Digitalaudio - the jitter of the clock in your "hardware .wav file player" is critical. If it uses a common crystal oscillator, it will not show you what this medium is capable of. The Master Clock is THE MOST IMPORTANT thing in ANY digital audio system.
Steve N.
Empirical Audio |
Depends entirely on your playback chain. If you have a low-jitter source and a highly resolving system avoiding an active preamp (very important), you will not only get better playback from CD rips than with a CD transport, you will find that FLAC and ALAC files color the sound and compress it. IME the only truly transparent format is .wav. Even AIFF format from Mac causes a slight sonic degradation.
Steve N.
Empirical Audioi |
Jwm - Amarra and PM use the CODECS that are the problem. They work with Apple Mac. You have to turn off itunes and use the playlist function to play FLAC in Amarra.
I prefer a particular version of Amarra, not 2.4, over PM. I have compared it to PM. If you have Amarra, you can download this version here:
www.sonicstudio.com/releases/Amarra_233_Release_4319.zip
They are evidently working on a version of 2.4 that has similar sound quality, but I have not heard it yet. If we can get the stability of 2.4 with the SQ of the older version above, this will be the one to get IMO.
Steve N.
Empirical Audio |
Almarg - there is really easy jitter test for CD transports. Just rewrite a commercial CD track with a CDROM drive from a .wav file on a computer. Use a good disk like Mitsui Audio Master.
If you hear any difference between the commmercial disk and the copy when played on the CD transport, then you are hearing jitter artifacts.
I have yet to play a CDP that does not show this effect.
Steve N.
Empirical Audio |
Jwm - I download all kinds of FLAC files. Then I immediately convert them to .wav. Sounds a LOT better. If you dont hear any difference, its your preamp or your source has jitter too high to hear this. I routinely demonstrate this at RMAF and now at Newport Beach show. Everyone hears the difference.
I know, I know, everyone firmly believes that their system is highly resolving, but get real, it probably isn't. Not unless you have modded every single component including the crossovers in your speakers, used the best cabling and eliminated your active preamp. Then and only then will you maybe be able to easily hear differences in FLAC, .wav and AIFF. I have done all of these.
Steve N.
Empirical Audio |
"It still seems to me that Bit-perfect ripping does provide the best case scenario for reliably getting the data off the optical disk as best as possible"
You would be correct, however as stated before here, errors and error correction is not the thing that makes CD playback an inferior method. Its the CD servo and its effect on jitter from the transport. If you have a transport that is CDROM-based and buffers the data at high-speed in memory, then this can be as good as a ripped file, assuming a good hardware design.
"THere is still lots that can go wrong downstream from there in regards to jitter in particular even with a practically bit perfect ripped .wav file. More so in general perhaps in the case of FLAC which is lossless but compressed and requires more processing in the D2A conversion process."
My theory is that its not jitter differences that makes FLAC sound different that .wav. Particularly with the advent of Async USB.
Its the real-time behavior of the FLAC CODEC when running on the computer. Seems to happen on both PC and Mac.
On the other hand, FLAC CODECs in end-point networked devices, such as Squeezebox have been demonstrated to sound identical to .wav.
Only people using a computer for FLAC decoding hear these differences.
Steve N.
Empirical Audio |
Kij - getting a difference between re-written CDROM and a commercially made CD depends on a lot of things:
1) use Mitsui Gold audio master disk
2) clean the disk well using a good treatment before burning
3) use a good reader and CD burner at 1X or 2X speed, such as Plextor or Yamaha - I used one with a Superclock modded into it.
4) Rip with dbpower amp on PC or XLD on Mac with Accurate Rip enabled
If you do all of these, you should definitely hear a big improvement in the copy, unless of course your CD player is CDROM-based and reclocking the data out of memory, like the Parasound and a couple others.
Steve N.
Empirical Audio |
To the original post: You can achieve very low jitter using either computer/server playback or transport playback, however IME it is easier and less expensive to achieve the maximum performance using a server.
The main reason is that jitter from the CD transport is generally affected by the pits in the CD. When you rip the CD, the jitter in the pits is unimportant and does not affect the image that results or the server playback. The pits are only important for transport playback.
Once you have a digital image, you can apply low-jitter clocking hardware to it that will result in playback with lower jitter than even the best transports can deliver. The exception is the transport based on a CDROM drive that spools the data to a memory and then it is spooled out of the memory. This is essentially a server type playback.
Also, the playback software and CODECs will affect the playback depending on the lossless compression used. IF none is used, you will get the best result, by using native wav format.
Therefore, the result you will get with server/computer playback depends entirely on the playback software used, the format that you rip to and the playback hardware. The playback hardware can be a USB converter, USB DAC, Ethernet network interface or simply a S/PDIF interfaced from the computers PCI bus. The quality of the interface hardware and the power supply for each of these makes the difference.
Steve N.
Empirical Audio |
Dvavc - I'm in central Oregon, but I will be in CA at the Newport Beach show next weekend, Fri, Sat and Sun. My exhibit is in the Hilton room 1001.
"Just prove to me (by audition) that your own CD ROM with source CD won't beat files ripped from it."
I cannot do this at the show because I will not have a CD transport at the show, only an Antipodes server. If you come to Oregon (Bend area), I'll be happy to do this demo. Bring your best CD transport and disk. Just email me.
Virtually all of my customers have given up on CD transports in favor of computer audio because of the improved SQ. Some have even sold their vinyl setups. I have the posts on forums to prove this.
I highly recommend you come to the Newport show anyway. It will be an eye-opener for sure:
http://theshownewport.com/visitors/index.html
Steve N.
Empirical Audio |
"The point here is that you loosing the quality of the sound (SQ) right there when you ripping/converting PCM files from CD in to WAV on your comp. And said loss of SQ can't be compensated by jitter free playback."
That is simply not true. The data fields in the digital copy are identical to the original CD data, assuming that you ripped it properly with a C2 error correcting drive and used Accurate Rip in the ripper.
Some playback software does muck with the data in various ways, such as changing the offset or doing resampling, but these players are well-known and are avoided by most audiophiles.
If you want examples of how some rippers, formats or players compromise the SQ, I can do this at Newport show after hours. Those that have been following computer audio for a while understand which of these to avoid, so the SQ issue with these is really a non-issue for the most part. It's just a matter of education.
Steve N.
Empirical Audio |
Dvavc - If you are coming to Newport show, bring a thumb flash drive with your ripped files on it. Make one FLAC and the same track wav. See you in Hilton 1001.
Thanks,
Steve N. |
Dvavc came to my room at Newport and proposed an experiment, which we tried to demonstrate.
His assertion is that ripping CDs is a flawed process because he hears a difference when playing back the same track on a PC or Mac using first the CDROM as a transport and then with the ripped track playing back from hard disk.
The problem with this comparison is that it is NOT apples to apples. Because the playback hardware paths in the computer for CDROM and memory/disk playback are entirely different and they actually use two different clocks, I would expect these to sound very different. This has nothing to do with the rip quality.
A better experiment is to playback using the same method, but drive through a reclocker like my Synchro-Mesh before the DAC. This removes any differences in the playback hardware. This would allow one to hear if there are any differences in the data from the CD and the ripped file.
If Dvavc had the patience, we could have done more experiments like this. Too bad his mind was already made up.
Steve N.
Empirical Audio |
