Amiga-Development

Please login or register.

Login with username, password and session length
Advanced search  

News:

Created for developers of all Amiga camps

Poll

How should the Masking function on channel B be handled?

Global, whatever value just happens to be there is valid.
Global with initialization, after boot or reset this feature is ignored until it's first usage..
Per Job, the mask have to be defined for each blitter task before the task is issued.

Pages: 1 ... 18 19 [20]

Author Topic: Blitter Enchancements  (Read 45040 times)

0 Members and 1 Guest are viewing this topic.

NorthWay

  • Jr. Member
  • **
  • Posts: 70
    • View Profile
Re: Blitter Enchancements
« Reply #285 on: February 19, 2016, 02:19:52 PM »

Two more minor ideas about the blitter:

Given that there is a new address range for registers ($DFF2xx?) should the "USEx" bits of bltcon0 be separated from the other bits (i.e. make them two registers)? With a Blipper implemented a list of cookie-cuts could transform to a list of copy (repaint) ops by changing the initial minterm.

Would it make sense to introduce "relative" addresses? I.e. having a base address register for a channel and then add whatever value written to the address registers to get the full address. I'm still thinking in reference to a Blipper and trying to re-use lists as much as possible.
If pairing the base with a max register you could get some pseudo-security/exception handling if that has any value. Also, just setting the base to 0 would make it identical to what it does today anyway.
Logged
Pages: 1 ... 18 19 [20]