06475: gladiatr and clones: gladiatr and clones start with one credit

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
06475	Gameplay	Minor	Always	Jan 25, 2017, 23:54	Jan 27, 2023, 21:01

Tester	cuavas	View Status	Public	Platform	MAME (Self-compiled)
Assigned To	hap	Resolution	Fixed	OS	MacOS X
Status [?]	Resolved			Driver	taito/gladiatr.cpp
Version	0.182	Fixed in Version	0.252	Build	64-bit
		Fixed in Git Commit	536c6ee	Github Pull Request #

Summary	06475: gladiatr and clones: gladiatr and clones start with one credit
Description	Since enabling MCU emulation, gladiatr and clones all boot with one credit. This has something to do with MAME's RAM and/or register initialisation. On an F3 reset it may boot with zero or one credit depending on contents of memory.
Steps To Reproduce	Start gladiatr or a clone. It will boot to the "PUSH START" screen rather than attract mode.
Additional Information
Github Commit

Flags
Regression Version
Affected Sets / Systems	gladiatr and clones

Attached Files

No.14726 cuavas Administrator Feb 4, 2018, 23:33	The MCU code we have is weird, and obviously has various things hacked out of it, including RAM/register initialisation. It looks like the kind of thing you'd find on a bootleg, but more than one person has insisted they've found this MCU program on genuine boards. Because the MCU code doesn't initialise its registers and onboard RAM on boot, it starts with whatever MAME initialises it to. This happens it start up thinking it has one inserted credit it needs to tell the game program about.
No.21031 Augusto Tester Jan 21, 2023, 10:55	Only reporting the issue happen in 0.251.
No.21032 hap Developer Jan 21, 2023, 16:29	The first opcode of cctl and ccpu is nonsense. ucpu looks like it has it right? It has the same init routine. So, this solves the problem: ROM_FILL(0, 1, 0x22) after loading cctl/ccpu rom. Is it a bad dump problem, or did bootleggers do this to make it work on their board? Also, coins slots appear to be the wrong way around.