On Fri, Jan 24, 2003 at 02:53:26PM +0100, Martin Oehler wrote:
> Hi!
> 
> I've run tapetype on a LTO Ultrium drive and some questions concerning
> the results. The tapetype test has been made twice, once with 
> hardware compression (HC) turned on (by mistake) and once with  HC 
> turned off.
> 
> The OS is Linux and 'mt' is telling me actually:
> # mt -f /dev/nst0 datcompression 1
> Compression off.
> 
> The results of the first run (with HC) were:
> 
> define tapetype unknown-tapetype {
>     comment "just produced by tapetype program"
>     length 104261 mbytes
>     filemark 545 kbytes
>     speed 1601 kps
> 
> }
> 
> Results without HC:
> 
> define tapetype unknown-tapetype {
>     comment "just produced by tapetype program"
>     length 104302 mbytes
>     filemark 547 kbytes
>     speed 1603 kps
> }
> 
> Why is the length the same? Seems like switching the compression modes
> with mt is not working. 

Maybe you hit what Gene describes in another posting today.
Once a tape has been written to with HW compression, the drive senses it
and automatically switches to HW compression regardless of settings.

What capacity were you expecting for "native", uncompressed data?

> Additionally, I can't understand why the speed is that slow. The drive
> should write about 15 MB/sec without HC. Is it possible to ignore
> the measured speed and use the one from the datasheet of the tape?

According to docs/TAPETYPE:

        The speed is currently unused.

Is this an HP drive by any chance?  ISTR they have some auto-slowdown
system to match scsi-bus speed.

-- 
Jon H. LaBadie                  jon AT jgcomp DOT com
 JG Computing
 4455 Province Line Road        (609) 252-0159
 Princeton, NJ  08540-4322      (609) 683-7220 (fax)