Up to version 5.12 the support was classified as "experimental," because it was based solely on user feedback. In time for 5.13 release LinuxFund.org has donated DVD-dash capable unit, which permits me to perform pre-release tests on site and thus renders the "experimental" status obsolete:-) |
Q. | How are dvd+rw-tools different from cdrecord-ProDVD? When it comes to DVD-R[W] burning that is... |
---|---|
A. | DVD-R[W] specification defines following recording strategies:
Now, cdrecord-ProDVD supports Disk-at-once strategy only, while dvd+rw-tools all the others. That's it:-) |
Q. | What is the difference between these recording strategies? Which one is the one for me? |
---|---|
A. | Disk-at-once is the one guaranteed to provide full compatibility with DVD-ROM/-Video. But apparently only as long as recording didn't suffer from buffer underruns (see this paragraph on +RW page for technical details). DAO recordings are unappendable. Oh! It should be explicitly noted that "full compatibility with DVD-Video" doesn't really mean that DVD-RW media recorded in this mode will be universally playable! DVD-RW reportedly suffers from compatibility issues very similar to DVD+RW. In other words these incompatibilities, both -RW and +RW, have way more to do with low reflectivity of these media, than with their logical formats. As for Incremental strategy [limited] DVD-ROM/-Video compatibility is provided at single packet/extent "level." Good news are that growisofs burns the whole filesystem image as a single extent, so that you have all the rights to expect it to come out as compatible as DAO recorded with [eventual] buffer underruns. Yes, buffer underrun protection gets engaged, so that you don't have to worry about how much extra load your system can sustain during the burning process. In addition Incremental strategy provides for multiple sessions and dvd+rw-tools do utilize it. But you should keep in mind that very few DVD-ROM players actually support DVD multi-sessioning! Even among those explicitly supporting DVD-R most will play till the first Border-out only. Restricted Overwrite (DVD-RW only) provides for arbitrary overwrite (well, with certain restrictions:-), meaning that it's possible to grow ISO9660 volumes within a "single session," in the very same way as with DVD+RW (or -RAM/plain file for that matter). So that if your DVD-ROM unit can play media recorded in this mode (it most likely just does as long as it can play DVD-RW at all), but doesn't support multi-sessioning, then this is the mode for you (provided that you wish to add data at different occasions that is). Buffer underrun protection is unconditionally on. As for compatibility with DVD-ROM/-Video. Apparently user Data Zone is equivalent to one recorded in DAO, but as if every 32KB block has suffered from buffer underrun. Disk-at-once and Incremental recording strategies are applicabe to DVD-R media and DVD-RW media in Sequential mode. Restricted Overwrite strategy is applicable to DVD-RW media explicitely formated for Restricted Overwrite. |
Q. | How do I switch between DVD-RW modes? |
---|---|
A. | Initially blank DVD-RW media is in
Sequential mode. To format for Restricted Overwrite invoke
' To change [back] to Sequential mode [or to
reuse the media in Sequential mode for a new dataset] invoke
' |
Q. | How do I burn? |
---|---|
A. | For practical burning instructions see the growisofs paragraph on +RW page. |
Q. | Any [other] caveats? |
---|---|
A. | When burning DVD-R[W] growisofs recognizes an extra command line option, namely -speed=X, which lets you choose the burning speed. If you use brand-name media, you're unlikely to have to use this option. Your unit will determine optimal velocity automagically. But if you use no-name media, -speed=1 might turn out to be a must. Problem is that some units reportedly fail to determine optimal speed for no-name media and pick higher one, higher than media can stand, which results in unplayable recording. This may vary from firmware to firmware (Pioneer users should consult this page). As for -speed option in more general sense.
You might find yourself in situation when media rated as 4x is
recorded at lower speed, even if you explicitly specify
-speed=4. This is because that particular brand is poorly
supported by your particular unit firmware. It's not an
application problem, as growisofs asks unit for list of
supported velocities for currently loaded media and picks
closest to one specified in command line. The list is
updated upon every media load and can be found in
Futhermore quoting SONY support page: "... some of discs on the market do not meet the recognized quality and performance standards." Question is how does it show? Apparently in bizarre ways... DVD-RW media might [reportedly] perform as write-once or maybe twice or three times. Blank/format procedure might appear successful, but deferred failure is reported upon actual recording attempt. It might be possible to blank the media in one unit, but not in another one [of different brand]. DVD-R recording might [reportedly] fail around 95% of advertised capacity... Or recording might fail at start-up with "INCOMPATIBLE MEDIUM" error code. At all these occasions switching to another media brand [reportedly] helped. As for DVD-ROM/-Video compatibility of Restricted Overwrite mode. Playability might be limited for reasons similar to DVD+ lead-out issues (see from 2nd paragraph), when legacy unit is tricked to calibration on virgin surface. In case you find media recorded in this mode unplayable, attempt to
In order to facilitate for the latter test and be excused from wasting an hour on full blanking procedure, growisofs version 5.6 implements rudimentary support for DVD-RW DAO recordings (yes, for DVD-RW only). Recall that DAO recordings come out unappendable, which goes so to say against growisofs' spirit:-) As of version 5.6 growisofs no longer relies on logical unit to do all the magic, but explicitly asks the burner to perform OPC, Optimal Power Calibration, prior recording. It seems that it's not necessarily a "good thing™," as it's maintained that media profiles pre-defined in unit firmware might be more optimal than those obtained as result of auto-calibration. I explicitly check if the unit returns any OPC descriptors in reply to READ DISC INFORMATION command [assuming that if it does, then pre-defined media profile is in effect], but it wasn't actually observed to be a case so far. If anybody can elaborate on this further, I appreciate a note... |