apparently gprider uses a setup similar to f1 exhaust note etc. where the game uses a dual PCB stack and bridge board providing a direct connection containing shared RAM rather than a network connection.
in order to facilitate this we need both PCB stacks running, so as per a request from SailorSat I've made this possible, and added a set 'gpridert' which uses this configuration. Unfortunately, for reasons unknown to me, the communication still breaks down (it was thought that this could have been caused by the hacks needed if running multiple instances of MAME with shared RAM as a file instead, but apparently not)
as this is something of a failed experiment I'm submitting it for SailorSat to take a look at, might revert it before the next release however, although once we do have full dumps of some of the Sega gambling / redemption games we'll need the various board types as devices anyway as those games mix multiple types of boards for the different screens.
New Clone Added
--------------------------------------------
Soccer Superstars (ver UAC) [Smitdogg, The Dumping Union]
NOTE: These sets don't properly check the mask roms and aren't playable any more. Recent regression?
I *almost* was able to do this as cleanly as I would have liked. There remains
a switch statement on line 824 in video/model2.c that I would have liked to get
rid of, but I have yet to figure out a way to pass model2_renderer's member
variable m_renderfuncs[renderer] to the FUNC macro (living inside a
render_delegate constructor) successfully.
Of course, it could be done in an entirely different manner, but that can wait
for the great Model2 rewrite of 20XX :-).
Also isolated Andrew Zaferakis' copyright to hng64_3d.c since that's where his
~700 lines of code lives. (nw)
Finally, touched up some spacing in drivers/Model2.c - converting this one
to polynew will be a little more involved and will be done on a later date. (nw)
the legacy one. (nw)
This seems to work, and seems to adhere to the style of other "polynew" uses.
If anyone sees anything totally wrong, let me know, because I will be slowly
moving other drivers to "polynew" in the future.
