Skip to main content

Topic: SACD Ripping (Read 55077 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • Donunus
  • [*][*][*]
SACD Ripping
I have never used sacd but was wondering how big the uncompressed file out of an sacd is(Bitrate). Also, Is there a lossy encoder out there that can encode these files and actually do it justice?

  • rjamorim
  • [*][*][*][*][*]
SACD Ripping
Reply #1
First, you (normally) can't read SACD data using your standard DVD ROM drive.

Second, even if you could read it, the data is strongly encrypted.

Third, even if you could decrypt it, it's compressed by a proprietary lossless compression method.

Fourth, even if you could decompress it, the stream format (DSD) is unreadable by most audio processing software.
  • Last Edit: 02 October, 2005, 01:18:16 AM by rjamorim
Get up-to-date binaries of Lame, AAC, Vorbis and much more at RareWares:
http://www.rarewares.org

  • Defsac
  • [*][*][*][*]
SACD Ripping
Reply #2
SACD has copy protection on the physical level, similar to how Playstation 2 discs are copy protected. The data is encrypted and the key stored in an area of the disc not readable by anything except an SACD device. This is similar to the PS2 copy protection where a code is stored in an area of the disc which can't be read by conventional devices. This also prevents the code being written by a standard burner. Both technologies are designed so only pressing the disc will copy the code (or key in the case of SACD).

To read an SACD, you'd need a reader with an adjusted focal length suitable for reading SACD discs (the standard 1.2 mm length will give you the CD audio track if it's a hybrid or nothing if it's SACD only). The reader would also have to be capable of reading the encryption key. Even if you did this you'd be stuck with a Direct Stream Digital stream and as rjamorim said there are very few tools capable of working with a DSD stream.

The only feasible method of losslessly capturing SACD audio at the moment would be a device that captured data after the decryption stage but before the digital to analog stage. You'd need to modify a hardware SACD player to do this.
  • Last Edit: 02 October, 2005, 02:51:44 AM by Defsac

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #3
OK, What about the DVD audio format? Is there anything we can do with that one?

  • user
  • [*][*][*][*][*]
SACD Ripping
Reply #4
DVD-Audio is the format one should prefer for high resolution digital audio, stereo or multi-channel.
You can master and burn even your own DVD-A discs.

SACD has technical flaws by design, read here some sacd vs. dvd-a topics,
whereas DVD-A offers everything ones needs, ie.
eg. commercial DVD-aS COULD BE PLAYED ON pc; IF YOU HAVE SUITABLE SOUNDCARD, capslock off again;)
You can even record (eg. vinyl) by good soundcards in high resolutions, and burn your own DVD-A.
www.High-Quality.ch.vu -- High Quality Audio Archiving Tutorials

  • skamp
  • [*][*][*][*][*]
  • Developer
SACD Ripping
Reply #5
Yes, you can now rip DVD-Audio discs, although the whole process isn't exactly straight forward. I also found out that you can directly decode MLP files using Surcode MLP Encoder...
See my profile for measurements, tools and recommendations.

  • SebastianG
  • [*][*][*][*][*]
  • Developer
SACD Ripping
Reply #6
Quote
I have never used sacd but was wondering how big the uncompressed file out of an sacd is(Bitrate). Also, Is there a lossy encoder out there that can encode these files and actually do it justice?
[a href="index.php?act=findpost&pid=330983"][{POST_SNAPBACK}][/a]


The uncompressed bitrate is 64*44100 (= 2.82 mega) bits per second per channel (At least for stereophonic content. It seems too high for multichannel content, I'm not really sure here)

The best thing (IMHO) would be to ditch the DSD encoding and to convert the DSD stream to something like 88.2 kHz, 24 bit (or even 16 bit with clever noise shaping -- you have enough room above 25 kHz at that sampling rate!) which then result in 2.12 mega respective 1.41 mega bits per second per channel.

Lessless compression (FLAC, WawPack, ...) would be an option. If you want to go lossy then convert to 24 Bit/44.1 kHz instead and do AAC/Vorbis/MP3/....
Although 88.2 kHz is supported by AAC and Vorbis I don't think available encoders are optimized enough for that kind of input. They'll probalby eliminate everything above 20 kHz anyway in order to improve quality below 20 kHz. :-p

Quote
Fourth, even if you could decompress it, the stream format (DSD) is unreadable by most audio processing software.
[a href="index.php?act=findpost&pid=330986"][{POST_SNAPBACK}][/a]


This would be the smallest problem of all since DSD->PCM conversion is pretty simple. I already tinkered with a raw DSD stream sample (mono, about 14 sec) I received in response to a request for a test sample from a kind Sony employee.
Then again, problem 1-3 are still remaining to be solved and with all due respect I still think that DVD audio is way sexier from a technical point of view (not considering the "audiophile overkill" character of both, SACD and DVD audio)  -- So, problem 1-3 are probably remaining for a looong time.


Sebi
  • Last Edit: 02 October, 2005, 09:51:54 AM by SebastianG

  • tgoose
  • [*][*][*][*]
SACD Ripping
Reply #7
Can DSD to PCM be done losslessly?

  • rjamorim
  • [*][*][*][*][*]
SACD Ripping
Reply #8
Quote
This would be the smallest problem of all since DSD->PCM conversion is pretty simple.[a href="index.php?act=findpost&pid=331047"][{POST_SNAPBACK}][/a]


And how can that be achieved? Are there readily available tools for that conversion?
Get up-to-date binaries of Lame, AAC, Vorbis and much more at RareWares:
http://www.rarewares.org

  • Garf
  • [*][*][*][*][*]
  • Developer (Donating)
SACD Ripping
Reply #9
Quote
Quote
This would be the smallest problem of all since DSD->PCM conversion is pretty simple.[a href="index.php?act=findpost&pid=331047"][{POST_SNAPBACK}][/a]


And how can that be achieved? Are there readily available tools for that conversion?
[a href="index.php?act=findpost&pid=331060"][{POST_SNAPBACK}][/a]


Don't you just have to downsample? Granted, most out of the box tools wouldn't like 2.8Mhz 1 bit input data, but constructing one wouldn't be an issue.
  • Last Edit: 02 October, 2005, 02:25:16 PM by Garf

  • dekkersj
  • [*][*][*]
SACD Ripping
Reply #10
Quote
Quote
Quote
This would be the smallest problem of all since DSD->PCM conversion is pretty simple.[a href="index.php?act=findpost&pid=331047"][{POST_SNAPBACK}][/a]


And how can that be achieved? Are there readily available tools for that conversion?
[a href="index.php?act=findpost&pid=331060"][{POST_SNAPBACK}][/a]


Don't you just have to downsample? Granted, most out of the box tools wouldn't like 2.8Mhz 1 bit input data, but constructing one wouldn't be an issue.
[a href="index.php?act=findpost&pid=331091"][{POST_SNAPBACK}][/a]


Yes, you have to downsample. If you have a hardware player, just pick up the DSD stream and acquire it. To do so is of course a difficult task, but not undoable I would say. As far as I know this is the only way of getting the sacd on the harddisk. But maybe in some time we are in luck since there has been a move towards commercializing sacd. If I am not mistaking, Sonic has gained ownership of some rights.  In other words, in the near future there will be software available for the professional market to produce sacd's much easier than nowadays. Once this is done, I expect that it would be much easier to rip sacd's at home.

Greetz,
Jacco
Logical reasoning brings you from a to b, imagination brings you everywhere.

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #11
So If you have to downsample when converting to pcm and convert the audio file to a lossy format, wouldn't the output mp3 technically sound the same as a 16bit cd rip/encode?(considering the sacd/dvd-audio and the cd version of an album was from the same master.)

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #12
What I was getting at when posting this thread was not just whether I could rip the audio files out of sacds/dvd audio discs but whether the sound of an mp3 ripped from these sources would sound better than ones ripped from a 16 bit cd. Another thing that I was wondering is if there was a 24 bit audio lossy format that can go direct dvd audio to lossy without downsampling to 16bit

  • Defsac
  • [*][*][*][*]
SACD Ripping
Reply #13
Quote
What I was getting at when posting this thread was not just whether I could rip the audio files out of sacds/dvd audio discs but whether the sound of an mp3 ripped from these sources would sound better than ones ripped from a 16 bit cd.
There's still a lot of debate as to whether the uncompressed DVD-A/SACD tracks sound better than uncompressed CD audio.
  • Last Edit: 02 October, 2005, 10:48:22 PM by Defsac

  • dekkersj
  • [*][*][*]
SACD Ripping
Reply #14
Quote
Quote
What I was getting at when posting this thread was not just whether I could rip the audio files out of sacds/dvd audio discs but whether the sound of an mp3 ripped from these sources would sound better than ones ripped from a 16 bit cd.
There's still a lot of debate as to whether the uncompressed DVD-A/SACD tracks sound better than uncompressed CD audio.
[a href="index.php?act=findpost&pid=331175"][{POST_SNAPBACK}][/a]


Right, the only reason (IMHO) that DVD-A/sacd tracks sound better is that these formats are better taken care of from recording to disk.

Regards,
Jacco
Logical reasoning brings you from a to b, imagination brings you everywhere.

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #15
Quote
Quote
Quote
What I was getting at when posting this thread was not just whether I could rip the audio files out of sacds/dvd audio discs but whether the sound of an mp3 ripped from these sources would sound better than ones ripped from a 16 bit cd.
There's still a lot of debate as to whether the uncompressed DVD-A/SACD tracks sound better than uncompressed CD audio.
[a href="index.php?act=findpost&pid=331175"][{POST_SNAPBACK}][/a]


Right, the only reason (IMHO) that DVD-A/sacd tracks sound better is that these formats are better taken care of from recording to disk.

Regards,
Jacco
[a href="index.php?act=findpost&pid=331218"][{POST_SNAPBACK}][/a]

Aren't golden eared audiophiles going to say "BLASPHEMY" to this statement? Are you saying that no human can do a good abx with 16bit cd against DVD audio or sacd if the mastering/ recording process is the same for both? I cannot try it for myself either cause I don't have a dvd-audio or sacd player or any of the media for that matter

  • dekkersj
  • [*][*][*]
SACD Ripping
Reply #16
Quote
Quote
Quote
Quote
What I was getting at when posting this thread was not just whether I could rip the audio files out of sacds/dvd audio discs but whether the sound of an mp3 ripped from these sources would sound better than ones ripped from a 16 bit cd.
There's still a lot of debate as to whether the uncompressed DVD-A/SACD tracks sound better than uncompressed CD audio.
[a href="index.php?act=findpost&pid=331175"][{POST_SNAPBACK}][/a]


Right, the only reason (IMHO) that DVD-A/sacd tracks sound better is that these formats are better taken care of from recording to disk.

Regards,
Jacco
[a href="index.php?act=findpost&pid=331218"][{POST_SNAPBACK}][/a]

Aren't golden eared audiophiles going to say "BLASPHEMY" to this statement? Are you saying that no human can do a good abx with 16bit cd against DVD audio or sacd if the mastering/ recording process is the same for both? I cannot try it for myself either cause I don't have a dvd-audio or sacd player or any of the media for that matter
[a href="index.php?act=findpost&pid=331224"][{POST_SNAPBACK}][/a]


I did (and do) not mean to upset anybody. IMHO there are two improvements in dvd-a/sacd over ordinary cd: more bandwidth and more dynamic range. I can not hear above 20 kHz (I tested this) so only one argument survives, ie. a larger signal-to-noise ratio. Since the DR of an orchestra is let us say 80 dB, there is no need to go beyond cd. However, the multichannel feature justifies the existance of dvd-a/sacd. Stereo is for me history.

Regards,
Jacco
Logical reasoning brings you from a to b, imagination brings you everywhere.

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #17
What is DVD audio by the way 24 bit 96khz? that in itself should make it sound more analog since there is more data bits per second. Question is who hears this difference around here? please reply

  • dekkersj
  • [*][*][*]
SACD Ripping
Reply #18
DVD-A is at best 192 kHz 24 bits and in multichannel 96 kHz 24 bits. This format is well suited for dvd-a production since most recording studios work with the 96 kHz resolution. For cd, a sample rate conversion should be performed and one can debate about the quality of this process. Adobe Audition has a "quality level" as a slider if you want to perform this conversion in software, suggesting that they cannot do it flawlessly. For me this is a strong indicator one should not alter sample rates, or one should be extremely carefull.

Your statement about "sounding analog" is marketing language to me. If one speaks to an audiophile, analog sound is mostly a matter of coloration, second order distortion and so on. You probably mean that the more sample moments (and levels) there are, the better the analog shape is approximated. Both are not scientific, but mood-based ways to describe what is really going on. By the way, if you want an enormous amount of bits per second go for sacd...You see, sounds like marketing without any technical argument.

So, on system level there is no reason to go to any high-resolution format but implementation issues make it that more bits, oversampling etc do sound better. Sometimes, the reason is trivial: better electronics.

Regards,
Jacco
Logical reasoning brings you from a to b, imagination brings you everywhere.

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #19
What I mean by the word "analog" is that is sounds more like the original master or real life. More bits so that less quantization is needed. In theory, doesn't analog sort of mean unlimited bits and bandwidth since the sound of somebody speaking to you for example doesnt have x samples per second. It is unlimited. All thats done with digital audio is making enough samples per second to make the digital audio sound as if it is the "analog" or real sound that was originally recorded.
The argument is whether the difference of data between 2 channel 16 bit/44khz CD audio and 24/192 really have an audible difference

  • dekkersj
  • [*][*][*]
SACD Ripping
Reply #20
Quote
More bits so that less quantization is needed.
The argument is whether the difference of data between 2 channel 16 bit/44khz CD audio and 24/192 really have an audible difference
[a href="index.php?act=findpost&pid=331244"][{POST_SNAPBACK}][/a]


Here, you make a huge mistake. A properly dithered digital signal will be a perfect analog signal after (ideal) interpolation. There is a lot of theory about this subject. To my knowledge this theory is still valid.

Regards,
Jacco
Logical reasoning brings you from a to b, imagination brings you everywhere.

  • beto
  • [*][*][*][*][*]
  • Members (Donating)
SACD Ripping
Reply #21
Quote
What I mean by the word "analog" is that is sounds more like the original master or real life. More bits so that less quantization is needed. In theory, doesn't analog sort of mean unlimited bits and bandwidth since the sound of somebody speaking to you for example doesnt have x samples per second. It is unlimited. All thats done with digital audio is making enough samples per second to make the digital audio sound as if it is the "analog" or real sound that was originally recorded.
The argument is whether the difference of data between 2 channel 16 bit/44khz CD audio and 24/192 really have an audible difference
[a href="index.php?act=findpost&pid=331244"][{POST_SNAPBACK}][/a]


Yes, analog may be thought as unlimited but your argument is a moot point because we can only hear frequencies up to 22khz (and sometimes even lower than that).
In the past Nyquist already proved that to fully represent the spectrum of a bandwidth limited signal you have to sample at a minimum of twice the highest frequency. Which is precisely the samplerate of a CD.
Going above that smells like snake oil, because for sure it has no scientific base...
  • Last Edit: 03 October, 2005, 08:11:47 AM by beto

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #22
Quote
Quote
What I mean by the word "analog" is that is sounds more like the original master or real life. More bits so that less quantization is needed. In theory, doesn't analog sort of mean unlimited bits and bandwidth since the sound of somebody speaking to you for example doesnt have x samples per second. It is unlimited. All thats done with digital audio is making enough samples per second to make the digital audio sound as if it is the "analog" or real sound that was originally recorded.
The argument is whether the difference of data between 2 channel 16 bit/44khz CD audio and 24/192 really have an audible difference
[a href="index.php?act=findpost&pid=331244"][{POST_SNAPBACK}][/a]


Yes, analog may be thought as unlimited but your argument is a moot point because we can only hear frequencies up to 22khz (and sometimes even lower than that).
In the past Nyquist already proved that to fully represent the spectrum of a bandwidth limited signal you have to sample at a minimum of twice the highest frequency. Which is precisely the samplerate of a CD.
Going above that smells like snake oil, because for sure it has no scientific base...
[a href="index.php?act=findpost&pid=331247"][{POST_SNAPBACK}][/a]


Its not the frequency response I worry about, its the number of data bits making the sound "less jagged" like a higher polygon count 3D model for example that makes it smoother.

  • Donunus
  • [*][*][*]
SACD Ripping
Reply #23
Quote
Quote
More bits so that less quantization is needed.
The argument is whether the difference of data between 2 channel 16 bit/44khz CD audio and 24/192 really have an audible difference
[a href="index.php?act=findpost&pid=331244"][{POST_SNAPBACK}][/a]


Here, you make a huge mistake. A properly dithered digital signal will be a perfect analog signal after (ideal) interpolation. There is a lot of theory about this subject. To my knowledge this theory is still valid.

Regards,
Jacco
[a href="index.php?act=findpost&pid=331245"][{POST_SNAPBACK}][/a]


Well what I was saying is that you don't need to interpolate(basically guess) as much when there is more data per second so there are less chances of interpolation errors. Still, let's ditch theory. Is there nobody that can say that DVD audio or SACD sounds better than conventional CD? Can anyone here hear the difference? If Not, then whats the point of SACD then? Is it just something that can cure some audiophiles paranoia that CD sound is not good enough?

  • dekkersj
  • [*][*][*]
SACD Ripping
Reply #24
Interpolation is something that simply belongs to a digital system. However, I agree with you that the more stupid-proof a system is (more bits to avoid difficult to understand dither principles) the best change you have to an improved sound quality.

Regards,
Jacco
Logical reasoning brings you from a to b, imagination brings you everywhere.