@doctorxyz, I haven't taken a look at Xploder yet, but I have a comment to make about why the Persona 4 fix works.
The VSync rate for 480P isn't different from the VSync rate of 480I, if I'm not wrong. The difference is in the HSync rate.
So why does the VSync interrupt handler of Persona 4 screw up if the video mode is changed to a progressive video mode? I believe that it's not an error in the Sony GS user's manual, and the FIELD field within the GS CSR really flips accordingly.
Unfortunately, progressive scanning means that the even and odd scan lines are drawn at once, so the FIELD field within the CSR is probably being flipped whenever the GS renders scan lines from each set.
The problem is... at the start of the VSync blank (INTC Interrupt 2), the contents of the FIELD field within the CSR register is probably going to be always the same as line rendered before the start of the VBlank is always the same, when progressive scanning is used.
So now I believe that this fix is going to be forever a game-specific fix, since only Persona 4 is known to check the FIELD field of the CSR and to act on it.
@all, please feel free to give comments of what I've said here, as I came to this conclusion after being baffled by why the VSync interrupt handler can be affected by a change in the HSync frequency and drawing method.
As for other games that are being broken by the HSync frequency changing, we may be able to counter that by tampering with the kernel itself (Unless again, the game is checking the hardware directly, which will be more challenging to counter).
Unmodified SCPH-77006 with SM 3.6
SCPH-39006 with M-chip modchip, SCPH-10281 NA and refurb Seagate 80GB HDD
SCPH-10000 v1.00 with SCPH-10190 PCMCIA NA and SCPH-20400 HDD unit