Re: 2-session DVD: Linux shows new session's data while windows shows old sessions

[Zhang Weiwu <zhangweiwu@realss.com>]

On Mon, 23 Apr 2012, Thomas Schmitt wrote:

> That might make a difference to the reading drive.
> Some drives present some types of closed DVD as "DVD-ROM". It might be that
> MS-Windows does not expect a DVD-ROM to have multiple sessions.
> So if the drive does this with DVD-R but not with DVD+R, then this
> might be the trigger.
>
> Maybe some MS-Windows software can tell you more about how the drive
> presents the medium. If not, consider to boot a Live-CD with Linux
> that contains dvd+rw-mediainfo, cdrskin, or xorriso. E.g. "RIPLinux".

> > 1. Doses Windows/Linux handle multi-session differently?
>
> Either this, or the drive makes the difference, or both.

I tried to reproduce another multi-session DVD-R from the same pack, 
except I use 'cdrskin -multi' this time. Then I did a few tests:

FreeBSD 9.0	Windows				Hardware
last session*	first session (xp&vista)	DVD-R in DVDRW
first session	first session (2000)		DVD-R in DVD-ROM
last session*	last session  (xp&vista)	DVD-R (-multi) in DVDRW
first session 	first session (2000)		DVD-R (-multi) in DVD-ROM

Test notes:

	* asterisk in FreeBSD 9.0 column means Linux is also tested and
	behavior is the same as FreeBSD.

	I believe Linux behave like FreeBSD.
	Why bother testing FreeBSD on Linux mailing-list? I have to,
	because the only DVD-ROM I have is equiped with FreeBSD and all
	portable devices are DVD writers, thus a complete test result is
	only possible with FreeBSD.  For those who prefer Linux over
	FreeBSD with a passion: you might noticed Linux is too well
	equipted to do full test result.

The following conclusion is with DVD-R media only:
	1. DVD-R in DVD-ROM always result first session mounted.
	2. In Linux, DVD-R in DVDRW always result last session mounted.
	3. In Windows, usually first session is mounted, except DVD-R
	with open session (-multi) inserted in DVDRW driver. Since you
	cannot garantee that condition, the best practise is not to
	write an addiontal session hoping to use it on Windows. If you
	want to smartly hide the last session from Windows users, you
	should not use -multi on the last track, to avoid it incidently
	revealed on some Windows hardware.

> If my theory is right, then the effect depends on drive firmware and
> media type. Far from being reliable.

Since exactly the same (portable) hadware and media type (DVDs from the 
same package) behave differently on different OSs, there is at least a 
case when OS makes the difference.

Hope all these experiments being informative:)

Best regards
Zhang Weiwu