15 Aug
2015
15 Aug
'15
7:24 a.m.
This looks like fun..
http://mightyframe.blogspot.com/2015/08/qic-24-tape-data-block-format-decod…
16 Aug
16 Aug
8:24 a.m.
I left a note there about using CRCs to do data correction. It is
a simpler method for software than normally use.
Dwight
Date: Sat, 15 Aug 2015 07:24:24 -0700
From: aek at bitsavers.org
To: General at classiccmp.org; classiccmp.org:On-Topic and Off-Topic Posts cctalk at
classiccmp.org
Subject: QIC-24 data analysis
This looks like fun..
http://mightyframe.blogspot.com/2015/08/qic-24-tape-data-block-format-decod…
9:19 a.m.
On 08/16/2015 08:24 AM, dwight wrote:
I left a note there about using CRCs to do data
correction. It is a
simpler method for software than normally use.
I posted a link on VCF about using GNU CRC RevEng--a great little
utility to aid in determining polynomials and initial conditions for
CRCs. Not too different from figuring oddball floppy encodings.
What I'd like to see is a QIC mechanical solution that treats DC carts
as a reel-to-teel drive--no tension band worries. I imagine this could
be done by motorizing the takeup and supply reels and putting in a
driven capstan with suitable RPM control. Head positioning, of course,
could be handled the normal way. Perhaps an old top-of-the-line audio
cassette transport could be adapted.
--Chuck
10:15 a.m.
On 8/16/2015 11:19 AM, Chuck Guzis wrote:
On 08/16/2015 08:24 AM, dwight wrote:
More likely, apply the technology used on some 9 Track drives: a little
tension arm and then use servos to drive the supply and takeup reels,
and an encoder (*not* driven) that measures speed (in place of the
capstan on older drives).
I left a note there about using CRCs to do data
correction. It is a
simpler method for software than normally use.
I posted a link on VCF about using GNU CRC RevEng--a great little
utility to aid in determining polynomials and initial conditions for
CRCs. Not too different from figuring oddball floppy encodings.
What I'd like to see is a QIC mechanical solution that treats DC carts
as a reel-to-teel drive--no tension band worries. I imagine this could
be done by motorizing the takeup and supply reels and putting in a
driven capstan with suitable RPM control. Head positioning, of course,
could be handled the normal way. Perhaps an old top-of-the-line audio
cassette transport could be adapted.
--Chuck
1:22 p.m.
On 08/16/2015 10:15 AM, Jay Jaeger wrote:
More likely, apply the technology used on some 9 Track
drives: a little
tension arm and then use servos to drive the supply and takeup reels,
and an encoder (*not* driven) that measures speed (in place of the
capstan on older drives).
Maybe--I know the method--you see it on some Cipher drives and it's even
on my tape cleaning machine. A capstan-pinch roller, I think is no
worse and might actually help with wrinkled QIC tapes that have made a
mess inside of a QIC cartridge when their tension belts have rotted out.
--Chuck
3:20 p.m.
On 8/16/2015 3:22 PM, Chuck Guzis wrote:
On 08/16/2015 10:15 AM, Jay Jaeger wrote:
One issue with the capstan idea, though, would be direction. Capstans
really want to pull. Not very good a pushing. ;) So if you did that,
and you wanted to direction operation (required if you can't stream the
entire tape), then you'd need two sets of capstans like the big IBM and
the like 7 and 9 track tape drives had.
More likely, apply the technology used on some 9
Track drives: a little
tension arm and then use servos to drive the supply and takeup reels,
and an encoder (*not* driven) that measures speed (in place of the
capstan on older drives).
Maybe--I know the method--you see it on some Cipher drives and it's even
on my tape cleaning machine. A capstan-pinch roller, I think is no
worse and might actually help with wrinkled QIC tapes that have made a
mess inside of a QIC cartridge when their tension belts have rotted out.
--Chuck
4:29 p.m.
On 08/16/2015 03:20 PM, Jay Jaeger wrote:
One issue with the capstan idea, though, would be
direction.
Capstans really want to pull. Not very good a pushing. ;) So if
you did that, and you wanted to direction operation (required if you
can't stream the entire tape), then you'd need two sets of capstans
like the big IBM and the like 7 and 9 track tape drives had.
Or CDC... :)
Or maybe just a tachometer and let the reel motors do the work. QIC
data is pretty much self-clocking (but for those HP 2-track drives).
Still mulling it over--the number of QIC cartridges I've had to handle
has gone down substantially over the years.
--Chuck
9:20 p.m.
On 8/16/2015 6:29 PM, Chuck Guzis wrote:
On 08/16/2015 03:20 PM, Jay Jaeger wrote:
Exactly.
One issue with the capstan idea, though, would be
direction.
Capstans really want to pull. Not very good a pushing. ;) So if
you did that, and you wanted to direction operation (required if you
can't stream the entire tape), then you'd need two sets of capstans
like the big IBM and the like 7 and 9 track tape drives had.
Or CDC... :)
Or maybe just a tachometer and let the reel motors do the work. QIC
data is pretty much self-clocking (but for those HP 2-track drives).
Still mulling it over--the number of QIC cartridges
I've had to handle
has gone down substantially over the years.
--Chuck
3416
days inactive
3418
days old
7 comments
4 participants
participants (4)
-
aek＠bitsavers.org -
cclist＠sydex.com -
cube1＠charter.net -
dkelvey＠hotmail.com