Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[silversurfer0116]

Hi:  I'm new to JRiver and computer audio in general.  Like most people I started ripping CD's with Windows Media Player and then iTunes.  I've found the quality of playback from my laptop using iTunes for Windows  through my DAC is less than I had hoped.  So here I am at J River.  It seems to me that it would be best if one could rip a CD in lossless format directly off of CD transport to a hard drive, bypassing as much of the computer's noise, sound card, etc. as possible.  I've searched for this info without much success.  Maybe it doesn't matter?? as long as the software you use does a good job ripping a CD from the computer's optical drive but so far I am less than thrilled with the results (MP3's but still less than thrilled).  I will be going lossless for computer playback on my home audio system from now on.

*  If someone could direct me to a good tutorial (website, book, whatever) on ripping CD's this way, and vinyl too, I would appreciate it.  I'm trying to get a good start on this without too many hiccups.

fyi...I have an older but excellent CD transport made by Esoteric; a Cary Audio Design DAC, the Xciter, also a little old (is 5 years old?),  and a Toshiba P770 laptop, and a Toshiba auxillary hard drive.  These are quality components except for the Toshiba which was supposed to be great but I am less than thrilled with it's build quality.

[blgentry]

Like most people I started ripping CD's with Windows Media Player and then iTunes.  I've found the quality of playback from my laptop using iTunes for Windows  through my DAC is less than I had hoped.  So here I am at J River.  It seems to me that it would be best if one could rip a CD in lossless format directly off of CD transport to a hard drive, bypassing as much of the computer's noise, sound card, etc. as possible.  I've searched for this info without much success.  Maybe it doesn't matter?? as long as the software you use does a good job ripping a CD from the computer's optical drive but so far I am less than thrilled with the results (MP3's but still less than thrilled).  I will be going lossless for computer playback on my home audio system from now on.

To be sure of the issue of you "not being thrilled" with the playback of files on your system, you need to start with some known good high quality lossless files.  I'm guessing that once you have good quality lossless files, you're going to be much more thrilled.    So download some demo tracks or something to make sure the rest of your system is set up in way that pleases you.  That way you can be sure it's your old MP3s from itunes that are the source of your poor quality sound.

I've been on a very long journey with digital audio.  I initially thought MP3s were absolutely "good enough" and ripped a bunch of my CD collection as MP3.  When I hooked up my gen4 ipod to my (then) new home stereo I was taken aback at the lack of life in the sound.  The highs seemed to be less prominent.  The overall "feel" of the sound was dull and not involving.  I was honestly shocked.  My first thought was that I had something set up wrong.  I went and got a physical CD and put in in my DVD player and listened to that for comparison.  There was a very easily discernible difference.  The highs and the "life" were back in the music again with the CD.

Many years later, I decided to re-rip my collection as lossless FLAC files.  I'm listening to some as I type.  I can tell you honestly, that with good playback equipment, the sound is better than I've gotten from CD players in my previous systems.  Granted, I'm mostly listening with headphones these days, which adds a sense of detail that's hard to get from speakers.  But the real point is that the source of music, the files, are extremely high quality with none of the problems I previously heard on my lossy files.

It's actually quite jarring to switch from playing my FLAC files to playing a youtube video of a song (for example when a friend sends me a link to a cool song).  The change in quality is HUGE.  The FLACs really are that much better to someone who knows and appreciates sound.

Now that I've told you a whole story you didn't ask for, let me answer your question. 

As far as I know, there's no easy way to use a home audio optical transport to rip CDs.  The good news is, there's also no need to do so.  If you use a ripping program that has what's called a "secure mode", you're almost guaranteed a perfect quality digital transfer (rip).  I've ripped hundreds of CDs with my Mac's internal optical drive and all of the discs that are undamaged have yielded perfect rips.  A lot of them *with* damage also yielded perfect rips.

MC20 for Windows has a secure ripping mode I believe, but I haven't used it.  (I'm on a Mac).  I *personally* prefer to rip with a program that uses the AccurateRip database.  This is a way of doing a checksum on each file that's ripped and then comparing it to a community database of rips from other people.  If the checksums match the database, you can be pretty sure you got a perfect rip.  If they don't, you probably didn't get a perfect copy due to dirt or damage on the disc.

There are several well known programs for Windows that use the Accurate Rip database.  JRiver sometimes doesn't like us to mention other product names here, so I'll leave it to you to search them out if you want one.

In summary:  Use your computer to rip, using a program with Secure Mode and the Accurate Rip database and you'll get essentially perfect lossless files almost every time.  I recommend FLAC.

Good luck and good listening.

Brian.

[glynor]

It seems to me that it would be best if one could rip a CD in lossless format directly off of CD transport to a hard drive, bypassing as much of the computer's noise, sound card, etc. as possible.

That is exactly what happens when any modern ripping application rips from CD. The sound card, and noise in the computer, makes no difference as there is no analog conversion step.

MC does ripping in Secure Mode, which verifies that the exact data that was on the disc matches the file created on the hard drive. So long as you don't turn that feature off (it is on by default), and you rip to a lossless file type, then there is no difference between the ripped file and the original media on the source disc. It is a perfect, digital copy.

I *personally* prefer to rip with a program that uses the AccurateRip database.  This is a way of doing a checksum on each file that's ripped and then comparing it to a community database of rips from other people.  If the checksums match the database, you can be pretty sure you got a perfect rip.  If they don't, you probably didn't get a perfect copy due to dirt or damage on the disc.

There are several well known programs for Windows that use the Accurate Rip database.  JRiver sometimes doesn't like us to mention other product names here, so I'll leave it to you to search them out if you want one.

AccurateRip only ensures that your copy of the CD matches other people's copy of the CD. Secure ripping alone verifies that the rip matches the disc inserted.  So, the database is of ambiguous utility, as it cannot help (more than a secure rip) verify anything about the files output by the ripping application, only that the sources match.

I don't think JRiver cares if you mention other applications, like dbpoweramp, that do other things. They do get upset when someone implies that MC's ripping is less capable because it doesn't support AccurateRip. But that's because AccurateRip doesn't have anything to do with ripping accuracy. If you've properly implemented secure mode ripping (and they have) then AccurateRip is redundant, or at best, verifies that the sources match what other people have out there.

Community databases can be nice, but they're also troublesome, because they can easily get filled with inaccurate information. In this particular case, because it doesn't offer anything useful, and could lead to unsolvable support inquiries, they have no interest in adding it*.

* And I don't know if they care about this, but I suspect that AccurateRip has some privacy implications as well. I think the real "goal" of that feature is to collect a dossier of information on your music which is valuable to advertisers.

[blgentry]

AccurateRip only ensures that your copy of the CD matches other people's copy of the CD. Secure ripping alone verifies that the rip matches the disc inserted.  So, the database is of ambiguous utility, as it cannot help (more than a secure rip) verify anything about the files output by the ripping application, only that the sources match.

I don't get your reasoning.  "Secure" is a word selected to describe a process that attempts to verify that it read the disc correctly.  But it can't actually know that it got the correct bits.  Accurate Rip, on the other hand, compares a checksum of the whole file to another checksum submitted by someone else with the same disc.  If those checksums match, it becomes extremely likely that you both got the correct bits off of the disc.  How can you say this is not useful?  It is an end to end verification of the contents of the digital file you just created by ripping.  Accurate Rip measures the only thing that matters:  The end result.

Brian.

[JimH]

I don't get your reasoning.  "Secure" is a word selected to describe a process that attempts to verify that it read the disc correctly.  But it can't actually know that it got the correct bits.  Accurate Rip, on the other hand, compares a checksum of the whole file to another checksum submitted by someone else with the same disc.  If those checksums match, it becomes extremely likely that you both got the correct bits off of the disc.  How can you say this is not useful?  It is an end to end verification of the contents of the digital file you just created by ripping.  Accurate Rip measures the only thing that matters:  The end result.

This is just not a problem any more.  Please read some of the old threads on this.

[glynor]

I don't get your reasoning.  "Secure" is a word selected to describe a process that attempts to verify that it read the disc correctly.  But it can't actually know that it got the correct bits.

That is not correct. It can, because discs include EDC error detection (checksums). This allows the application and the disc controller to verify that it was able to read the same data that was originally written to the disc, or to know when it failed to do so.

If a disc is damaged, secure ripping can use the ECC data on the disc to reconstruct the original data, or re-read sectors of the disc multiple times until it gets a "good transfer" (which passes the cryptographic checksum).

But a 100% secure rip is verified that it passes all of the EDC checks, and is a cryptographically sound "perfect copy" of the original data on the disc.  That is why AccurateRip is redundant.

[glynor]

I thought of a better way to explain, perhaps.

You seem to think that a secure rip reads the disc multiple times in order to verify that it read the disc correctly (using the multiple reads as a verification step).  Sort of doing a "I read this sector 4 times, and 3 of the times it came out one way, so that's probably the right answer".

This is not what happens at all.

Each sector is "signed" with a 392 byte Error Detection layer, which is essentially a cryptographic hash function (a reversible one, which allows it to be used as ECC as well).  Like all hash functions, the output is deterministic based on the input. If the input changes, even by one bit, then the output signature changes. This is the same technology that allows secure websites to work, and SHA checksums to work. The hash function on a compact disc isn't as "secure" as a cryptographic hash used in security applications, because it doesn't have to be (there's no active attacker trying to break in, just random dumb bad luck and quantum effects)*.

Without error detection when reading from disks of all types, computers would not work at all.

In a secure ripping mode, the application absolutely, positively knows when it gets a "bad read" and then it does either:
* Reconstructing the missing data, perfectly, from the ECC XOR data available on the disc.
* Re-reading the sector multiple times until it gets the "correct" (matching the hash function) answer.

Then, at the end once it gets a confirmed-successful read, it also verifies that what it read is actually what was written to the new (ripped) file.

With a damaged disc, neither may be possible, or the EDC data itself may be damaged. If this happens, then that is what reduces the quality of the rip from 100%.

The multiple reads are error correction, not error detection. It already knows it got a bad read, because the hash doesn't match, before it even tries to re-read the sector.

* It is possible there could be random collisions in the hash function, but this is incredibly unlikely (like it would take longer than the age of the universe to create collisions by random chance unlikely). And, by the way, unless you are using a Xeon-class CPU with ECC RAM support, it is millions of times more likely that the data will be corrupted upon playback in RAM (which does have no error detection or correction, then) than it is that it will be incorrectly written to disk in a rip.