update comments to clarify the epos.cpp tristar 8000 clock situation, based on schematics (nw)

This commit is contained in:
Lord-Nightmare 2017-05-17 13:20:31 -04:00
parent 843dcdf32f
commit 827539b216

View File

@ -13,11 +13,15 @@
- Super Glob seems like a later revision of The Glob, the most obvious
difference being an updated service mode.
- These games don't have cocktail mode.
- The divisor 4 was derived using the timing loop used to split the screen
in the middle. This loop takes roughly 24200 cycles, giving
- The CPU clock divisor 4 was derived using the timing loop used to split
the screen in the middle. This loop takes roughly 24200 cycles, giving
2500 + (24200 - 2500) * 2 * 60 = 2754000 = 2.75MHz for the CPU speed,
assuming 60 fps and a 2500 cycle VBLANK period.
This should be easy to check since the schematics are available, .
This also matches the IGMO schematic, as it is the /CLK signal, which is
derived from the 11MHz xtal by dividing it down by two 74LS193 chips, one
(U92) dividing the clock by 2, and another (U91) having 3 taps, further
dividing the already divided clock by 2 (/CLK), 4 (/PLOAD) and 8 (CLOCK).
The CLOCK signal drives the AY.
- I think theglob2 is earlier than theglob. They only differ in one routine,
but it appears to be a bug fix. Also, theglob3 appears to be even older.
@ -472,7 +476,7 @@ static MACHINE_CONFIG_START( epos ) /* EPOS TRISTAR 8000 PCB */
/* sound hardware */
MCFG_SPEAKER_STANDARD_MONO("mono")
MCFG_SOUND_ADD("aysnd", AY8912, XTAL_11MHz/16) /* 0.6875 MHz??? - clock not confirmed (similar to the clock listed below) */
MCFG_SOUND_ADD("aysnd", AY8912, XTAL_11MHz/16) /* 0.6875 MHz, confirmed from schematics */
MCFG_SOUND_ROUTE(ALL_OUTPUTS, "mono", 1.0)
MACHINE_CONFIG_END