Re: C64 (or possible other CBMs too): PETSCII -> screen codes conversion table in ROM?

Hi

I dont remember very well, but I think it was just an offset in 1 or 2 
ranges of values
I think the offset was #$40 to add or substract...

the code was quite short and without tables

last time I wrote that it was in 1984...

--
didier

Le 12/12/2011 17:11, "Gábor Lénárt" a écrit :
> Hi,
>
> I want to manipulate the text video RAM directly, so I mostly have
> information in the form of "screen codes" already (instead of PETSCII) since
> using a KERNAL function like CHROUT is much slower for heavy usage.
> However, I have some information presented in PETSCII. I have the idea that
> KERNAL should know about the PETSCII->screen codes translation somehow,
> since it also prints onto the screen. But I want to avoid the "overhead"
> of calling CHROUT, so my question is: can I found a conversion table
> somewhere in the ROM, I can use, like this?
>
> ; X=byte to output
> LDA conv_table,X ; conv_table in KERNAL
> ; now I have the screen code in A
>
> Of course I can include a 128 byte long conversion table by my own too, but
> I wouldn't make my software larger, if there is already a table like this in
> the ROM somewhere. Or does the conversion based on a routine and not table?
> For sure, in this case, I don't need to interpret special "control
> characters", like reverse on, return ...
>
> It's possible that KERNAL does this by a subroutine instead of a table, but
> then it's surely slower than using just a table.
>
> Currently I do an ugly trick: after clearing the screen, I print out
> characters by KERNAL, then I copy bytes from the video RAM (what kernal
> written as bytes into the video ram) into my reserved area used as the
> translation buffer.  After this, I clear the screen again.  In this way, at
> least I don't need to include a pre-generated table in my program, however
> this solution feels a bit hacky for me :) If kernal provides any translation
> table (rather than using code for "translation") it's still means less
> memory footprint if I can use that.
>
> thanks!
>
>         Message was sent through the cbm-hackers mailing list


       Message was sent through the cbm-hackers mailing list