A good blockfactor for 6250 bpi tapes?

[Chris Torek]

In article <687 at maxim.erbe.se> prc at erbe.se (Robert Claeson) writes:
>I generally use blockfactor 20 when I'm reading and writing
>1600 bpi tapes, but I haven't found a good blockfactor for
>6250 bpi tapes yet.

Nine track tapes have poor error recoverabilty beyond 10240-byte
records at 1600 bpi.  Error correction on GCR drives is better, so that
although the bits are packed more densely, records of up to 32768 bytes
are generally safe (`GRAS' is the sci.med-ish term :-) ).

Tar's `blocking factor' values are in terms of 512 bytes: `b 20'
is 10240 bytes, and `b 64' is 32768 bytes.

Cartridge tapes (generally *not* 6250 fci) are often implemented
such that the actual records on the tapes are always 512 bytes long,
no matter what size is used to write to them.  Here only write
(or readback) speed matters when choosing a block size.  Otherwise,
larger blocks are generally better as they waste less tape on
inter-record gaps.
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163)
Domain:	chris at mimsy.umd.edu	Path:	uunet!mimsy!chris