The only reason I'm answering is that I have reasons to believe that I
indeed failed to find up-to-date or accurate enough information about
DVD- formats. The only specification *publicly* available is 3.95GB
DVD-R ECMA specification which describes certain Linking scheme that
applies in Incremental Recording mode (a.k.a. packet writing, right?).
Is this scheme applicable even to 4.7GB DVD-R and/or DVD-RW?
The write speed does not matter if you use BUrnProof or packet writing.
Of course DVD+RW _only_ supports packet writing...
The page in question was updated with following two paragraphs:
What does plus stand for in DVD+RW/+R? The key feature of DVD+RW/+R
media is high [spatial] frequency wobbled [pre-]groove with addressing
information modulated into it. This makes it possible to resume
interrupted [or delibirately suspended] burning process with accuracy
high enough for DVD[-ROM] player not to notice anything at playback
time. Recovery from buffer underrun condition in DVD-RW/-R case in turn
is way less accurate procedure, and the problem is that the provided
accuracy is very much what average player can tolerate. Now given that
both provided and tolerated inaccuracies are proportional to respectively
writing and reading velocities there basically no guarantee that
DVD-RW/-R recording that suffered from buffer underrun will be
universally playable.
This is not correct.
High frequency wobbling does not help to make the DVD recording better
in case of buffer underruns. You always use the pre-groove only to find
the course place where to resume recording. The exact position is found
by reading the already recorded data and switching from read to write.
Well, explain following then. I can take *virgin* DVD+RW media,
partially format it so that only lead-in gets written out as well as
probably a small tail after it gets de-iced. This takes ~40 seconds.
Then I can immediately seek to say 4GB, write some data and immediately
eject the media. Another 10-20-30 seconds passed depending on how much
I've chosen to write. I can flip the disc and note that media between
lead-in and written data remained *virgin*. Now if exact positioning
could be determined only by playing earlier written data how come it
went so fast and surface in between remained virgin? If it weren't
accurate enough how can they guarantee that I'll be able to squeeze 4GB
in between lead-in and data written at 4GB offset?
Sometimes DVD+RW/+R is erroneously compared with packet writing. Packet
writing means that every chunk of actual user data gets surrounded by so
called link, run-in and run-out sectors, which wastes some capacity, not
to mention that it requires explicit support by player's DSP [Digital
Signal Processor] performing the actual decoding of user data. Again
thanks to high frequency wobble, no such things are needed for DVD+RW/+R.
This is exactly why it's commonly referred to as designed from scratch
for maximum compatibility with DVD-ROM specification.
Call it ho you like, it _is_ some kind of packet writing.
Well, yet the fact is that whatever we call it, the result is
*indistinguishable* from whatever that might appear as TAO/SAO/DAO.
Of course, there is no run-in/run-out blocks as the DVD uses a different
scheme for cross interleaved error correction.
Yes, I have to admit that I'm apparently wrong about these
run-in/run-outs [they do belong in CD recording] and will correct the
page as soon as I get clear picture of DVD-R[W].
MMC specifications mention link data in DVD-R[W] context as well as
border-in/-out in DVD-R incremental recording context. Does either hold
true? If yes, what's role of border-in/-out? Most importantly. Is
explicit support by DSP [performing the actual decoding of user data]
required for playback of DVD-R[W] recorded in packet mode?
Then I fail to understand why did you bring up packet writing at all?
cdrecord[-ProDVD] doesn't support it so that those willing to perform
say live NFS backups have to solely rely on buffer underrun protection,
don't they? Is there reason to believe that recovery is so much more
accurate than in CD [where it's off by default] that you're ready to bet
your backup on it?
- write packets to a blank disk (no overwrite possible)
- write random addressed packets to a pre-formatted disk (overwrite works).
DVD- supports both modes, DVD+ only supports the second mode.
In addition, DVD- supports a streamed de-facto uninterrupted write called
SAO (Session At Once).
As already mentioned. DVD+RW supports write of randomly addressed 2KB
blocks even to *virgin* media (well, when you write 2KB, 32KB gets
naturally written/de-iced], and naturally rewrite in random order at
later occasion. Randomly written DVD+RW disc is *indistinguishable* from
one written progressively/streamed. DVD+R media can be written
progressively in 32KB ECC blocks, *but* uninterrupted streaming is not a
requirement [you can even eject media
