MMC/MtFuji: Confusion in the interpretation of DVD Forum book for implementation of Close Track/Session command
David.Burg at microsoft.com
Wed Jun 24 13:36:28 PDT 2009
* From the T10 Reflector (t10 at t10.org), posted by:
* David Burg <David.Burg at microsoft.com>
(Posting one more time due to e-mail account majordomo registration issue
fixed by John)
(Re-sending in plain text to avoid character conversion problems)
Microsoft would like to request the hardware manufacturers from the MMC and
Mt Fuji committee to study the following difference of interpretation of the
DVD Forum book specification for DVD-R/-RW related to the implementation of
the Close Track/Session command from the Mt Fuji/MMC specification.
Microsoft$B!G(Bs product Windows issues this command to DVD recorder
devices (to close both track and session, creating a border out on DVD-R/RW)
and does not get a consistent behavior between drives from different vendors.
In this introduction of the issue, Microsoft will refer to the vendors as
vendor $B!F(BA$B!G(B and vendor $B!F(BB$B!G(B.
We would like the difference to be clarified so vendors may provide a
consistent implementation for the host command reaction. The current
difference results in radically different execution time, in the most extreme
case a difference from 2 minutes to 13 minutes(!).
1) Vendor $B!F(BA$B!G(B
Vendor $B!F(BA$B!G(B reads DVD-R Part1 PHYSICAL SPECIFICATION Version2.0:
Outer diameter of Information Area
1) less than 68.0 mm $B"*(B 70.0 mm min.
2) 68.0 to ($B#A(B - 1.0 mm) $B"*(B Outer diameter of Data Area $B!\(B
2.0 mm min.
3) ($B#A(B - 1.0 mm) to A $B"*(B $B#A(B + 1.0 mm min.
70mm LBA = near 82588hex
2048byte x 82588hex = 1.093419008 x 1E + 9 byte $B"b(B 1.1GB
The speed of 1x is 11.08Mbps in DVD,
If users write DVD-R with 1x,
1.1GB/11.08Mbps $B"b(B 13 min 34 sec.
2x is 22.16Mbps.
1.1GB/22.16Mbps $B"b(B 6 min 47 sec.
4x is 44.32 Mbps.
1.1 GB/44.32 Mbps = 3 min 24sec.
2) Vendor $B!F(BB$B!G(B
Vendor $B!F(BB$B!G(B looked at the 3.4 of DVD-R Spec version 1.1.
Leadin area structure started from 022FA0 h to 30000h. ( 53344 sectors )
Border out Area can be changed according to the written data size.
Vendor $B!F(BB$B!G(B thinks that worst case should use 92MBytes for the
Border Out (165700h+ PSN of beginning Border Zone, for first border zone
case). Worst case will be 47904 sectors for Border out.
So close session recording time excluding recording seek time will be
53344+47904 / 1385 = 73 sec
Microsoft does not know which of the two interpretations is correct. I hope
that the hardware manufacturers present on the committee alias can resolve
this difference and let know what the proper interpretation of the
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org
More information about the T10