James Pond II / II+ VIDC timings

Discuss VIDC, translation to VIDC20 and emulation of VIDC
Post Reply
JonAbbott
Posts: 1737
Joined: Thu Apr 11, 2013 12:13 pm
Location: Essex

James Pond II / II+ VIDC timings

Post by JonAbbott » Wed Apr 27, 2016 5:22 pm

Continuing the discussion around the VIDC emulation issues from the ADFFS 2.54 thread...
  1. Scrolling is jerky if the VSync IRQ is triggered before the screen geometry has been calculated (ArcEm also suffers this problem, Arculator and Red Squirrel do not)
  2. It's T1 timer values appear to be slightly out under VIDC emulation...which is possibly due to the timing issue from point 1 (ArcEm is spot on, Arculator and Red Squirrel are both way off on timings)
(2) can be seen by the fact the blue sky on immediately starting the game, starts 3 pixels too late. The copperbars also appear to be slightly too thin compared to physical.

JonAbbott
Posts: 1737
Joined: Thu Apr 11, 2013 12:13 pm
Location: Essex

Re: James Pond II / II+ VIDC timings

Post by JonAbbott » Thu Apr 28, 2016 11:00 am

Looking through the Arculator source, the VSync IRQ occurs at VDER. This explains why the horizontal scrolling is working under Arculator.

I've not seen a timing diagram for VIDC1/VIDC20 that indicates at which point it determines the screen geometry, but we can deduce it's before the VSync IRQ is triggered, otherwise JP2's horizontal scrolling wouldn't work on physical. This implies the geometry for the next frame is determined at VDER, prior to the VSync IRQ.

Post Reply