>Not really. AIT and DAT and I assume M2 spin the heads and slow down the
>tape but the relative speeds are in the same neighborhood. (I assume
>it's easier to spin the heads faster than move the tape faster which is
>why DLT appears to be falling behind in the race.) Anyway, you still
>need to keep the data flowing at the speed of the drive or it will stop
>streaming and get into tape stuttering or rewinds to reposition. This
>also causes a large loss in tape capacity as there's a lot of recording
>overhead in starting or stopping a stream.

The loss in capacity comes from the drive's attempt to continue tape
motion while waiting for new data to arrive in the write buffer.
No data is recorded while the tape continues to move forward.
This "long gap" strategy has been in use since the advent of
6250 bpi 1/2" reel-to-reel drives. The latest generation of cartridge
drives, whether or helical scan all implement some form of
intelligence, in the drive, to slow the tape transport or insert
gaps in the data and thus adapt to the rate at which data is arriving
to the drive. If network congestion or system loading eventually
exceed some limit, the drive must fully stop the tape motion
and rewind slightly in order to reposition before restarting
recording when the data level in the buffer reaches a high water mark.
Repositioning of DLT drives is a simpler process than for
helical scan drives due the latter's capstain effect (binding friction
due to head rotation and tape motion interference). Sailor's
used to the effect of a "cat head" will find this concept to be intuitive.
A linear drive (DLT, LTO, 3590E etc)  does not need to reduce tape
tension during the reverse motion. A helical scan drive
(Exabyte (all), DDS-DAT, AIT and high end drives from Sony and Ampex)
must be much more careful in its handling of the tape to avoid stretching
of the media due to slapping or binding of the tape loops during the change of
motion direction.

The wear-and-tear effect of tape repositioning (aka "shoe shining") is
the predominant factor in the differences in drive reliability and
performance observed by different users. Those who have had bad
exeriences with one drive type or another can most often trace an
explanation back to system and network loading or limited bandwidth,
memory or other resources...not to the drives.

Houston TX

To subscribe:    [EMAIL PROTECTED]
To unsubscribe:  [EMAIL PROTECTED]
Archives:        <http://list.working-dogs.com/lists/retro-talk/>
Search:  <http://www.mail-archive.com/retro-talk%40latchkey.com/>

For urgent issues, please contact Dantz technical support directly at
[EMAIL PROTECTED] or 925.253.3050.

Reply via email to