-
Notifications
You must be signed in to change notification settings - Fork 53
/
Copy pathbugs.txt
38 lines (25 loc) · 1.99 KB
/
bugs.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
****** As of Version 3.7 (XX XXth, 2025) ******
Major bugs in PinMAME that we are aware of:
-------------------------------------------
#1) Not all Williams/Bally WPC functions are implemented in PinMAME.
The most important (?!) functions missing are the ticket dispenser,
and accurate real time clock emulation (current rtc functionality is a workaround).
#2) The WPC printer function is only tested with default settings.
The printout will probably not work with other settings.
#3) The name and year for each game were taken from the Internet Pinball Database
(http://www.ipdb.org) OR from supposedly more accurate personal research/sources, and so may not be 100% accurate!
#4) Occasional pops/clicks on Capcom sound.
#5) Occasional pops/clicks/crackling on Whitestar II sound, only on high volume levels.
(~24 and above, for example on Nascar and Grand Prix, or even ~20 on LOTR, so avoid this)
This seems to happen on real machines (verified on 2 different Grand Prix pinballs), too.
#6) Some Data East have slow DMD startups (Star Wars,...), Sega 192x64 timing is not 100% accurate and
sometimes shows startup screen, Alvin G. has timings issues resulting in some flicker.
#7) Sound gets cut off in Alvin G games sometimes, and some weird notes in World Tour occasionally.
#8) Slugfest will work with "pinmame -cheat sf_l1". Same goes for Theatre Of Magic home version (tom_14h(c)).
Alternately, check the "skip pinball startup test" in vpm.
#9) GTS80S DAC sound is off for most of the sound effects.
#10) Some setups crash when using the AT91 JIT compiled code (e.g. Whitestar II and SAM), disable AT91 JIT
by setting "at91jit" to 0 in the registry.
#11) AT91 JIT is 32-bit and x86 only. The non-JIT core doesn't have tight IRQ timing that may cause emulation issues.
It would be possible to tighten up the timing in the interpreted core, but it causes a huge performance hit
(have to keep checking on memory writes whether an IRQ needs to fire).