30 Mar
2000
30 Mar
'00
12:18 a.m.
Since alt.sys.pdp10 is in a big flamewar right now, I guess I'll ask here.
Does anyone know the correct geomerty of an RP04 pack in 18-bit mode?
I have it as 19 heads, 411 cylinders, and 20 sectors a track, with 260
word sectors. (18 bit words, 256 data words and 4 header words.)
I think that's wrong though.
-------
30 Mar
30 Mar
1:35 a.m.
On Thu, 30 Mar 2000, Daniel A. Seagraves wrote:
Since alt.sys.pdp10 is in a big flamewar right now, I
guess I'll ask here.
Does anyone know the correct geomerty of an RP04 pack in 18-bit mode?
I have it as 19 heads, 411 cylinders, and 20 sectors a track, with 260
word sectors. (18 bit words, 256 data words and 4 header words.)
I think that's wrong though.
From the TOPS-10 7.03 monitor source (file RPXKON.MAC),
I see BLKPRU
(blocks per unit) set to 154280 with the comment "406
CYLINDERS", BLKPUM
(blocks per unit including maint cylinders) is 156180 with the comment
"411 CYLINDERS", and BLKPUC (blocks per unit in 10/11 compatibility mode -
what is this? Does it mean 16-bit mode?) as 171798, with the comment
"22*19*411".
If you do the math, 411*19*20 does equal 156180, so it would appear that
the geometry you have is correct.
I don't remember offhand about the number of words per sector - I was
thinking that there were 5 header words (2 cylinder address, two
"keywords", and one CRC or something?). But I'm likely to be wrong there,
and my peripherals handbook is at home and I'm at work right now.
--Pat.
3:42 a.m.
On Thu, Mar 30, 2000 at 06:18:55AM -0800, Daniel A. Seagraves wrote:
Since alt.sys.pdp10 is in a big flamewar right now, I
guess I'll ask here.
Does anyone know the correct geomerty of an RP04 pack in 18-bit mode?
I have it as 19 heads, 411 cylinders, and 20 sectors a track, with 260
word sectors. (18 bit words, 256 data words and 4 header words.)
I think that's wrong though.
Why? It's the exact same drive and media as used on a PDP-11, just formatted
to 20 sec/track instead of 22. So the cyls and tracks ought to be the same,
just fewer sectors since they're 11% longer. I certainly don't *know* that
it's the same (although I'll bet looking at the source to the ITS
"MARK"
routine would be enlightening), but why wouldn't it be? And of course the
headers should be invisible unless you use opcodes 53/63/73.
John Wilson
D Bit
8799
days inactive
8799
days old
2 comments
3 participants
participants (3)
-
DSEAGRAV＠toad.xkl.com -
pat＠transarc.ibm.com -
wilson＠dbit.dbit.com