Re: DSSI Disks

On 2026-05-03 20:42, Paul Koning wrote:
>=20
>=20
>> On May 3, 2026, at 1:19=E2=80=AFPM, Johnny Billquist <[email protected]> =
wrote:
>>
>> On 2026-05-03 19:03, Andrew Ball wrote:
>>> =C2=A0=C2=A0ASB> I notice that ra0 and ra1 are reported as the wrong =
type of
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> drive (RF31 instead of RF35).
>>> =C2=A0=C2=A0=C2=A0JB> The names are basically just strings returned f=
rom the
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> simulator. =C2=A0They have no relevan=
ce beyond being just a string
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> that you can print out.
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0>
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> Size/capacity similarly is reported a=
s what the actual capacity
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> is, without any regards to names.
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0>
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> So this is all just cosmetic, and it'=
s a thing that you need to
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0> fix in the simulator if you want it t=
o say something else.
>>> =C2=A0=C2=A0=C2=A0=C2=A0Thanks Johnny (and thanks Michael for the cor=
rection). =C2=A0I'll try
>>> some other disk types to submit a more helpful report to the open-sim=
h
>>> people. =C2=A0RF71 works:-
>>> ra3 at mscpbus0 drive 3: RF71
>>> ra3: no disk label: size 781440 sectors
>>
>> I wasn't even aware that simh could report "RF71". simh have been=20
>> pretty stubborn on not accepting that it should be an arbitrary=20
>> string, and have just a few fixed values hardcoded in the code.=20
>> (Atleast when I last checked.)
>>
>> But honestly, I would just ignore this "problem". Like I said - it's=20
>> very much just a cosmetic thing.
>=20
> I would have thought the "type" is chosen based on the disk image size,=
=20
> but judging by the examples given that isn't the case. =C2=A0I wonder w=
hy not.

simh could possibly do that for known sizes, yes. But even that isn't=20
clear cut. The RF70-73 are the same as the RA70-73, for example.
But simh just seem to have a very strange idea that identifiers somehow=20
are very fixed. I was surprised they added some RF disks. It used to be=20
that simh used RA81 for almost everything. Then a few more types of RA=20
disks were added, and a few other odd bits. But it's always been rather=20
hardcoded.

> And yes, technically it's a string that should be configurable to=20
> anything the MSCP protocol can convey. =C2=A0But it would be nice if at=
least=20
> it produced identifiers matching the media size.

When that exists, sure, it could provide some defaults. But it's also a=20
question of why. Because the names have no actual meaning.

For MSCP, the only thing that actually matters is the size. Names can be=20
completely ignored. They are only useful for printing something for human=
s.

   Johnny

--=20
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: [email protected]             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol