issues
make it tricky to do. One problem is the various audio device drivers
that are in src\mame\audio that don't encapsulate their state and thus
get pulled accidentally into the driver_device by my scripts. Here's
a patch that fixes most of them. The first patch makes sure the
interface is always a device (mostly changing space handler to device
handlers) and does some minor cleanup. The second patch then places
the state into an appropriate token.
~aa
