The following is yet another way to do it [the same identities of course
apply]:
TAD ARGONE
AND ARGTWO
CMA
DCA TEMP
TAD ARGONE
MQL
TAD ARGTWO
MQA
AND TEMP
This method has the advantage that it also works on the straight-8 EAE
avoiding the singular problem of SWP unique to the two original EAE
implementations,. [In theory, EAE could be on the LINC-8.]
The MQ has the advantage of the .IOR. that helps start things off, but the
MQ cannot be .AND. with anything. Nothing need be dependent on the MQ
outside of the code. [If you assume the MQ is one of the arguments doesn't
help here; most real-world needs wouldn't do that.]
Ironically, the shortest and fastest seems to be avoidance of the MQ
altogether [thus making it work on ANY model].
TAD ARGONE
AND ARGTWO
CLL RAL
CIA
TAD ARGONE
DCA ARGTWO
This works because .XOR. is addition ignoring the carry bits. So, knowing
they will happen, just allow them at first, then remove them.
OK, all that said, I have another problem: On the LINC-8 there is no
divide, but there is an "inverse multiply" [an alternate operating mode of
the regular divide]. Can anyone explain what this is exactly [I suspect it
is a "poor man"s divide].
It's standard in all LINC machines [Classic LINC, LINC-8, PDP-12, B D
Speare Micro-LINC 100 and 300].
cjl [also it's a one's complement machine].
ps: The major registers of the LINC-8's MQ-type register uses the very
same R212 modules used for straight-8 EAE.
On Wed, Dec 23, 2015 at 12:24 PM, CLASystems <clasyst...@gmail.com> wrote:
> The MQ is standard on the PDP-8/E and up through the VT78 and the DECmates.
>
> Not possible on the 8/L, but an easy option for the PDP-8/I and the PDP-12.
>
> This could be a problem on the straight-8 [or LINC-8] with EAE though.
> It's the only model where the SWP [MQA and MQL in the same instruction]
> doesn't work [in case that matters].
>
> cjl
>
>
>
> On Wed, Dec 23, 2015 at 11:30 AM, Bob Vines <bobvine...@gmail.com> wrote:
>
>> Has anyone written PDP-8 code for .XOR. that uses the MQ Register (when
>> the EAE isn't available)? If so, can you share the source code and/or the
>> algorithm?
>>
>>
>> Thanks,
>>
>> Bob
>>
>
>
>
> --
> "In the future, OS/2 will be on everyone's desktop"
>
> Bill Gates, 1992
>
--
"In the future, OS/2 will be on everyone's desktop"
Bill Gates, 1992
