[hatari-devel] Another question about cycles

[Laurent Sallafranque]

Hi,

Thanks for the help, I think I've got the keys to do it now.


 >> It would be : OR.W (A0), (A1)   :  2   2   8   8   ---> Falcon 
cycles : 2   2   12   14  ?
 > Whoo, whoo! There's no or.w (a0),(a1) instruction :)

LOL. I was so in the theory that I didn't even have a look at what I wrote.
But I think I've got the idea now.

Regards

Laurent


Le 08/11/2011 00:23, Miro Kropác(ek a écrit :
> Hello,
>
>     So, for Falcon, a OR.W D0,D0 gives 0 0 2 4.
>
> Yes.
>
>     But how does it work for a OR.W D0,(A0) ?
>
>    OR Dn,EA 0 1 3(0/0/1) 4(0/1/1)
> + (An)            1 1 3(1/0/0) 3(1/0/0)
> OR Dn,(An) -> 1 2 6(1/0/1) 7(1/1/1)
> or.w dn,(an) -> 1 2 10 13 (falcon)
>
>     Should I add the times of the OR.W DO,DO and the time of the (A0)
>     from the table ?
>     It would be : OR.W D0,(A0)   :  1   1    5   5   ---> Falcon
>     cycles : 1   1   7    9  ?
>
> Well, I can't tell that for sure but I'd say always try to look for a 
> special case (like OR Dn,Dn explicitly listed) and if not found, use 
> the general formula, as I did above.
>
>     Same for a OR.W (A0), (A1) ?
>     It would be : OR.W (A0), (A1)   :  2   2   8   8   ---> Falcon
>     cycles : 2   2   12   14  ?
>
> Whoo, whoo! There's no or.w (a0),(a1) instruction :)
>
>     I's enjoy you control my file when it's finished.
>
> Please don't take my word as definite, most of the stuff I'm talking 
> here about is just my educated guess :) Too bad I'm not a hardware 
> guy, I could measure some stuff then.
>
> -- 
> MiKRO / Mystic Bytes
> http://mikro.atari.org
>
>
> _______________________________________________
> hatari-devel mailing list
> hatari-devel at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/hatari-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.berlios.de/pipermail/hatari-devel/attachments/20111108/175a8886/attachment.html>