Jump to content
Sign in to follow this  

Do Hard-Disk Drives Sound Better Than CD's?

Recommended Posts

Bill, this one is for you...since you've ripped and like the computer playback result.


Also...check out the comments section. Good stuff down in it too.




Touches on a bunch of the issues of CD's and digital playback, error correction, jitter, the flawed USB and SPDIF interfaces and so on! Unless you're 100% vinyl, then these issues are involved in your system and music playback.



  • Thank You 1

Share this post

Link to post
Share on other sites

Thanks for that, Mark. Interesting read. I myself have wondered about the efficacy of USB as a transport mechanism, but always told myself that it has to be better than S/PDIF (especially through TOSLINK). I know that Firewire (IEEE 1394) can transmit without CPU interference (it gets a buffer address and just streams to it), but I don't think that's really the issue (and Apple's latest product omitted Firewire connections — what's up with that?). It's the asynchronous nature of USB that worries me. We went away from parallel interfaces because clocks got better and we could deal with serial bit streams as good as (and cheaper than) word width data. Thus the demise of IDE (PATA) and Parallel printer ports (ECP+EEP). Perhaps the dominant transfer mechanism in the future will be eSATA, but the jury is still out on that one.

Share this post

Link to post
Share on other sites

Over in the Dark Tower, there has been some discussion about how computers were not fit to rip CDs.  If you still go over there, you might want to read the thread, and the beginnings of this thread too, especially the excellent article by Robert Harley.  Anyway, the discussions over there prompted me to post what has become known to some over there as "The Post"

Being an "ancient computer guy" (I cut my teeth on IBM 7094s, DEC PDP-1s and UNIVAC 1108s), I am with you on the bit is a bit thing (even though it's a byte is a byte). The formats you talk about are just wrappers. The data is all the same. Consider if it weren't. You want to download a file from another computer. It's file system could be FAT32 or NTFS or some UNIX variant. Neither computer knowing what the other's file system is organized as are still able to transfer this information, because the convert it to a common carrier communication format (UFT, FTP or some such). These are upper level protocols. Down in the trenches there are different layers of TCP/IP that handle the handshaking between computers, the packet accept/reject/resend mechanisms, the packet assembly/disassembly process (they don't necessarily come in order). All this ensures 100% accurate delivery of the packets of data. The data hasn't changed one bit, but the wrappers surrounding them have. If this process didn't work flawlessly, downloading songs, pictures or programs could never work. One bit out of place in downloaded code could render the entire program inoperable. So the data is not changed one iota from point a to point b.

Now consider ripping a disk on your computer. The format is CDDA (Redbook), which means the data is 16 bit signed 2s complement integers (-32768 to +32767). This is the data. It was written and can be read back at 44,100 of these integers per second. Based on sampling theory math (Nyquist), this means that the frequency response a signal resulting by converting these integers back to an analog signal can be no more than half (22.05kHz). While there may be sophisticated error-correction going on (there are subchannels on the disk for this), the read speed is so much greater than the transfer speed, this process is usually invisible to the application. So, your ripping program just merrily goes on reading the CD into a buffer as fast as it can, and when the buffer fills up, it is written out to disk via the computer's file system (again, the application doesn't need to know what file system it is, that's just a wrapper). Your ripping program may also contain a program that compresses the information to save space. This compression should not be confused with analog audio compression where the music is squeezed into a smaller dynamic range. This compression is digital and has to do with eliminating redundancies in strings of zeros and ones to save space. Consider a 2 second silence. That's about 2.5 million zeros in a row. Now, a code for zero followed by a number (like 2.5 million) takes up far less space on disk than stringing them out. This is what lossless compression does (i.e., nothing is lost). I'm not even going to talk about lossy compression. When the file is uncompressed for playing, these compressions are reversed real time.

These digital compression techniques have been in use since I entered the field at the dawn of the computer age. I can tell you that every major corporation backs up and restores their data using compression without one single thought of dropping a single bit in yottabytes of information.[/quote']
  • Thank You 4

Share this post

Link to post
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
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.