|[10 earlier articles]|
|Re: Processor specific optimisations email@example.com (2002-01-28)|
|Re: Processor specific optimisations firstname.lastname@example.org (MickaŽl Pointier) (2002-01-30)|
|Re: Processor specific optimisations Ulrich.Teichert@gmx.de (2002-01-30)|
|Re: Processor specific optimisations email@example.com (2002-01-30)|
|Re: Processor specific optimisations firstname.lastname@example.org (MickaŽl Pointier) (2002-02-06)|
|Re: Processor specific optimisations email@example.com (Clark L. Coleman) (2002-02-06)|
|Re: Processor specific optimisations firstname.lastname@example.org (2002-02-06)|
|From:||email@example.com (Rick Hohensee)|
|Date:||6 Feb 2002 23:39:24 -0500|
|References:||02-01-077 02-01-126 02-01-151 02-01-164|
|Posted-Date:||06 Feb 2002 23:39:24 EST|
firstname.lastname@example.org (Frank Wilde) wrote in message news:02-01-164...
> Rick Hohensee <email@example.com> wrote:
> > [...] Speaking of the 6502 and deliberately falling off the edge
> > of the earth as a programming technique, has anyone ever used 6502
> > "pre-indexed indirect" addressing? I didn't find any occurance of it
> > in the C64 ROMs, 8k by Commodore and 8k by Microsoft.
> That would be ($nn,X), right? I'd suspect management of the
> connectedness of screen lines to logical lines (present in all CBM
> machines with less than 80 cpl) of having used that mode; why else
> would they reserve (about) 24 precious zero-page variables for just a
> flag per byte?
I don't know. I do recall searching the ROMs, BASIC and kernal, for
the bit patterns I thought would show use of pre-indexed indirect, and
got no hits. I definitely did that, but I may not have done it
I doubt if linking lines requires P-I-I. I think what P-I-I really
does on 6502 is implement a data structure something like a set of
cylindrical data objects, with diameters of 256 bytes, but this is
VERY vague memory. (I recall thinking it resembled a drumkit.)
Return to the
Search the comp.compilers archives again.