Hi all,

I'm about to back up some HUGE files (ca. 1 GiB each) onto DVD+R's

I plan on using also QuickPar to make some .par2 files to protect them.

But I'm a greenhorn when it comes to .par2. So my questions are:

1. Should I make a .par2 per file, or should I make a .par2 for the whole collection of files?
2. I've read somewhere else in this forum that QuickPar can repair a damaged ISO. For that to work, must I put the .par2 files in the same DVD or may I put them on a different DVD?
   
   1. In relation to Q.1 above, if I create .par2 files per file, and put them in the same DVD, if I try to repair a damaged ISO, will it work?
3. And what parameters do you recommend to protect DVD+R's (e.g., block size, .par2 file size and/or number of files, % redundancy, etc.)

Thanks beforehand for your kind help

I've never used it myself, but have you considered dvdisaster instead?

Hi pepoluan. I played around a little bit with Quickpar a long time ago bit I ended up using dvdisaster instead. See download here

Dvdisaster is pretty easy to use and though I've only needed to recover a bad DVD once it worked flawlessly for me. I just used the default parity file settings which I think were about 15% of the main image file size.

@pdq & uart: Thanks for the reminder. I used to use DVDisaster and somehow forgot all about it. Will revisit

Anyways, I also find a nice discussion about QuickPar, and this post kinda summarize the whole thread:
http://www.hydrogenaudio.org/forums/index....st&p=271470

I'll check out all options; these files are waaaay to0 important to not be protected by something.


Edit: misspelling

Hmmm... anyone ever have any problem with DVDisaster-augmented DVDs ?

I mean, have anyone ever experienced a DVDisaster-augmented DVD to be not readable in a standard DVD drive?

Although I've never had any problems with RS02 (augmented iso image) protection method, DVDisaster page warns that media burned from augmented images may be not readable on some DVD units...

http://dvdisaster.net/legacy/en/background30.html#image (last row in comparison table).

Better test your drive on some less important data before trying to protect anything precious...

Messer

I don't quite see the value of creating parity files for DVDs, but I'll try to answer your questions anyway.


I'd do parity sets for each file.


The parity files can be stored anywhere. You can store them on the same DVD, a different DVD, a flash drive, a hard drive, etc. For the easiest, smoothest validation and reconstruction, you should keep the PAR2s in the same directory as the files.


If you have enough blocks to repair the ISO, yes.


The amount of redundancy is what you need to be concerned with. That's entirely dependent on how much space you wish to use for parity.

Rather than testing, I think I'll go the RS01 path instead.

Besides, if the DVD ever needs resurrection, I still need to do it with DVDisaster anyways. So better a guaranteed-no-problem DVD for the datafiles plus an ECC DVD.




Thanks for the answers, Ron. After some thoughts, I think I'll use the DVDisaster RS01 path instead. But most likely will use QuickPar to protect the (separate) ECC files.

When I was using DVDisaster I would burn the parity (ecc) files onto different discs. Typically I would burn about 8 DVD's and make an ecc file for each of them, then I'd just burn a 9th DVD containing all 8 ecc files.


I know where you're coming from Ron. It's a bit of extra work and there's really no guaranty that it will even work when you need it. The problem with optical media is that unless you check it fairly regularly it can degrade too much to even be recognized in a drive and then your parity files are useless. Never the less it does add some measure of reliability. One time I couldn't copy a particular DVD because it had read errors (crc errors) but fortunately I'd previously made a dvdisaster ecc file and I was fully able to repair and re-burn it. Personally though for valuable data I think it might be better to just make two complete copies of each DVD.

Are you sure 12.5% redundancy is a good idea?

Anyways, I plan to use 14.something% redundancy as according to the DVDisaster makers, the latest version of DVDisaster is optimized for that and 33.something% (so it will make them fast)



If both copies happen to have error at the similar location (unlikely, but still possible), you'll lose the valuable data.

No I think I the DVD's I were burning weren't quite using all the DVD capacity, something like about 4000 MB average. So I was using about 14% redundancy and still fitting 8 ecc files onto one DVD. Anyway it wasn't always exactly 8, that was just a typical figure.


Yeah nothing is really going to be 100% failsafe. A big worry for me is that I have seen optical media that has just gone completely bad during storage and was unreadable (as in the file system wasn't even recognized) in any drive I tried it in. I know that's rare but it does happen.

If you store the PAR on the same disc as the actual data, what's the point? If the disc gets damaged (scratches, degradation due to UV, etc.), you lose both the actual files and the PARs. BTW, any special reason why you want DVD+R over HDDs?

While at it, I also thought DVD-RAM was safer but according to an article in the German c't, RAM media is a bad target for backups because ODD manufacturers (according to some insiders) tune the write strategies only for +R and -R discs and not for RWs and RAMs. Only the stand-alone video DVD recorders have to pass a wider set of tests with RW and RAM discs.