Jump to content
  • The Computer Audiophile
    The Computer Audiophile

    Computer Audiophile CD Ripping Strategy and Methodology

    At first blush ripping a Compact Disc with a computer seems like such a simple task. So simple that it can nonchalantly be done by someone from eight to eighty years old. Putting a CD into a Macintosh computer automatically launches iTunes. By default iTunes can automatically start ripping the CD to the hard drive and eject the disc when it's finished. Listeners seeking the ultimate in convenience and immediate results should select this method of ripping. Note the two words "in" and "convenience" next to each other in the previous sentence. That choice of words was not by accident. Using the aforementioned ripping method will lead to major inconvenience, lost music, and hours of precious time wasted. Ripping CDs, like everything in life, involves many choices and well thought out decisions if one wants to do it right the first time. There are simple ways, there are hard ways, and there are many ways in between the two extremes to rip a CD collection. What follows is the Computer Audiophile CD Ripping Strategy and Methodology. This manual details the Computer Audiophile strategy and method of ripping a CD collection and explains why this method is recommended.

     

     

     

     

     

     

    Strategy & Methodology

     

     

    The importance of a well thought out ripping strategy and defined methodology cannot be stressed enough. Without a solid strategy and solid methodology one is bound to lose countless hours of time and money by re-ripping CDs. In addition the inability to use an old CD ripped in an obscure format and the inability to find one's music stored in a mass of untagged files with no identifying data is a guarantee for those who dive in head first without forethought. Fortunately most of these pitfalls can be avoided by following the recommendations in this document. Note: This manual covers software strategy and methodology. Hardware considerations will be addressed in another document.

     

     

     

     

    Computer Audiophile CD Ripping Strategy

     

    Strategy |?strat?j?|

    Noun ( pl. -gies)

    A plan of action or policy designed to achieve a major or overall aim.

     

    The mindset needed for someone about to rip an entire Compact Disc collection must be that of a preservationist seeking to preserve a piece of art, music on a disc, by creating the most accurate copy possible. Before embarking down the ripping road one must develop a clear strategy. Developing a clear strategy requires a goal. The goal here begins with a definition of digital preservation. Digital preservation is defined as: long-term, error-free storage of digital information, with means for retrieval and interpretation, for the entire time span the information is required [1]. Applying this definition specifically to ripping Compact Discs brings the goal a bit more into focus.

     

    A. Long-term can be indefinite according to The Reference Model for an Open Archival Information System (OAIS) Principles of digital preservation.

    B. Error-free storage of digital information in this context is equivalent to error free digital copies of Compact Discs on a storage medium not prone to errors itself. Terms associated with error free ripping are secure rips, accurate rips, and error correction among many others.

    C. The means of retrieval, in the context of ripped CDs, literally means actual access to the ripped files on a hard drive or other storage medium readable by computers available when the data is needed. Expanding beyond the literal sense, the word retrieval necessitates the ability to retrieve audio and playback it back from the file itself. Think of file formats that can be used today and the chances of support for these formats in the future.

    D. The term interpretation relates partially to the previous concept of retrieval because one's application must be able to interpret the file format. Interpretation also relates to metadata and how one interprets specific files or albums among a sea of similar looking data. Interpretation is synonymous with explanation and here the metadata explains what one is looking at by clearly stating the track name, album name, and artist name in addition to many other tags.

    E. The Time span this information is required is up to each person. The times span may be until a specific file can no longer be played by any available hardware or software or until a specific file format can be converted into another format that meets the aforementioned goals. The time span that one wants to keep around that old Mili Vanili <a href="http://en.wikipedia.org/wiki/Girl_You_Know_It%27s_True">Girl You Know It's True</a> album may have nothing to do with technology and more to do with one's pride.

     

    Closely tied to digital preservation is digital sustainability. This is a continual process focused on building an infrastructure and approach that is flexible with an emphasis on interoperability. Digital sustainability incorporates activities in the present that will facilitate access and availability in the future.[2] In the context of this document flexibility, interoperability, and access and availability in the future are key concepts. The end result of this strategy and methodology must be a music collection that embodies all of these terms.

     

    In addition to the somewhat scholarly definitions listed above one must keep in mind that ripping CDs is done to enjoy one's music collection via a music server. Album art, immediate access to long lost tracks, great sound, and playlists are paramount. Thus, the stated goal must incorporate as much of these needs as those of a digital preservationist.

     

     

     

    Stated Goal

     

    The concise goal is to rip one's CDs accurately, to a reliable storage medium, in a file format that's as future resistant as possible, with metadata, for use as long as one requires. In layman's terms, "Do it right the first time." Doing things right the first time often involves more work up front but increases usability and saves time and money down the road. The Computer Audiophile ripping strategy is no different.

     

     

     

    Accomplishing The Goal

     

    The Computer Audiophile strategy to accomplish this goal requires more than a single copy of one's music collection. I've adapted a concept from Peter Copeland's, dated but still relevant, Manual of Analogue Sound Restoration Techniques to more appropriately fit an audiophile's needs. The number of copies of one's music collection should be no less than two and may reach four depending on one's preferences.

     

    (1) The original Compact Disc should be kept as long as possible.

    (2) A ripped bit for bit "Archival Copy” of the CD only used to produce other copies, not used for playback.

    (3) A ripped bit for bit "Working Copy" of the CD used for everyday playback.

    (4) An optional "Edited Copy" that may contain volume leveling, EQ settings, or additional alterations to the data.

     

    The original Compact Disc should not be discarded as it is the only real reference version of the specific album. As good as current ripping techniques are they may not be able to completely clone a Compact Disc according to the late Peter Copeland, former Conservation Manager at the British Library Sound Archive. Readers thinking about sending their music collection off to a service that rips the CDs in exchange for keeping the physical disc are strongly discouraged from this accident waiting to happen.

     

    Ripping the CD as accurately as possible is a critical piece of this ripping strategy. Accurate equates to ripping the Compact Disc in its native resolution 16-bit / 44.1 kHz. Frequently readers of Computer Audiophile have asked about ripping CDs to a higher sample rate such as 24-bit / 48 kHz. Ripping to anything other than 16/44.1 does not preserve an accurate "Archival Copy" of the CD. An inaccurately ripped copy leaves no future options to produce a bit for bit version of the original CD. If one wants a 24/48k version of a CD, it can be created any time from the ripped 16/44.1 version. While it is technically very possible to downsample from a higher resolution than 16/44.1 there are unavoidable mathematical algorithms for rounding numbers during the conversion and each downsampling program operates differently. Thus a downsampled version will likely never equal an exact "Archival Copy" ripped at 16/44.1. In addition to ripping the CD in its native resolution one must use a ripping program that offers a secure ripping feature. This feature compares the checksum of the ripped file to that of others in a global online database. This ensures the rip is at least consistent with a number of other rips around the world. CDs that are not in such an online database are often ripped several times and the rips are compared to each other before the program will consider the rip secure and encode the track. More information about how to accomplish this is below in the methodology section.

     

     

     

    Archival Copy, Working Copy, Edited Copy

     

    Listening to music daily via a music server allows a lot of time to peruse one's music files via a playback application like iTunes or MediaMonkey. This also allows the listener plenty of opportunities to accidentally delete the files or do something, that can't be reproduced, that causes the files to disappear. In the second scenario the files are usually still there, but if the user can't find them the files are essentially gone. In addition to human error computer hardware is far from perfect. Hard drives fail whether they are traditional spinning drives or solid state drives. For these and many other reasons, it is recommended to keep an Archival Copy of one's music files and a Working Copy. The Archival Copy never changes. It is an "exact" replica of the original music on the disc. There is no chance of corrupting the files via usage or accidentally deleting them while browsing or cleaning up your collection. This Archival Copy should reside on a separate storage device from the working Copy. For example, storing the Archival Copy on a NAS drive and the Working Copy on a local drive is a solid solution. One should also think about off-site storage and stable storage media like tape, but that's a discussion for another day. In addition to the ripped files in the Archival Copy location it is recommended one keep downloadable PDF liner notes and the automatically generated ripping report with information about the success or failure of each ripped CD in this location. More information about generating an automatic ripping report is in the methodology section below. The Working Copy is just that, the bit for bit copy of music you work with and playback through your audio system. Changes can be made to this copy without worry. The Working Copy can be replaced easily from the Archival Copy. This may seem a little over the top at first, but in the long run it will save countless hours of time by avoiding the need to re-rip one's music collection. It will also save money for those who hire the ripping process out to a third party. There are a few ways to accomplish this double or triple rip that make it as easy as iTunes once setup correctly. This will be covered in the methodology section as well. Note: The Archival Copy will also be invaluable when the need arises to use a different file format. A detailed explanation of the reasoning behind this can be found below. An Edited Copy is totally optional and is likely not needed by many readers. In the days when EQ adjustments and volume normalization were unreversible this Edited Copy was more valuable. Nonetheless there may still be value in this fourth copy of one's music collection.

     

     

     

    Metadata |?met??dat?; -?d?t?|

    Noun

    A set of data that describes and gives information about other data.

     

    Metadata, commonly referred to as tags or the act of tagging, is essentially data about the data. For the purpose of the Computer Audiophile Ripping Strategy and Methodology metadata is all the information about an album, artist, song, etc... Without metadata songs are often titled Track 01, Track 02, and likely have no album art. Metadata is a huge enabler that nudges people down the music server path. Viewing all one's albums with artwork and track after track listings or playlists are all part of the experience. In addition to increasing usability now, metadata is critical for long term future viability of ripped material. Even if one follows this document closely up to this point and has an Archival Copy of all their music, it won't do much good if all the tracks are Track 01, all the albums are Untitled, and all the artists are Unknown. Another big part of using metadata is whether or not the metadata is embedded. Embedded means the music file itself stores all the metadata such as track names, artists names, album name, and album art. This is by far the preferred way to store metadata when ripping CDs. The other way metadata is stored is by associating this data with certain files, albums, and artists and keeping the information in a separate database or file. An example of this concept can be illustrated using iTunes. iTunes is capable of both embedding metadata and associating metadata with files. If one rips a CD in iTunes as AIFF files iTunes will embed the text metadata like track title, artist and album. If the iTunes Store sells the specific ripped album, iTunes will download the album art automatically but it will not be embedded. iTunes associates the album art with the specific album and places the picture in its unintelligible album art directory making it irretrievable in the future. iTunes is capable of embedding album art into the ripped AIFF files, but it must be done manually by adding the art via the track or album properties. Embedded v. associated metadata is not that big of a deal until one loses or wants to move or recreated a corrupt library. Embedded metadata moves with the files and appears in the new library without user intervention. Associated metadata is gone for good. In the example above the text metadata specifying album, artists, and track will remain with the file, but the associated album art will be lost because iTunes stored it in a folder instead of embedding the art. The importance of metadata ties nicely into the next topic, file formats.

     

     

     

    File Format - Archival Copy

     

    Recently I published an article recommending Computer Audiophile readers rip all their music as AIFF files. The ripping strategy outlined in this ripping manual is not a total departure from the previous recommendation rather it's an enhancement or expansion. The Computer Audiophile Ripping Manual recommends that all music ripped as the Archival Copy be ripped in FLAC format. The Working Copy file format depends on the operating system and playback application in use. Ripping the Archival Copy to bit for bit FLAC files enables one satisfy many of the principals of digital preservation and sustainability such as flexibility, interoperability, and access and availability in the future . Since FLAC (Free Lossless Audio Codec) is lossless it enables a 16/44.1 1:1 copy of the music. This satisfies the need to preserve the original work of art (music) in it's native format. Sure FLAC is not the format of the music on a Compact Disc, but it allows an identical bit for bit copy to be extracted from the FLAC file. The FLAC format is highly usable as it is and can be converted into virtually any format available as of this writing. FLAC is not future proof but it could be called future resistant. FLAC will likely be supported as long or longer than any of the current uncompressed or compressed lossless formats. The chances are also pretty high that when support for FLAC eventually fades there will be a readily available program to convert FLAC files into the successor format. The same cannot be said about proprietary formats. When decisions are made to benefit shareholders, think Apple and Microsoft, there is no telling what the future holds. That's neither good nor bad, it's just a fact of life. FLAC also supports metadata as good or better than any other format. All types of metadata can be embedded into FLAC files and this metadata is readable by almost all applications that support FLAC. (Songbird currently has problems reading embedded album art). Readability of FLAC files by applications is a trivial matter. FLAC has a standard that many formats like WAV and AIFF do not. As an example, when researching file formats I talked to application developers who all said FLAC is simple to support in playback applications. These developers also said WAV and AIFF each have many inconsistencies, such as the way a file header is written, that make it more difficult support. Two real world examples come to mind when I think of WAV files. Over the last year PS Audio and Boulder have finally succeeded in playing the Reference Recordings HRx 24/176.4 files back via their disc players. Many of the difficulties these companies had in development were related to the inconsistent way file headers were written in the WAV files. Again, FLAC files don't have this problem.

     

     

     

    File Format - Working Copy

     

    The file format of the Working Copy of one's music collection should address the specific needs of each user and does not need to be an open format as long as it's a 1:1 bit for bit copy. As a music lover and audiophile I recommend lossless and/or uncompressed formats only and will talk about these formats here. The file format selected is dependent on a combination of the operating system and playback application one will be using. Users of Mac OS X should rip the Working Copy of their music in AIFF format. AIFF supports embedded metadata usable by iTunes and is a 1:1 copy of the original work of art on the Compact Disc. AIFF is a fairly usable format for a few applications, although the metadata may not be widely usable. Many applications can still play AIFF if nothing else. I mention this because users frequently move files around to try playing music on mobile devices or a network streaming device at a friends house. AIFF should work OK in these situations. Additionally I don't believe there have been any changes to the AIFF format in years. The same cannot be said about CODECs like Apple Lossless Audio CODEC (ALAC). If one views the properties of ALAC files in iTunes, part of the metadata of the files lists what version of the CODEC was used to encode the file. With each new iTunes release there are likely changes made to this CODEC. Good, bad or harmless who knows but nonetheless there have been and will be more changes. I don't know what version of ALAC I would use as a reference to compare each successor or predecessor version. I have not personally tested this myself but some readers have reported hearing differences between files ripped with the new and old versions of the ALAC CODEC. Voodoo or not it's al least something to consider. Users of Microsoft Windows based computers have two reasonable choices for ripping their Working Copy. If the user needs the best embedded metadata I recommend ripping the Working Copy to FLAC. This is against my strong preference for uncompressed files during playback, but it's the only way to go if embedded metadata is critical. If the user can live with the high likelihood of losing the metadata for the Working Copy and wants to guarantee the best sound quality then the recommended format is WAV. This does change my previous recommendation of using AIFF on a PC. Because WAV support is so widespread on PCs and AIFF metadata is not supported very well on PCs it is recommended to use WAV. The suggestion that the best sound quality for PCs comes from WAV files is not uncontroversial and is a conclusion I encourage everyone to come to on their own. It is my opinion that playing WAV files will not hurt the sound quality. The same file portability and interoperability as AIFF files is also pertinent to WAV files. The one caveat with a Working Copy full of WAV files is that the very limited support of embedded metadata will eventually cause one to lose this metadata. For example, using MediaMonkey with WAV files one can associate metadata and album art with the files. However, if one wants to switch applications or if the MediaMonkey database is lost or becomes corrupt much of the metadata is gone. It is often said that WAV files do not support embedded metadata. This is not a 100% correct statement as I believe there is room for embedded metadata within the WAV format. I've had inconsistent results with embedded metadata and WAV files. Album artwork is one that I've had the most problems with. Embedded v. associated metadata is a major concern that shouldn't be taken lightly.

     

     

     

    File Format - Edited Copy

     

    This is completely open to any uncompressed, lossless, or lossy format one selects.

     

     

     

    The Computer Audiophile ripping strategy in its simplest form comes down to four points (listed below). This strategy is the first step toward accomplishing the goal of ripping one's CDs accurately, to a reliable storage medium, in a file format that's as future resistant as possible, with metadata, for use as long as one requires. With this strategy in hand one must execute a solid methodology to accomplish the goal 100%.

     

     

    Summary of the Computer Audiophile CD Ripping Strategy

     

    1. Rip in 16/44.1 for a 1:1 copy of the original.

    2. Rip an Archival Copy and a Working Copy in addition to keeping the original Compact Disc.

    3. Embed metadata for usability now and data retrieval and interpretation in the future.

    4. Use open file formats for flexibility and interoperability in the future.

     

     

     

     

     

     

    Computer Audiophile CD Ripping Methodology

     

    methodology |?me???däl?j?|

    noun ( pl. -gies)

    a system of methods used in a particular area of study or activity.

     

    A simplified explanation of the Computer Audiophile Ripping Methodology: Use dBpoweramp in secure mode to rip CDs at 16/44.1, outputting two or three different file formats simultaneously to different disks.

     

    Executing the Computer Audiophile Ripping Strategy is much easier than one might think. The following methodology accomplishes the stated goal in one fell swoop. There are a few applications available that are just as capable as the one recommended in this document. Readers are encouraged to experiment on their own and select an application they are comfortable using. It is critically important that one sets one's self up for success when it comes to tasks like ripping a complete music collection. These boring, menial, and mindless tasks lead to cutting corners and taking the easy road unless one is setup for success. Ripping three copies of each disc will soon lead to one copy if the process is not streamlined and convenient. I recommend setting up a Windows computer that is always ready to rip CDs, like a ripping station with dBpoweramp ripping profiles ready to go. This way coming home from the store with five CDs doesn't lead to abandonment of this whole process because one wants to listen right away. No reconfiguring of the computer will be necessary and ripping the Computer Audiophile way will require the same amount of work as ripping the "traditional" way.

     

     

     

    Computer Audiophile CD Ripping Methodology Requirements

     

    1. Windows based computer (XP, Vista, 7)

    2. Adequate internal hard drive space

    3. Adequate external hard drive space on a NAS, USB, FireWire, or e-Sata disk.

    4. dBpoweramp Music Converter software, AIFF, WAV, and FLAC CODECS, Multi Encoder Utility

     

     

    Step By Step

     

    The following instructions show readers how I rip CDs to a Windows XP computer, a Mac Pro, and a NAS drive simultaneously. The Windows XP music server will contain a Working Copy as WAV files, the Mac G5 will contain a Working Copy as AIFF files, and the NAS will contain an Archival Copy as FLAC files. Note: My Windows XP computer does double-duty as my ripping PC and music server.

     

    Windows XP - WAV files for Working Copy

    Mac G5 OS X - AIFF files for Working Copy

    NAS - FLAC files for Archival Copy

     

     

     

    1. Setup for success with a ripping computer ready to go whenever it's needed.

     

    2. Setup folders on an external drive and/ or another music server to receive the ripped files.

     

    On the NAS drive create a folder called Archival Copy to receive FLAC files. On the Mac G5 running OS X 10.4 Tiger enable Windows Sharing (simply called File Sharing on OS X 10.6 Snow Leopard). The Windows PC already has a My Music folder accessible by dBpoweramp.

     

    Using OS Tiger X 10.4 enter System Preferences >> Sharing and enable Windows Sharing by checking the box to the left. Then click Accounts and enable at least one account.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/tiger-sharing-01-original.png" title="Tiger OS X 10.4 Sharing" class="thickbox" rel="tiger-sharing-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/tiger-sharing-01-33.png" alt="tiger 001" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/tiger-sharing-02-original.png" title="Tiger OS X 10.4 Sharing" class="thickbox" rel="tiger-sharing-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/tiger-sharing-02-33.png" alt="tiger 002" ></a>

     

     

     

    Using OS X Snow Leopard enter System Preferences >> Sharing and enable File Sharing by checking the box to the left. Then click the + symbol underneath the Shared Folders box. On the following screen select the Mac's Music folder and enable Share files and folders using SMB (Windows) as well as enabling at least one user account. Click Done and make sure the the Music folder is now listed in the Shared Folders box with Read & Write Access listed to the right of the newly enabled user account.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/snow-leopard-sharing-01-original.png" title="SL OS X 10.6 Sharing" class="thickbox" rel="sl-sharing-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/snow-leopard-sharing-01-33.png" alt="sl 001" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/snow-leopard-sharing-02-original.png" title="SL OS X 10.6 Sharing" class="thickbox" rel="sl-sharing-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/snow-leopard-sharing-02-33.png" alt="sl 002" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/snow-leopard-sharing-03-original.png" title="SL OS X 10.6 Sharing" class="thickbox" rel="sl-sharing-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/snow-leopard-sharing-03-33.png" alt="sly 003" ></a>

     

     

     

    3. Download and install dBpoweramp, AIFF CODEC, and Multi Encoder Utility (WAV & FLAC included by default). These are three separate files with simple installation procedures. Download and double-click the file.

    Program - http://www.dbpoweramp.com/download.htm

    AIFF - http://www.dbpoweramp.com/codec-central-aiff.htm

    Multi Encoder - http://www.dbpoweramp.com/codec-central-utility.htm

     

    4. Configure dBpoweramp general settings.

     

    Launch dBpoweramp CD Ripper. Make sure PerfectMeta )Intelligent Lookup) is enabled. It may not be obvious but one needs to select the little down arrow to the right side off the Meta button for this menu to appear.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-perfect-meta-original.png" title="dBp PerfectMeta" class="thickbox" rel="dBp-perfectmeta-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-perfect-meta-33.png" alt="dBp 001" ></a>

     

     

     

    Next click the little arrow next to the Options button go to Encoder Priority and make sure the highest priority available is selected. Then select After Ripping and make sure Display Error Log and Do Nothing are enabled. This allows an error log to pop-up after a rip goes awry. It's nice to know as soon as possible in addition to having the detailed log file in the Archival Copy album folder.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-encoder-priority-original.png" title="dBp Encoder Priority" class="thickbox" rel="dBp-encoder-priority-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-encoder-priority-33.png" alt="dBp 003" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-after-ripping-original.png" title="dBp Encoder Priority" class="thickbox" rel="dBp-encoder-priority-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-after-ripping-33.png" alt="dBp 004" ></a>

     

     

     

    To configure the dBpoweramp ripping options click the large Options button. The options I use are Secure (Recover Errors), AccurateRip, and Eject After Rip.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-01-original.png" title="dBp Secure" class="thickbox" rel="dBp-secure-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-01-33.png" alt="dBp secure 001" ></a>

     

     

     

    I don't Enable Ultra Secure Ripping because there comes a point when enough is enough. If I actually thought three to six rips were needed for my collection to be better then I would consider enabling this option. It's possible I may gather my tough to rip damaged CDs and rip a batch of them in Ultra Secure mode. Doing this will take a serious investment of time as one CD rip can last twenty-four hours if the disc is in bad enough shape. Up to this point I have been very satisfied with standard secure ripping and having my results verified by the checksum with the online database.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-02-original.png" title="dBp Ultra Secure" class="thickbox" rel="dBp-ultra-secure-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-02-33.png" alt="dBp ultra secure 001" ></a>

     

     

     

    The word Interpolation frequently has a bad connotation with audiophiles. Interpolation is a method by which the ripping program inserts an intermediate value or term into a series by estimating or calculating it from surrounding known values[4]. For example if one has a an unreadable spot on a CD named Frame B, the ripping program with interpolation enabled will use Frame A and Frame C to estimate what should have been read from Frame B and use that new value. This does present questions about preserving the actual CD content as perfect as possible. I look at it this way, the interpolated frame is more likely to sound like the actual recording from an undamaged disc than a pop or tick would sound without an interpolated frame. If the damaged disc in question is a one of a kind piece of history one obviously make two copies one with and one without interpolation. I don't recommend making the Archival Copy using the Computer Audiophile Methodology without interpolation. To the best of my knowledge there is no way to interpolate during a file conversion if one wants to create an AIFF copy from the Archival FLAC copy. Thus every subsequent copy would have a pop or tick in the un-interpolated frame position.

     

    In addition to interpolation, this is where one enables the Write To File option. When this is enabled a file will be written to the album folder for the currently ripped disc. This file is great to keep in the Archival Copy as it explains if anything hen wrong with the rip and more. I use this when I think I hear a click or pop in a song. Looking at this log provide a data point to help me decide is I am hearing something real or not or if it's just part of the recording.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-03-original.png" title="dBp Interpolation" class="thickbox" rel="dBp-interpolation-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-03-33.png" alt="dBp interpolation 001" ></a>

     

     

     

    <a href="http://files.computeraudiophile.com/2009/1029/Shelby-Lynne-Just-a-Little-Lovin.txt">Click here for log file example</a>

     

     

     

    Two nice features I enable are Eject After Rip and Prevent auto-run on all CD drives. Th first is handy when I am not paying attention. The opening disc tray usually grabs my attention better than anything else. Preventing auto-run of CDs is very nice so Windows or another program does not launch an unwanted interruption every time a new disc is inserted. Neither of these are absolutely necessary, but I think they are part of setting one's self up for success.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-04-original.png" title="dBp Eject" class="thickbox" rel="dBp-eject-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-opt-04-33.png" alt="dBp eject 001" ></a>

     

     

     

    5. Create a ripping profile for quick access to different ripping settings..

     

     

    In the lower left corner of the dBpoweramp window click the drop down arrow next to Profile and select [Create Profile]. In the pop-up box give this profile a name.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-01-original.png" title="dBp Create Profile" class="thickbox" rel="dBp-create-profile-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-01-33.png" alt="dBp create profile 001" ></a>

     

     

     

    Configure all the ripping settings for this profile. Remember the need for three file formats in three different locations for my two music servers and one Archival Copy on a NAS. This is where the Multi Encoder utility installed earlier comes into play. Select the drop down arrow to the right of the words Rip To, then select [Multi Encoder]

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-02-original.png" title="dBp Configure Profile" class="thickbox" rel="dBp-configure-profile-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-02-33.png" alt="dBp configure profile 001" ></a>

     

     

     

    In the middle box at the bottom of the dBpoweramp windows is where the file formats will displayed once configured. Click the Add Encoder button on the right side of this box. Select FLAC as the encoder as this will be the Archival Copy. I select Compression Level 0 (Fast) because I recommend using FLAC for all its other great features, not its compression algorithms. It's OK to compress these files if one chooses. Check the box to Verify Written Audio after encoding. The Output Location must point to an external drive, in this example the M: drive is my NAS and Archival Copy is the folder on the NAS that was crated back in step two. Select Dynamic to give the artists, albums, and tracks a directory structure that works for a give use. Clicking the Set button allows one to customize this directory structure. By default all the tracks are put into one folder with the artist's name and the tracks have the album and track name in their titles. I prefer to have a folder hierarchy with the artists at the top level, then a folder for each album, and the individual tracks within each album folder. The Dialog box pictured here has all the choices for choosing a directory structure. One can get creative and very granular.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-03-original.png" title="dBp Add FLAC" class="thickbox" rel="dBp-ad-flac-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-03-33.png" alt="dBp add flac 001" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-04-original.png" title="dBp Add FLAC" class="thickbox" rel="dBp-ad-flac-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-04-33.png" alt="dBp add flac 002" ></a>

     

     

     

    Add the AIFF encoder, leaving the bit depth, sample rate, and channels the Same As Source. The folder location selected here is the Automatically Add To iTunes folder on my Mac G5 music server. This folder was shared back in step two. After the files are ripped to this folder one has to open iTunes on the Mac G5 server and the tracks automatically get added to the iTunes library and removed from the Add To… folder. Don't rip to this location with iTunes open on the Mac G5 as I have run into problems of corrupted tracks and tracks not importing.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-05-original.png" title="dBp Add AIFF" class="thickbox" rel="dBp-ad-aiff-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-05-33.png" alt="dBp add aiff 001" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-original.png" title="dBp Add AIFF" class="thickbox" rel="dBp-ad-aiff-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-33.png" alt="dBp add aiff 002" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-filled-original.png" title="dBp Add AIFF" class="thickbox" rel="dBp-ad-aiff-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-filled-33.png" alt="dBp add aiff 003" ></a>

     

     

     

    Add the WAVE encoder, leaving the bit depth, sample rate, and channels the Same As Source. This folder location is the My Music folder on the local computer's hard drive. I use my ripping computer as my Windows XP music server as well. I use the same artist, album, track [title] directory structure here as I did in the previous folders.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-06-original.png" title="dBp Add WAVE" class="thickbox" rel="dBp-ad-wave-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-06-33.png" alt="dBp add wave 001" ></a>

     

     

     

    Once the profile is completely setup the Multi Encoder settings will look like the following graphic with FLC, AIFF, and WAVE in the box and all configured.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-07-original.png" title="dBp Multi Encoder" class="thickbox" rel="dBp-multi-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-profile-07-33.png" alt="dBp multi 001" ></a>

     

     

     

    6. Click the Rip button in the upper left corner and sit back while the Computer Audiophile CD Ripping Methodology does all the work from here on out. The first graphic shows dBpoweramp in the middle of the ripping process. The green letters and numbers below the column CRC indicate the track has been ripped and the checksum matches that of others in the database. The number in parenthesis in the Rip Status column next to AccurateRip indicates, "The rip had no detectable errors and agrees with [X number of] other people (they have ripped the same track and had the exact same result), an agreement from AccurateRip ensures an error free rip" according to the dBpoweamp website. The second graphic displays what dBpoweramp considers a perfect rip with ten Accurate tracks out of ten possible.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-02-original.png" title="dBp Ripping" class="thickbox" rel="dBp-ripping-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-02-33.png" alt="dBp ripping 001" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-01-original.png" title="dBp Ripping" class="thickbox" rel="dBp-ripping-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/dbp-rip-01-33.png" alt="dBp ripping 002" ></a>

     

     

     

    7. Check the rips to make sure everything worked as planned. Before opening iTunes on the G5 one can look in the Auto Add folder to check if the freshly ripped album is located in the right place. As soon as iTunes is opened these tracks will disappear into the iTunes hierarchy of folders leaving the Auto Add folder either empty or with a folder of unimproved tracks. The reasons for not importing a track can be many, some times the folder contains album artwork that doesn't get imported anyway. Thus no worries.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-filled-original.png" title="dBp Auto Add To iTunes" class="thickbox" rel="dBp-auto-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-filled-33.png" alt="dBp auto 001" ></a> <a href="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-original.png" title="dBp Auto Add To iTunes" class="thickbox" rel="dBp-auto-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/tiger-auto-add-folder-33.png" alt="dBp auto 002" ></a>

     

     

     

    Windows My Music Working Copy in WAV format.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/windows-my-music-original.png" title="dBp My Music" class="thickbox" rel="dBp-my-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/windows-my-music-33.png" alt="dBp my 001" ></a>

     

     

     

    NAS archival copy in FLAC format.

     

     

     

    <a href="http://images.computeraudiophile.com/graphics/2009/1029/NAS-archival-copy-original.png" title="dBp Archive" class="thickbox" rel="dBp-archive-gallery"><img src="http://images.computeraudiophile.com/graphics/2009/1029/NAS-archival-copy-33.png" alt="dBp archive 001" ></a>

     

     

     

    8. Leave the Archival Copy, import the Working Copies into the playback applications of choice. That's it, The Computer Audiophile CD Ripping Strategy and Methodology is now in place.

     

     

     

     

    Summary of the Computer Audiophile CD Ripping Methodology

     

    Once readers go through the initial setup and rip a few CDs they'll realize this methodology isn't difficult to use and understand. Again in simple terms, use a Windows PC with dBoweramp to securely rip CDs at 16/44.1 into three formats simultaneously on to three different disks. When setup for success this methodology is just as easy as ripping with iTunes and offers a tremendous amount of advantages over iTunes.

     

     

     

     

     

    Ripping Wrapping Up

     

    The Computer Audiophile CD Ripping Strategy and Methodology, if followed, allows one to reach the goal stated at the beginning of this document. " To rip one's CDs accurately, to a reliable storage medium, in a file format that's as future resistant as possible, with metadata, for use as long as one requires." This methodology also ensures readers are not full time digital preservationists. Readers go through this initial procedure so they can listen to music above all else and enjoy the convenience of a music server. A short forty-five hundred words earlier I suggested the ultimate in convenience, iTunes ripping, only lead to major inconvenience, lost music, and hours of precious time wasted. I now suggest the ultimate in convenience is The Computer Audiophile CD Ripping Strategy and Methodology and it leads to increased enjoyment of one's music collection for years to come. A little hard work up front makes for very little effort down the road and saves one from major inconveniences like re-ripping two-thousand discs or investing time and money in obsolescence. The ability to recover from user mistakes and certain hardware failures with a couple mouse clicks is wonderful. Perhaps the biggest benefit of this strategy and methodology is the piece of mind knowing that one's music collection ripped while maintaining flexibility for future interoperability without sacrificing sound quality or usability.

     

     

     

     

     

     

     

     

     

     

     

     

    Footnotes

    [1] Wikipedia - http://en.wikipedia.org/wiki/Digital_preservation

    [2] Wikipedia - http://en.wikipedia.org/wiki/Digital_preservation#Digital_sustainability

    [3] Manual of Analogue Sound Restoration Techniques by Peter Copeland. Published September 2008 by The British Library 96 Euston Road, London NW1 2DB Copyright 2008, The British Library Board http://www.bl.uk

    [4] Oxford Dictionary - http://www.askoxford.com/concise_oed/interpolate?view=uk

     

     

     

     

     

     




    User Feedback

    Recommended Comments



    Difficult no to add congratulations to this solid piece of work.<br />

    Yet I would like to have this question answered: for those of us lazy enough to be happy (for the time being maybe...) with Itunes could someone state clearly that "enable error correction" is a benefit for sound quality.<br />

    Thousands of people are ripping thousands of CD's: I am surprised that nobody has provided any substance to the error correction subject. I wrote it a few days ago: In Europe many people say with confidence that error correction should be disabled, in the US everyone says it should be enabled. Has someone compared both way seriously? For me no difference and I have ripped 3600 cd's so far.<br />

    Any light someone?

    Share this comment


    Link to comment
    Share on other sites

    bluedy1:<br />

    <br />

    It is better to enable error correction in iTunes. However, there is no guarantee that the rip will be without errors. In fact just yesterday I was ripping a brand new Melody Gardot (Thanks, Chris) CD and in the opening track there was an audible error. I deleted the track and re ripped it and it was good. So, iTunes unlike EAC or dbpoweramp is not flawless, at least in my experience. I have ripped over 1,000 cd's in iTunes to date but I will start down this long road again.....

    Share this comment


    Link to comment
    Share on other sites

    Regarding your methodology:<br />

    I'm not going to be a critic, but rather I choose to adopt the ideas that work for me.<br />

    <br />

    I never though about having an original copy and using another for actual playback. This idea combined with another actually saves me money. I was planning to move my drobo to the music server and purchase a drobo share so I could still see the unit over the network. Instead, I'm just going to convert a spare 1tb drive to e-sata or usb and connect it directly to my music server. Now I have redundant/secure storage in the drobo of the original data and a direct connection of another copy of the data for playback:) I would drop the drive in the music server, but we test different machines all the time and that would not work out.<br />

    <br />

    The idea of flac is fine and I understand the concept. However, I never had a problem with this because I don't give my music server the right to change the data on my storage drive:) It can only see the data. <br />

    <br />

    I will try the multi encoding idea though....very cool feature! I'm not sure if it will work for me over the network as I rip away from my music server in my production room. I'll let you know how it works out.<br />

    <br />

    I like to use the directory structure to fill in the data in the media players I use. My reason for this is simple....I understand it and it works:) <br />

    <br />

    Thanks for everything!<br />

    <br />

    Regards<br />

    Jesus R<br />

    www.sonore.us

    Share this comment


    Link to comment
    Share on other sites

    Thanks Alan for your suggestion to enable error correction.<br />

    I am not worried about bit perfect but about the reproduction of music in all its depth and width.<br />

    A french retailer just told me yesterday that error correction adds "noise".<br />

    I have not experienced such thing but I would like to know if others have thought and experienced both options carefully.

    Share this comment


    Link to comment
    Share on other sites

    This is the first time I’ve heard anyone recommend having error correction off. Unless of course you want to speed up your rips.<br />

    <br />

    My understanding is that with error correction on - iTunes goes back to check any data that it thinks maybe wrong. This is why an old scratched disc takes longer to rip compared to a new disc. Can’t see how or why error correction would add noise. I always rip with error correction turned on but have never compared the two.

    Share this comment


    Link to comment
    Share on other sites

    In answer to all the people regarding sound quality, all rippers will sound the same, assuming they ripped without errors. This is the big problem with iTunes, there is no error reporting.<br />

    <br />

    Dbpoweramp, EAC, and Max all use the AccurateRip database to compare your rip to a database of everyone else's rips using a CRC calculation. If the CRC matches, this means it ripped perfectly and you don't need to worry any further about sound quality. If it doesn't match it means you either have a different pressing or you had some errors somewhere. <br />

    <br />

    I don't think I read a paragraph in the article about configuring dbpoweramp to use Accurarterip, though I was glad to see you mentioned it. It requires sticking in a few popular discs until Dbpoweramp gives you the message that says 'this CD can be used to configure AccurateRip, configure it now?'<br />

    <br />

    It's also worthwhile to go into the Secure Settings to see if your CD reader's built in cache can be disabled and whether or not C2 error correction is supported for your reader.<br />

    <br />

    Regarding the concept of doing an archival backup as your primary source, this is a good idea, but there is some other food for thought here. FLAC is a good choice of format, but ripping track by track presents some problems for Archival purposes. Once you've ripped track by track you stand a good chance of losing some details regarding how the original album was put together. This includes lead-in, lead-out, and audio that exists 'between' tracks. For these reason the guys that are crazy about archiving the original 'album' rip to a single file with a cue sheet. The cue sheet preserves all this extra data. The file itself can be any format, flac, wavpack, and APE are popular. I'd agree that flac has the widest support.<br />

    <br />

    This is my only beef with Dbpoweramp, their CUE sheet support is lacking. I know it can generate a single file image rip of all tracks together, but I don't believe it can generate a CUE sheet, which is critical if you want to convert your single file archival copy to individual tracks for maximum compatibility. It's also critical if you want to convert your single file back to a CD with tracks. Dbpoweramp also doesn't have the ability to convert to tracks from a CUE sheet with a single file.<br />

    <br />

    That said, I'm not knocking Dbpoweramp, I definitely prefer it as a ripper for all the reasons highlighted, just pointing out some of the in's and out's of going the archival route. If you did want to go to that level then I believe EAC is the only tool that does this today, and for converting you need to start with the free CUETOOLS to convert the archive to individual tracks.<br />

    <br />

    One other point is that when you're choosing your strategy for archiving and conversion it pays to do some due diligence that there are no bugs with your chosen conversion utilities. There are a lot of encoders/decoders that don't actually convert to bit identical copies. I learned this the hard way a year ago with dbpoweramp's AIFF encoder, it dropped a quarter of a second off every AIFF file when ripping from CD straight to AIFF (This has since been fixed). And I just today discovered a similar issue with Media Rage, a Mac metadata/conversion tool that screws up AIFF when converting from FLAC. (which is a shame as it has really comprehensive format & metadata support, including wav, aiff, etc)<br />

    <br />

    To check if your converted files are still identical to the original, there are three tools I know of that let you test this:<br />

    - dbpoweramp's CRC32 utility codec. The least informative option, but it can generate CRCs of your actual decoded audio bitstream, if the CRCs match then the two audio streams are bit identical.<br />

    <br />

    - foobar's 'Bit Comparator' - this is a utility that can be added onto foobar. Just highlight two tracks in your playlist and select the utility. This then checks the decoded data and tells you whether there are differences. Unfortunately it gives up on more in depth analysis if the length's aren't exactly the same, which seems to be a common bug with converters is to drop some samples here or there.<br />

    <br />

    - EAC's 'Compare WAV' function. This is by far the most informative, as it checks all the samples regardless of a length mismatch, telling you where things are actually the same and where they're different. The problem is you need to convert your files to WAV before comparing them.<br />

    Share this comment


    Link to comment
    Share on other sites

    Idolse stated as fact... <em>"In answer to all the people regarding sound quality, all rippers will sound the same, assuming they ripped without errors. "</em><br />

    This is the belief of most people, however there are some who claim that different rippers sound different, different drives create different sounding rips, and even that adding sound deadening tape to components causes the resultant rips to be improved upon. And that is why people are asking the question of Chris what he feels.<br />

    <br />

    Eloise

    Share this comment


    Link to comment
    Share on other sites

    <cite>there are some who claim that different rippers sound different, different drives create different sounding rips, and even that adding sound deadening tape to components causes the resultant rips to be improved upon.</cite><br />

    <br />

    I realize that a lot of people asking this question don't come from a technical background, and don't fully understand what they're asking. The fact of the matter is that two files with the same CRC CAN'T sound different, as they are the exact same set of bytes. Audiophiles can get a bad rap in the technical community, and most of the time I actually side with audiophiles, as there are a lot of things beyond bit perfect that other people don't take into account. However in this case there isn't any way that two audio files with the same CRC can sound different.<br />

    <br />

    What could possibly make two files with the same CRC sound different are things like where they're stored, perhaps there are actually some users who hear differences because one file is heavily fragmented on the disk drive and the other isn't, but factors like that have nothing to do with the data itself, which can't sound different because it's the same data.<br />

    <br />

    For anyone looking for arguments, note it IS possible for two different files to have the same CRC, however the way it's designed it's not possible for two nearly identical files to have the same CRC. Here's a good article on CRC32:<br />

    http://marknelson.us/1992/05/01/file-verification-using-crc-2/<br />

    <br />

    Note there are 4.3 trillion possible CRC values, I'm not great at statistics, but based on that I believe the odds of the same track of the same cd coming up with the same CRC yet being different should then be 1 in 4.3 trillion. If you think you may have won that lottery you can always use the other comparison tools I noted, which compare sample by sample.

    Share this comment


    Link to comment
    Share on other sites

    Idolse said... <em>"I realize that a lot of people asking this question don't come from a technical background, and don't fully understand what they're asking. The fact of the matter is that two files with the same CRC CAN'T sound different, as they are the exact same set of bytes. Audiophiles can get a bad rap in the technical community, and most of the time I actually side with audiophiles, as there are a lot of things beyond bit perfect that other people don't take into account. However in this case there isn't any way that two audio files with the same CRC can sound different."</em><br />

    I 100% agree with you here ... I'm not in the category of people who think that identical CRC'd files can sound different, though there are some on here who do. It's not even people who aren't from a technical background though.<br />

    <br />

    Thanks for the pointer to the article on CRC ... useful to have some reference to point people to in the future - I understand how CRC type checksums work but my explanation of them was woful last time I tried. <br />

    <br />

    Eloise

    Share this comment


    Link to comment
    Share on other sites

    Anybody who has used one of the AccurateRip programs, such as EAC or dBpoweramp, knows that obtaining accurate digital extractions from CDs typically ranges from probably a low of 60% to a high of 90%. I sometimes hate to repeat info from Wikipedia but I prefer not to reinvent the wheel in stating the problems.<br />

    <br />

    CD audio has two major design constraints that make it difficult to obtain accurate copies in the form of a standard digital file. First, the system is designed to provide audio in real time in order to ensure continuous playback without gaps. For this reason, it does not provide a reliable stream of data from the disc to the computer.<br />

    <br />

    Secondly, the designers felt that it would be preferable for major scratches in the disc to be covered up rather than resulting in total failure. Normally, an error correction system such as Reed Solomon would provide either a perfect copy of the original error-free data, or no result at all. However, CD audio's Cross-interleaved Reed-Solomon coding includes an extra facility that interpolates across uncorrectable errors. This means that the data read from an audio CD may not in fact be a faithful reproduction of the original.<br />

    <br />

    Another practical factor in obtaining faithful copies of the music data is that different CD drives have widely varying quality for reading audio. Some drives such as Plextor are thought to deliver extremely accurate copies while others may do little or no error correction and even misreport error correction information.<br />

    <br />

    Obtaining an accurate digital extraction or "rip" under these circumstances is difficult. iTunes includes an "error correction" mode in its CD importing system. Technical information about this mode is not available from Apple, but it probably ensures that iTunes will attempt to error-correct all data it reads off the disc. However, iTunes does not report if interpolation occurred due to uncorrectable errors.<br />

    <br />

    There is specialized software that will attempt to correct errors, and also attempt to report if errors could not be corrected. They use a variety of techniques, such as making use of error correction information, knowledge of the peculiarities of different drives, and ripping multiple times and comparing the results. All of these programs are still susceptible to some degree to poor CD drives. [End Quote]<br />

    <br />

    Of course all these stupid ripping problems/inaccuracies go away when we get the data files directly from the recording studio such as the HRx files from Reference Recordings. With these studio master files we are assured of the highest qualities and accuracies. That the studio master files are relatively expensive is my complaint, since not requiring any further processing or distribution these files should not be more expensive than CDs, SACDs, DVD-As or LPs. But hey, it's just like when CDs first came out they were priced much higher than LPs, though the manufacturing and distribution costs were so much less.<br />

    <br />

    Now as to identical CRC’d files sounding different let’s repeat what we know. That the identical CRC’d files are data files. You cannot label identical CRC’s files A and B since A equals B and B equals A. Whether I transmit these data files via cheap or expensive cable, toslink, coaxial, Ethernet, USB or wireless the files still remain identical CRC’s files. Whether my computer is using dirty power or clean power, whether it is noisy or quiet, whether my hard drive is solid state or spinning, whether I use an external or internal clock, the data files remain as accurate as before. <br />

    <br />

    And yet the identical CRC’d files sound different. So let’s emphasize what I believe should be obvious that audio playback involves the data file and timing. What’s the difference between an excellent turntable and a budget turntable, an excellent optical transport and a budget transport? If you cannot answer this question for audio it is only because your mind is stuck on only the data aspects of this issue where the cheapest computer from Dell or Apple is capable of producing the same identical data files as the most expensive Dell or Apple computers.<br />

    <br />

    I highly recommend folks try using dBpoweramp. Try ripping 10 CDs with the Burst Mode, Secure Mode and Ultra Secure Mode and note the checksum results and number of re-rips. I can guarantee that you will see differences. I will also say that most of us will not hear the differences that you saw. So I agree with Chris that there comes a point when enough is enough. For all the endings audio debates or arguments, computer audio is not to be torturous, it is to enjoyed and regardless of what may be the best today, I would expect that things will only get better.<br />

    Share this comment


    Link to comment
    Share on other sites

    Everything you say about CD extraction being error prone is perfectly correct. I have in fact seen a number of tracks get ripped incorrectly without a single error being reported by either dbpoweramp or EAC (as you note with iTunes, you'd never know). But in every one of these cases the AccurateRip CRC DID NOT MATCH. A quick cleaning of the CD always resolved this type of error.<br />

    <br />

    As I said before, if the CRCs are the same, it's the exact same data file, therefore it can't sound different. If you sincerely don't believe that it's the same data, <strong>please</strong> use either Foobar and/or EAC as I described in the first post on this. These compare the files sample by sample, which goes far beyond what a CRC does. If all of your same CRC files are different according to EAC & Foobar then you've stumbled onto something interesting. If all the samples are the same then there is something else (i.e. not the data file) making you believe that they sound different. <br />

    <br />

    Regarding things that could make them sound different, you threw out a bunch of comments about different computer/DAC configurations, but it's not clear if they were all things you've actually ABX'd. As I said before, the only thing I can think of which may cause two copies of the same file to sound different in a verifiable ABX type of situation is disk fragmentation. Perhaps one rip was written to the disk highly fragmented, and the other wasn't. It's plausible that this could contribute to Jitter. Going further into crazy speculation (not that I personally believe this), one track may be out toward the the outside edge of the disk platter and one may be inside, causing a difference. To remove both those factors copy both tracks one at a time to a freshly formatted compact flash card and play them from there. I would think playing them over the network or playing from a RAM disk would be alternate ways to accomplish the same.<br />

    <br />

    <br />

    Regarding your challenge, I have ripped lots of CDs extensively, and I've re-ripped many of the same CDs many times, on three or four different CD readers. If the CD is clean and unscratched I will get the same CRCs over and over again. If it is damaged, then I wil get a different CRC for damaged tracks over and over. However the idea with the Secure & Ultra Secure mode is that it's playing a statistics game, the same sector is ripped many times, and one value will (hopefully) appear more than others. That, and often it will match the AccurateRip database, and when I know other people out there got the exact same data as me it's not a fluke. That said, I fail to see how this challenge proves that two files with the same CRC sound different. It only proves the error prone nature of CD ripping, and why AccurateRip and CRCs are a valuable tool.

    Share this comment


    Link to comment
    Share on other sites

    Does xld let you convert say flac to another format also? Thanks jimi

    Share this comment


    Link to comment
    Share on other sites

    I will have to try a quick cleaning of the CD to see if the quick cleaning always resolved this type of error for every one of these cases where the AccurateRip CRC DID NOT MATCH. But my opinion to date is that the Secure and Ultra Secure ripping modes are valuable additions to the AccurateRip programs that do not suddenly become useless if you just do a quick cleaning of the CD before ripping.<br />

    <br />

    But let me simplify what I am saying regarding identical CRC’d files sounding different by just discussing one audio file rather than two or more separate but identical files. The same audio file can sound different, which takes any discussion of CRC out of the equation and there is no need to compare files sample by sample or anything else since it is just a single audio file.<br />

    <br />

    What I had hoped to make clear is that the playback of an audio file involves at a minimum data and timing which is acknowledged by many in computer audio, and there may be other unknowns that if any exist, have yet to be acknowledged or quantified. What I hope that folks don’t believe is that nothing that we do on the digital side of the equation makes any difference because bits are bits, 0s and 1s, and the same audio file will always sound the same regardless of any differences that exist on the digital side of the fence.<br />

    <br />

    The typical computer geek, engineer or lay person will never believe this unless they are also an audiophile. Many of us have heard similar thinking over the years such as all amps are the same, all power cords are the same, all interconnects are the same, vibration does not make a difference, etc. My challenge was not so much to prove that two audio files with the same CRC can sound different, but that with any digital audio file, changes on the digital side of the fence can result in playback differences and quality.<br />

    <br />

    I have stated that if we were discussing the accuracy of a data file, it does not matter whether I transmit or receive the data a few milliseconds or a few hours late, the data will be accurate in almost all instances. This is not true for the audio playback of a data file and I have and continue to ABX many things to explore these differences. But just because I and others believe that there are sonic differences from using different digital cables, hard drives, software players, computers or whatever does not mean that one has to belabor any of these issues or differences to enjoy computer audio.<br />

    Share this comment


    Link to comment
    Share on other sites

    <br />

    AudioZorro,<br />

    <br />

    Very well said!<br />

    <br />

    clay<br />

    Share this comment


    Link to comment
    Share on other sites

    If you're simplifying the discussion to the same audio file sounding different then that's a completely different discussion - you're speaking generally about the finicky nature of computer based audio then, not about CD ripping.<br />

    <br />

    The point I'm trying to make is that CD ripping is one area where there is no black magic to worry about. If you follow Chris's instructions, and pay attention to whether you got an AccurateRip match, then you don't have to worry about whether one ripper 'sounds' better than another.... provided they ripped error free there is no difference in their outputs. If the CD doesn't exist in the AccurateRip database then it's true that you can't be sure whether the data was 100% accurate, but it's extremely likely to be if you use a secure ripping mode, or at least 99.999% accurate.<br />

    <br />

    I'm not on the bit-perfect equals perfect sound side of the fence. When you do get to playing back those files, there are a million things that can make it sound different, but that's another discussion...<br />

    Share this comment


    Link to comment
    Share on other sites

    <br />

    "The point I'm trying to make is that CD ripping is one area where there is no black magic to worry about."<br />

    <br />

    In general, I don't disagree with your statement, in deed, I certainly don't worry about it, as I only recently switched from iTunes to XLD for ripping on my Mac, BUT....<br />

    <br />

    "When you do get to playing back those files, there are a million things that can make it sound different, but that's another discussion..."<br />

    <br />

    ...until we understand each and every one of the million things that can make it sound different, I'm not as confident as you in saying there is no black magic. We just don't know everything yet in the arena of digital audio. Indeed, some things that we all thought only a year ago are turning out not to be true, despite what the hydrogen audio folks would believe.<br />

    <br />

    just my opinion,<br />

    clay

    Share this comment


    Link to comment
    Share on other sites

    If you prefer to be a purely Mac-based computer audiophile and wish to honor the recommended dbPowerAmp, what are your thoughts on running dbPowerAmp in a Windows virtual machine (via Parallels or VM Fusion) within Mac OS X?

    Share this comment


    Link to comment
    Share on other sites

    A couple of comments and a couple of questions:<br />

    <br />

    My experience is (and I have demonstrated at CES, once to each Atkinson, Harley, Fremer, and Connaker) that a track in AIFF sounds MUCH better (more air and vocal/instrumental "reality") than the same tune converted to Apple Lossless (ALC version created from the original iTunes AIFF rip). While I presumed a good portion of the difference could be due to the extra internal "work" having to be done by the computer real-time during playback of ALC files, we still hear a difference (though less so) between the original AIFF and an AIFF made by converting back from ALC. [Full disclosure: All rips and conversions done with iTunes on 1.5mhz G4 Mac mini w/1gig RAM; no attempts made at CRC comparison.] We stumbled upon this years ago (mostly before all this hyper-technical attention had begun being paid to rips) when my biz partner complained that all the ALC music files I had loaded onto our company reference system had a "sameness" to them and seemed dimensionally "flat" compared to the (majority) AIFF-ripped discs on the same system.<br />

    I have not done such comparisons against FLAC, but I am a bit surprised to see people recommending a lossless compression format for archiving. Storage is cheap and there are unanswered questions--even about "lossless."<br />

    <br />

    As for my practical questions (for Clay and others perhaps):<br />

    <br />

    1) I am using a "headless" Mac mini (OS X 10.4.11; next to a prototype USB DAC (Swenson/Hovland collaboration--an Empirical/Centrance USB input feeding an NOS PCM1704 plus all discrete output stage). A 500 gig external Firewire drive holds the tunes. Across the room at my desk I run a 1.67 G4 Powerbook into a nice 24" display. I use JollyFastVNC client to view and control iTunes running on the music-mini. The problem I have is that after a few minutes of the mini playing VNC host when I am remotely controlling it, the mini's fan runs up high and I can launch Activity Monitor and see some crazy-high CPU usage figures (sometimes 45% depending upon what I am doing or just randomly and often "perl" is the hog besides VNC) for running as a VNC host. My current room arrangement precludes use of my reference speakers, so I can't say for sure if this CPU-hogging is affecting the sound, but that seems likely. I know others here do use headless CPUs for feeding tracks to their USB DACs. Please share your thoughts and experiences. Are there any lower-overhead VNC alternatives? (Timbuktu is even worse in this regard).<br />

    BTW, I am pretty committed to iTunes (don't all gag at once) as a tune library front-end (I am happy to give XLD a try for better ripping). It works for me graphically, playlist-wise, and search-wise--plus Apple keeps improving it.<br />

    <br />

    2) Given my simple set-up (quiet, headless Mac connected to some seriously revealing gear: great DAC>Hovland HP200P-i>Hovland RADIA amp), and the talk I hear about faster processors, gobs of RAM, SSDs, and linear power supplies, please comment on how and WHY those upgrades might make sonic improvements. And since I have a limited budget and a working G4 Mac mini, which of the aforementioned computer upgrades are likely to make the most difference? Would an SSD or linear PS added to my current mini do anything? (Who even offers linear external PS units with correct voltages and plug for Mac mini, or I just hack and solder the cable myself?) My daughter could use my G4 mini if there was a compelling reason to swap in an Intel mini. But I would not afford the newest model, so thoughts on the minimum for best results will also be welcome.<br />

    <br />

    Thanks and regards,<br />

    Alex Crespi (from the former Hovland Company)

    Share this comment


    Link to comment
    Share on other sites

    T-Mas - Running dBpoweramp through Parallels is a mess. I tried to get this to work well for quite a while. A big issue is ripping speed. The added layer between the hardware and the virtualized OS really hurt things.<br />

    <br />

    Alex said, <i>"I am a bit surprised to see people recommending a lossless compression format for archiving. Storage is cheap and there are unanswered questions--even about "lossless.""</i><br />

    <br />

    I hear what you are saying. It's certainly something I thought about while working on this strategy and methodology. There are certain lengths I am willing to go to "stretch" what technically makes sense to me and accept some things that have no explanation as of yet. This is one that I'm going to let others spend time working on. Note: I recommend using the least amount of compression possible for my FLAC rips. It's faster this way. As I stated in the article AIFF and WAV don't make good archive formats for many reasons. <br />

    <br />

    <br />

    <br />

    In my opinion a file in FLAC format converted to AIFF, WAV, or any uncompressed format won't sound different from an original AIFF, WAV, or other uncompressed format as long as the files have the same CRC. As I've stated before I'm not a fan of playing compressed files, but I am a believer in using lossless compression for transporting and archiving files. <br />

    <br />

    <br />

    In the spirit of the bulk of the above discussion, I will go on record as stating I do not believe two identical CRC'd files can sound different if all else is equal. In my opinion it's nonsense to believe two of the same thing could give different results if all else is equal. In a way it seems like people are imagining something smaller than a "bit" like a sub-bit that may be different, thus causing a sonic difference. There is no such thing as a bit (binary digit) that's in between zero (off) and one (on). Were are not talking about quantum computing, qubits, and the superposition of bit values here. It's my belief people are barking up the wrong tree. Looking for differences that make things unequal in the playback system is likely the cause of sonic differences people are hearing. Playing audio from a PC or Mac make it nearly impossible to guarantee all things are equal during playback of two tracks. Some of the unequal issues have been mentioned above such as placement of the file on the hard drive. I don't think disk placement is as critical as processes in use on the music server and everything else a computer must do to continue operating, playing back audio is not one of those must do items. Sure this may not explain if groups of people are hearing the same differences with the same tracks shared among themselves. But, the power of suggestion and the mind are always things we must consider. I must consider the power of my mind when reviewing components otherwise I will undoubtedly provide people inaccurate advice and opinions that were derived by things other than the sonics of a component.<br />

    <br />

    Note: The above opinion is not a statement that bits are bits in the sense that all digital sounds the same. Timing is everything (almost) when it comes to playback of audio. The above discussion is about two identical CRC'd files sitting on a hard drive then playing via a system with hopefully identical timing inaccuracies. <br />

    <br />

    Please take my opinion in the spirit in which it's intended. This is a very open and relaxed site where all opinions are encouraged. Nothing I said is to be taken personal by anyone here. We all benefit from everyone's opinion even when we don't think things are possible or someone is way out there (myself included). I encourage people to continue looking into these things. I actually like being proven wrong when readers research a topic and post their findings here on the site. That means I learned something and won't repeat inaccuracies in the future.<br />

    <br />

    Please continue to discuss this topic as you guys see fit. It's interesting nonetheless. Those taking part in the discussion are equally as interesting people that we can all enjoy conversations with.<br />

    <br />

    <br />

    Share this comment


    Link to comment
    Share on other sites

    Chris, thanks for going on the record and being succinct. I agree that two identical CRC’d files cannot sound different if all else is equal. It was the lack of the latter conditional that gave me concern what other readers may be led to believe, not that I doubted the knowledgeable author’s intent.

    Share this comment


    Link to comment
    Share on other sites

    Alex says:<br />

    "I know others here do use headless CPUs for feeding tracks to their USB DACs. Please share your thoughts and experiences. Are there any lower-overhead VNC alternatives? (Timbuktu is even worse in this regard)."<br />

    <br />

    What version of OS X are you running on the Powerbook?<br />

    <br />

    Native Mac screen sharing improved somewhere along the way - Leopard perhaps - and you can even use this to access the screen of your Tiger OS, despite it NOT officially supporting the Apple VNC-like capability.<br />

    <br />

    I use a Macbook Air for headless access to a Mac Mini, a 'headless' Macbook Pro (read - MBP with damaged screen that was removed completely) and a G4 Cube.<br />

    <br />

    It is painfully slow with the Cube, but it's 1/3 the processor speed of your Mini.<br />

    <br />

    I'd say that the intense processor activity is very likely affecting the sound to some degree, whether you will hear it is another question entirely.<br />

    <br />

    I'd try the Apple Native screen sharing (it costs nothing), and if that doesn't help your problem, I'd use this as supporting evidence in question number 2 - IOW, get thee a faster computer. You can also limit the intensity of the Apple Screen sharing - I forget what it's called - but something like screen quality.<br />

    <br />

    Re Q2, I can't speak to the order of benefits, I got a Mac Mini with 4GB RAM, and I run Amarra / itunes on it now. I have an SSD that I"ve yet to install. I have NOT planned to upgrade power supply yet. So that tells you my order, which is based on price = lower cost upgrades first. I'd get an SSD even if I didn't think it might improve the sound, due to non music benefits - speedy, quiet, reliable.<br />

    <br />

    Your Mini seems maxed out at 1Gb, but that's more than my Cube. I think your VNC software is more of a hog than the Apple version.<br />

    <br />

    Welcome to CA, Alex<br />

    <br />

    clay<br />

    Share this comment


    Link to comment
    Share on other sites

    Clay:<br />

    Thanks much for the response. Funny how, if one does not post a bunch, people think you are new to the forum. Click my user name and you'll see I've been a member for 1 yr., 30 wks.<br />

    <br />

    I am running OS X 10.4.11 on both the 1.67mhz PB G4 and the 1.5mhz mini G4. You are correct that Leopard (which I have a copy of but only run on my wife's Mac just yet) was when they added a built-in VNC client. I do not really think the hosting side is any different as I still think one enables Apple Remote Desktop in Leopard--just as I have done in 10.4. The difference is the client I use on the Powerbook side--in this case JollyFastVNC. I could be wrong about all this--I should install Leopard and experiment. But it is jolly-fast. You refer to issues of speed, but I really have none. Screen sharing is plenty fast for my use in iTunes (but I wouldn't try to watch a video this way!).<br />

    <br />

    Chris:<br />

    Thanks for your opinions and statements about lossless FLAC and about identical CRC files. Please note that I did NOT claim to have done a checksum comparison between the AIFF and ALC files (after conversion back to AIFF). I just state that I hear a consistent, repeatable, demonstrable (to others) difference between playback of original AIFF rips (best), Apple Lossless of the same file (worst), and AIFF made by converting ALC back to AIFF before playback (middle). And of course this is all on my admittedly lower performance Mac mini using just the iTunes ripper (w/"Error Correction" on of course).<br />

    Still, I think the issue and my questions are germane to the conversation. <br />

    a) Is FLAC really the best choice for archiving?<br />

    b) Why would computer processing power and RAM have any effect on the quality of rips and/or how lossless files sound when converted or directly played back? The possible answers to this question is more important and interesting to me than the debate of "a)" above.<br />

    <br />

    Ciao,<br />

    ALEX<br />

    <br />

    P.S. Sorry about posting such things into an article thread. I keep forgetting that CA now has a real set of forums. Yet it seems that some of the most interesting dialog occurs in article comments. Chris, have you considered some means of limiting article comments or steering things over to the forums? Maybe it will just take self-discipline on my part...<br />

    <br />

    Share this comment


    Link to comment
    Share on other sites

    Great article, Chris. We should all be so thorough....<br />

    <br />

    How do you feel about using a virtual copy of Windows on a Mac (VMWare) for ripping; and do you think that an external, dedicated CD/DVD drive might be less error prone than any of the internal Mac drives?<br />

    <br />

    Bob

    Share this comment


    Link to comment
    Share on other sites

    I actually did what you're asking last year - ran Parallel's with an external USB drive plugged in. I was ripping 600 cds and wanted every computer available. For whatever reason the Macbook internal drive was worthless for ripping from Parallels, but the external drive did work just fine. There must be some difference in the way the internal IDE/SATA is abstracted vs. USB.<br />

    <br />

    Not sure about Vmware, but it should be ok.

    Share this comment


    Link to comment
    Share on other sites

    Has anyone used one of the automated CD loaders interfaced into an application like dBpoweramp? I've seen units that go for well under $1,000. Would you expect the same quality as "manual" ripping? Are they difficult to set up? Thanks-<br />

    Tom

    Share this comment


    Link to comment
    Share on other sites




    Create an account or sign in to comment

    You need to be a member in order to leave a comment

    Create an account

    Sign up for a new account in our community. It's easy!

    Register a new account

    Sign in

    Already have an account? Sign in here.

    Sign In Now




×
×
  • Create New...