One rip failed

[GoogleCodeExporter]

When I rip track 10 "Strength to Dream" (incorrectly titled "Strenght to Dream" 
in the metadata) from the Propaganda album "A Secret Wish", the resulting DSF 
file is corrupt.

Anyway, it seems to be lacking a header of any kind.  All the other DSF files 
begin with this header (catted through "strings"), but this one doesn't:

DSD 
fmt 4
data^L

If you convert it direct to a WAV file (I'm using a Mac program called 
DSDConverter to do this), it assumes the file is mono, and the result is not 
quite white noise, but definitely unlistenable.

I tried to prepend the above header to the DSF using a hex editor, and 
converted it to a WAV again.  The results were better (eg. it is recognised as 
a 6 channel file now), but there is a now very audible crackly interference 
throughout the whole track.

Admittedly, I'm using an earlier version of sacd-ripper, svn version 204 if I 
recall correctly.  Would this have been fixed since then?    Regrettably I 
can't try again anyway, because my wife has forced me to upgrade the PS3 now.

Also, any suggestions on how I can repair the file, other than what I already 
tried?  Here is the first part of the file in hex (meaningless, probably):

66 96 A6 99 59 5A 65 56 56 65 55 95 D5 54 36 55 35 55 4D 35 4D 53 55 2B D5 4C 
D3 B4 D4 CC 4C 2D 2D 4D CB 4C 33 D3 32 CD B4 B4 4C 2D CD 4C 33 2D 4D 4B 4D 33 
33 53 B5 34 33 4D 35 33 35 CD D4 34 53 35 53 33 D5 D4 34 35 55 B5 54 53 E5 4C 
D5 54 D6 54 55 65 53 59 D9 64 59 65 56 55 9B AA AA A9

Here it is after patching the header (you'll see the first bytes of the data 
above beginning on the last line):

44 53 44 20 1C 00 00 00 00 00 00 00 61 01 B2 1D 00 00 00 00 5C 00 B2 1D 00 00 
00 00 66 6D 74 20 34 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 07 00 00 00 
06 00 00 00 00 11 2B 00 01 00 00 00 00 00 98 27 00 00 00 00 00 10 00 00 00 00 
00 00 64 61 74 61 0C 00 B2 1D 00 00 00 00 66 96 A6 99

Original issue reported on code.google.com by Jer...@Malcolm.id.au on 6 Jul 2011 at 1:28

[GoogleCodeExporter]

I have had a number of fails, most of which generated pretty much nothing in 
the sacd_log file.  However, today's attempted DSDIFF rip of the MFSL Blues In 
Orbit (a hybrid) by Duke Ellington failed and yielded the attached log.  The 
end is some kind of repeating loop, which explains why the log file is 120K.

The disc is recognized by the PS3 as an SACD.

Interestingly, I was able to .ISO this disc and recreate a playable SACD-R.

Original comment by djmcka...@gmail.com on 7 Jul 2011 at 9:34

Attachments:

• sacd_log.txt

[GoogleCodeExporter]

@qirtaiba:

DSF is the Sony output, I haven't used DSDConverter myself but it probably uses 
DSD(IFF) files instead of DSF files. 

Original comment by mr_wic...@hotmail.com on 8 Jul 2011 at 11:32

• Changed state: Invalid

[GoogleCodeExporter]

@donald:

Please re-create this issue so we can have a proper bug report on that. What 
you report is completely valid and has to do with Blues in Orbit not being 
encrypted at all.

Original comment by mr_wic...@hotmail.com on 8 Jul 2011 at 11:34

[GoogleCodeExporter]

Not so, 30 other tracks have ripped OK. Only this one failed.

Original comment by Jer...@Malcolm.id.au on 8 Jul 2011 at 1:50

[GoogleCodeExporter]

can you open the track in Audiogate? And convert to DSDIFF or save as DSF 
again? If it's playable in audiogate, it's probably ok.

Original comment by mr_wic...@hotmail.com on 8 Jul 2011 at 3:44

[GoogleCodeExporter]

Audiogate only supports 2 channels.  The discs I have been ripping are all 
multi-channel.

Original comment by Jer...@Malcolm.id.au on 9 Jul 2011 at 1:00

[GoogleCodeExporter]

I have attached the product of 3 failed rips using Analysis Dumper.  Two of 
them are MFSL Hybrid SACD and the third is an older Sony SACD only disc.

Original comment by djmcka...@gmail.com on 9 Jul 2011 at 6:06

Attachments:

• [Failed SACD Rips - Analysis Dumper.zip](https://storage.googleapis.com/google-code-attachments/sacd-ripper/issue-20/comment-7/Failed SACD Rips - Analysis Dumper.zip)

[GoogleCodeExporter]

Here are 3 more.  All of these are fairly new releases.

Original comment by djmcka...@gmail.com on 9 Jul 2011 at 6:24

Attachments:

• [SACD Failed Rip Dumps_2.zip](https://storage.googleapis.com/google-code-attachments/sacd-ripper/issue-20/comment-8/SACD Failed Rip Dumps_2.zip)

[GoogleCodeExporter]

Each of these ripping issues was fixed by upgrading my .pkg file to 0.20 except 
for the non-encrypted discs.

Original comment by djmcka...@gmail.com on 10 Jul 2011 at 12:06

[GoogleCodeExporter]

Can you take a diff between the failed rips and the successful ones to see what 
the difference is?

Original comment by Jer...@Malcolm.id.au on 10 Jul 2011 at 1:05

[GoogleCodeExporter]

@qirtaiba:

the problem Donald reported is something completely different. 

Original comment by mr_wic...@hotmail.com on 10 Jul 2011 at 11:01

[GoogleCodeExporter]

qirtaiba, I'm the guy who makes DSDConverter. I can't say for sure that it 
correctly works with multichannel files (i.e. more than stereo) because I don't 
have any official files to test with. Only the ones (DSDIFF) I generate myself, 
but those might not be to spec.
Please try in DSDIFF format again, since this is better tested. There might 
indeed be a problem in DSF format. Maybe get in touch with me trough my website 
and send me a sample file if Mr. Wicked is confident that DSF output from 
sacd-ripper is correct.

Original comment by Olaf.O...@gmail.com on 14 Jul 2011 at 7:54

[GoogleCodeExporter]

Thanks Olaf, actually I can't rip it again because my wife made me upgrade our 
PS3's firmware.  Anyway, I have obtained a copy of the DSF spec from Sony and 
will try again to create a correct header for the file and convert to WAV again 
soon.  BTW, DSDConverter seems to work fine on multi-channel files and I don't 
think the problem is yours.  The problem is that the header was completely 
missing from the DSF file so DSDConverter didn't know what to do with it.

Original comment by Jer...@Malcolm.id.au on 15 Jul 2011 at 1:06