View Single Post
Old 13 August 2016, 19:33   #10
Toni Wilen
WinUAE developer
 
Join Date: Aug 2001
Location: Hämeenlinna/Finland
Age: 42
Posts: 19,559
Some more technical details. (Will be moved to undocumented thread when emulation is complete and full details are known)

I have confirmed that active bitplane DMA really continues to next scanline. Tested by modifying bitplane modulo at the beginning of next scanline, if modulo was changed between hpos 0 to 9, current scanline changed, if changed later, it only affected following scanline.

Same effect with bitplane pointer, if pointer was changed early enough, bitplane data in far left overscan changed in current scanline, if changed later, it only affected graphics after DDFSTRT.

Current emulation does not support above tricks -> it needs to be better

I guess this does not cause corruption because if AGA fetchmode is large enough, there can't be refresh cycle conflicts. But I do expect audio DMA conflicts which should break sound. (Probably explains why no one noticed buggy graphics setup..)
Toni Wilen is offline  
 
Page generated in 0.04722 seconds with 9 queries