Quote:
Not only is possible for a bottom of the heap computer to function as part of a front end for high end system but given the right DAC, this bottom of the heap front end can and will outperform traditional digital front ends (read: ultra expensive CD transports) will relative ease.

Let's accept this as true.

Why is this the case? Both start with the physical CD which contains an analog of the digital data stream. Both a computer and a CDP use a disk drive to recreate the datastream. EAC and other good ripping programs work faster than real time.

So why would a computer do a better job at this with cheaper components than anything other than a top-line a CDP?

The best approach I am aware of is to clock the corrected/interpolated data out of a RAM buffer, an architecture that can be and is used in both CD players (e.g. Boulder 1021) and streamers (e.g. Klimax DS, Transporter). So frankly, I don't think either approach has any inherent architectural advantage over the other when it comes to ultimate sound quality.

When it comes to convenience, functionality (metadata, album art, ...), etc. streamers win hands down.

Gordon Rankin (Wavelength Audio) once said "Computers are far better CD players. There are only a handfull of CD players that use ATAPI and collect the data off the disk as data. The rest use the SPDIF off the drive which always sucks."

RG

Quote:

Quote:
I am just finding it hard to believe something as poorly made and with such cheap parts as a computer can produce the quality sound (or bit stream) as the mega buck CD player or transport.

You are forgetting that even if the buts are correct, they still have to be presented to the DAC at the right time. If they are not, then sound quality will be affected. Look for the articles on "jitter" in Stereophile's free on-line archives. A lot of what you are paying for in the high-priced player is the ability to present the rights bits at the right time to the DAC.

John Atkinson
Editor, Stereophile

Ok, why does it take megabuck parts in those CD's and transports to do essentially the same thing the computer does built of junk...They manage on a real budget, not the budget of the $5K 'budget' component.

Quote:
Jim,

What you are stubbornly trying no to comprehend are the vast differences between things digital and things analog. In the digital world many of the issues that matter, and matter a great deal, in the analog world do not matter or are of little importance. For example, digital differs vastly from analog in the area of copying: with analog there are always generational losses when one makes a copy but with digital the copies do not suffer from any generational losses - either the copy is perfect or it fails. Black and white, on or off, just like the 1s and 0s of a digital data set.

So while things like build quality can have a tremendous impact within the analog domain, in the digital domain the build quality of one's computer, as it relates to the "sound" of a digital file stored on that computer, is of absolutely no consequence.

Then why has the magazine written glowing reviews over the years of megabuck transports? If the exact same thing can be accomplished with a $500 or so computer, why would anyone buy a transport, ever? If the digital side of a CD player (all the thing before the built in DAC)is of no consequence, then megabuck CD players are also a sham. One could build monument to modern art on the outside and, like my computers, fill most of it with printed circuit boards, unmodified computer laser trays, etc and only get serious at the DAC and later. Why, the $20K CD player could suddenly be sold for 1/10 the price. As the sound would be unchanged, why bother buying odiously expensive gear. Save the money for the rest of the system.

Quote:
Perhaps this is why so many of the music servers built by traditional high end audio companies come up short with respect to their sound. Too much attention paid to digital build quality and not enough attention paid in the areas that truly matter, like digital to analog converters and the analog side of the unit.

An interesting take.

Quote:
So too sum up:

Not only is possible for a bottom of the heap computer to function as part of a front end for high end system but given the right DAC, this bottom of the heap front end can and will outperform traditional digital front ends (read: ultra expensive CD transports) will relative ease.

Edit: By the way, there is a plus side to all your questions since all of this back and forth has managed to turn this thread into an excellent primer on the advantages of a computer based digital music system.

If what you note is true, then the Magazine should say as much and stop reviewing expensive front ends. If their sound results can be duplicaed by gear of only a fraction the cost, then reviews of stuff like this:

Should never, ever occur again. They should simply say, skip this thing and buy a music server and put your money into the best DAC you can find.

Quote:

Quote:
You are forgetting that even if the [bits] are correct, they still have to be presented to the DAC at the right time. If they are not, then sound quality will be affected. Look for the articles on "jitter" in Stereophile's free on-line archives. A lot of what you are paying for in the high-priced player is the ability to present the rights bits at the right time to the DAC.

Ok, why does it take megabuck parts in those CD's and transports to do essentially the same thing the computer does built of junk...They manage on a real budget, not the budget of the $5K 'budget' component.

In general, the computer _doesn't_ manage to do as good a job when it comes to ensuring a jitter-free conversion to analog. This is not a high priority for computing engineers. Again I urge you to read all the articles on jitter in our archives. A DAC installed in a PC may receive the same bits, but it will generally produce an inferior-sounding analog signal from those bits compared with a standalone CD player, in my experience.

For the best sound quality using a computer as the audio file source, use a soundcard with a digital output and feed that datastream to a standalone high-end DAC like the Benchmark DAC 1, the Bel Canto DAC 3, etc.

John Atkinson
Editor, Stereophile

Thanks, that makes sense.

Quote:
Gordon Rankin (Wavelength Audio) once said "Computers are far better CD players. There are only a handfull of CD players that use ATAPI and collect the data off the disk as data. The rest use the SPDIF off the drive which always sucks."

To the best of my knowledge Gordon is correct although I don't think that statement taken in isolation necessarily conveys the full picture.

Very few (if any?) audio CD players use ATAPI because it is a standard for connecting computer storage peripherals, it has nothing to do with audio. It is a completely unnatural choice for an audio CD player as is not supported natively by the chipsets in the popular CD modules so would require addtional design work for no obvious gain. The digital interfaces offered by the most common CD modules are I2S, EIAJ and EBU. The first two are essentially the same and are prefereable from a jitter perspective as they present DATA and WCLK on separate pins. EBU, or IEC (60)958, is essentially just another name for S/PDIF which, as Gordon says, "always sucks".

In the modules I am familiar with error correction is identical in both CD-DA and CD-ROM operating modes. The only significant difference between them is that that mute, fade and interpolation (error concealment) functions are disabled in CD-ROM mode and any uncorrectable errors are just passed on and flagged on the EF pin.

Quote:
In general, the computer _doesn't_ manage to do as good a job when it comes to ensuring a jitter-free conversion to analog. This is not a high priority for computing engineers. Again I urge you to read all the articles on jitter in our archives. A DAC installed in a PC may receive the same bits, but it will generally produce an inferior-sounding analog signal from those bits compared with a standalone CD player, in my experience.

I'll agree that pristine sound quality is not a priority for PC designers, but I'm sure there are sound card engineers who have sound quality as a very high priority. A sound card on the PCI bus should have no difficulty receiving data at very high rates (e.g. multichannel 192kHz/24-bit) then clocking that data into very high quality DACs with a stable, low-jitter clock (not the bus clock). The LynxTwo cards, for example, deliver specifications that are at least equal to many players that qualify for high-end.

On another point in this thread, I agree that CD transports are obsolete and have been superceded by the media server/networked DAC model. I've you're going to have two boxes to play CD's, you should get all the convenience bennies of of the server/DAC model.

Quote:
The best approach I am aware of is to clock the corrected/interpolated data out of a RAM buffer, an architecture that can be and is used in both CD players (e.g. Boulder 1021) and streamers (e.g. Klimax DS, Transporter).

This does appear to be the thing to do.

PS Audio's Perfect Wave Transport (with color touchscreen) may turn out to be a breakthrough product that combines computer and CDP functionality and benefits.

Interesting looking product, thanks for the heads-up Elk, although I have to say I am a little bit put off by the constant use of the word 'perfect' and claims of 'jitter-free output' etc. That said, Paul McGowan has a good reputation for great sounding, good value products so it would be rash to dismiss it out-of-hand just because of a touch of over-ripe marketing. I look forward to reading more when it hits the streets which I believe will be in a couple of months.

"Over-ripe marketing" is a delicious descriptor.

I do expect it to be a great product.

Another interesting product is the just-released Bryston BDA-1 DAC, which also promises 'jitter-free.' It purportedly re-clocks everything as the signal enters the DAC. Impressive specs, too, quality-build, as usual, from Bryston. I also love the fact that you can choose to forgo upsampling/oversampling.

The Bryston does look interesting.

I like the first review as it incorporates an audio clich

Quote:
... (cue sound of angels).

Angels have a sound? Have a link? And don't send me to Emmylou's website!

RG

Quote:

Quote:
... (cue sound of angels).

Angels have a sound?

Yes, but it has a heavenly veil which is a coloration & must be lifted. It also lacks PRAT.

Quote:
Yes, but it has a heavenly veil which is a coloration & must be lifted. It also lacks PRAT.

Well done!

Now maybe someone can tell us how many angels can sing out of a three inch tweeter.

Quote:
Another interesting product is the just-released Bryston BDA-1 DAC, which also promises 'jitter-free.'

SD, Reading through the blurb this looks like just another DAC to me. Lots of DACs reclock the datastream, from the much vaunted Benchmark DAC-1 to my crusty old Boulder. Althouh this is doubtless a very worthy product I don't see anything particularly game-changing here. Am I missing something?

Amazing voice, BUT, she got fat and in the world of fat singers, Ella is the Goddess.

Quote:

Quote:
Amazing voice, BUT, she got fat and in the world of fat singers, Ella is the Goddess.

I beg to differ: The Divine One, Sarah Vaughan, also became rather large as she grew older and so shared the title of Goddess with Ella.

...and for all the wonder that is the Divine One, my favorite album of hers is still the not quite part of the canon 'I Love Brazil.'

s.

I just looked into buying a squeezebox.

Then I find out that they are wireless.

Scratch that. No wireless in this house. Period.

They are both. For a mere $300.00 the user is actually given a choice.

RG

Choice, what a concept.

Question:

I have a Squeezebox and I *thought* it contained a pretty good DAC of its own, a Burr-Brown something or the other.

Where does a separate DAC go in the chain...or (slaps head) are you talking about having the Squeezebox wired?

s.

Hi All,
I found this thread very interesting and informative. I know I'm coming late, when all has been said and dusted, but felt that after all this information has been provided, no one has actually commented on the actual sound quality of a computer based audio system.

My audio setup comprises a Linn turntable, Marantz 63 (used as transport) MSB Platinum DAC, Audible Illusions preamp, Audio Note Conqueror valve amp and Audio Note A-NE speakers.

Recently I bought a music-designated computer, fitted it with the quietest and fastes components, a Lynx L22 soundcard and several 500Gb hard drives. The idea is to start digitising/archiving my 2000+ LP jazz collection. (On that in a separate thread).

I am extremely happy with the sound quality from this computer source. For a number of CDs, I have compared the sound from the CD player DAC combination with two computer based setups: the first with the soundcard sending the digital signal to the DAC via S/PDIF , and the second with the computer sending the analog signal directly to the preamp.

The difference in sound quality was very noticeable and was mainly in the detail and dynamics of presentation. The result is quite surprising, particularly considering all that was said in the above posts. The weakest of the three was the computer to DAC setup, while the best was the computer to preamp.

Having convinced myself, I am now in the process of transferring all my CDs and LPs to the hard drive, from where the possibilities are endless.

Thanks jazzman, glad to be here.
Thanks also for the suggestion re Squeezebox, I'll have to give it a try.

oops, something went funny with the quote here, sorry. I'll continue anyway.

WRT transferring LPs to HDD, basically, I have been using Audacity, which I found to be better than some of the other programs that could be downloaded and would also offer a trial run. It is free and supports 24/96, which is what I'm saving the files as (WAV). Audacity also has the capability to clean the sound file, but I've only used the declicking feature so far.

After they are saved on the hard drive, I simply play them through iTunes, after I've enabled it (through Quicktime) to play the hi rez format. I also converted all the files to AAC lossy format (320 kbps) and transferred to my iPod, so that I can play it in the car, or elsewhere.

Finally, the sound that I'm getting is amazing - I am not able to differentiate the direct LP sound from that coming from the computer. I've also played with various iTunes lossy conversions and can report that the differences are not that great. Of course, the 24/96 sound is just amazing, with the main difference between the 16/44.1 being in the 3D presentation of the soundstage. If you are into hi rez sound, you should also try a few tracks from iTrax, this music was actually recorded in 24/96.

The process of transfering music is a bit slow as you can imagine, since every side of the record needs to be sliced into tracks, during which time I also clean up most of the large clicks / pops. The saved WAV files can be cleaned up at a later time (Audacity or any other program), which is something I plan to do. Those small persistent clicks and pops are the bigget problem. I will probably use iZotope or Adobe Audition 3 for this, but it is something where I am still unsure, and any suggestions will be welcome.

All this is a work in progress, I was able to archive only about 20% of my record collection so far, but it has enabled me to listen to some records that I have not played in years.
Sorry for the long post, but hope that this answers your questions.
Cheers,
D

Audacity is a good basic program. Good choice.

iZotope RX is an excellent noise reduction/spectrum repair program, but it is a bit expensive ($300.00 street) and takes a lot of computing power - as well as expertise.

As a general rule I suggest not using any noise reduction or declickers. They all have artifacts, even the best, and unless the removed noise destroys one's enjoyment of the recording the damage done by removing it is not worth the trade-off.

Thanks Elk.
I came to a similar conclusion. I want to limit usage of a noise reduction program to minimum, because of the obvious impact on sound in general. I plan to use it manually, only on obvious clicks and pops. Audacity sometimes has trouble even dealing with these, hence the need for something more powerful. That's why I mentioned iZotope, which I could use for more extensive noise reduction trials, just to see what the result will be. I always intend to keep the original file though, in case the trial does not go as intended.

The great thing about all these programs is that they have a preview function that lets you listen to the changes prior to saving the file. This may or may not be of benefit, depending on what audio playback one has on the computer.

You may want to try manually re-drawing the waveforms to remove the big clicks. With practice you can get very good at it. While a bit crude in concept it does not do any damage to the surrounding music and only affects a very brief transient. Plus it's amusing to do.

I don't worry about the clicks and pops unless there huge and really distracting. For me the rare little bit of noise is just part of listening to analog and am I find it amusing when these are transferred to digital. I probably would be more aggressive if I had some old worn favorite LPs to transfer or there were otherwise a lot of issues.

Let us know what you find works after you experiment.

Hi jazzfan and thanks for the suggestions.
I also had a second look at the Squeezebox and transporter and decided that with my current computer/soundcard configuration, I actually don't think that I have a problem, so spending in excess of $2K would not be justified. I am happy with the DAC in the Lynx L22 and there is virtually no difference between the LP sound and the converted hirez digital sound that I'm getting.
However, I do have a question. Is there any difference between apple lossless and WAV hirez? Sonically? sizewise? I converted a couple of files and they still had the same size, so I'm wondering if something went wrong there?
Cheers,
D

Quote:
However, I do have a question. Is there any difference between apple lossless and WAV hirez? Sonically? sizewise? I converted a couple of files and they still had the same size, so I'm wondering if something went wrong there?
Cheers,
D

There should be no difference sonically since on playback the losslessly compressed file is restored to be an exact copy of the original.

Sounds like something might have gone wrong in your case since the ALAC file should be smaller than the corresponding WAV. For most music encoded at 16/44.1 the ALAC file is generally about 40-50% smaller, the actual degree of compression is program-dependent.

I have seen it suggested that when compressing 24/96 WAV files one should expect a lower rate of compression since the analog noise floor will be significantly higher than the digital one and noise because of its random nature does not compress. I am slightly sceptical to this but am not familiar enough with how Linear Predictive Coding deals with samples of a signal with a higher noise floor (as opposed to samples of pure noise) to be able to comment. My gut feel is that you should still be seeing some degree of compression though.

Elk, can you offer any empirical datapoints on compression rates of 16/44.1 program versus 24/96?

DraganH, How (exactly) did you perform the conversion?

I first adjusted the settings under the iTunes advanced/import/ tabs. Under apple lossless conversion, there was only 'high(est) resolution' setting, so I picked that one. Sorry, I'm going from memory here, because I don't have iTunes on my work computer here. Having adjusted the setting, I then did a right click on the track and chose 'convert to apple lossless'.

I did a quick test. I took one of my 24/96 FLAC files (the Introitus to Mozart's Requiem, SCO/Mackerras) and converted it to WAV. The file size, bit rate and sample rate as reported by Windows Explorer (Vista) for the WAV file were 146.02 MB, 4,608 kbps and 96 KHz (sic) respectively. For the FLAC original Windows reported 87.43 MB (41% compressed), 4,608 kbps* and 96 KHz.

I then converted the WAV file to ALAC using iTunes 7.7.1.11 using right click and 'Convert Selection to Apple Lossless' as you describe. iTunes reports the above parameters for the resulting ALAC file as 89.2 MB, 2,813 kbps, 96 kHz, however for the same file Windows Explorer reports 89.2 MB (39% compressed), 2,116 kbps*, 44.1 KHz(!)

Just to get a 'second opinion' I opened the ALAC file in foobar (version 0.9.5.5 with ALAC decoder plug-in 1.0.1), which also reports this as a 44.1 kHz file but with a bit rate of 2,814 kbps, although interestingly it refuses to play it claiming an 'Unsupported format or corrupted file'. I suspect the foobar ALAC decoder is based on David Hammerton's reverse-engineered code in which lack of support for 24-bit files is a known bug that nobody seems to have gotten round to fixing. The file plays fine in iTunes/Quicktime.

So, observations from this are:
- Someone at Microsoft needs to take the remedial class on SI prefixes.
- Both ALAC and FLAC do usefully compress 24/96 WAV files (in this example by 39% and 41% respectively).
- Windows Explorer and foobar both appear to have bugs in their reporting of the audio parameters of ALAC files (at least hi rez ALAC files). The file size is the giveaway here, it is consistent with a 24/96 file compressed 39% but it's too large to be 24/44.1, even uncompressed. Although the sample rate and bit rate for this file as reported by Windows are consistent with each other, both would appear to be incorrect. One can only speculate as to where Windows is getting these incorrect data.

* Slightly confusingly 'bit rate' is interpreted differently by different programs. Windows appears to report the PCM (i.e. uncompressed) bit rate even for compressed files whereas both iTunes and foobar report the bit rate as seen by the decoder (i.e. compressed).

Using the 24/96 WAV files from the new John Mellencamp release, I note that iTunes 7.7.1 reports both the native WAV and the encoded Apple Lossless files are 24/96.

The size of the sample track is of course smaller (65mb vs. 105mb), and the bit rate is lower for the lossless tracks (4600 kbps vs. 2908 kbps).

I have no other software to evaluate the files.

Brian

Quote:
About the only thing I do know about both of these formats is that dBpoweramp works very nicely for converting files in these formats to flac, provided of course that the files do not contain any DRM.

There is no DRM in ALAC, I think you are confusing it with AAC.

Gentlemen,
Thank you very much for this detailed explanation.
I'll have to see where I went wrong, because this seems to be quite a saving in HD space. However, I still think that I will not be able to go down this path, since I plan, some time in the future, to firther clean up these files and the programs that I plan to use (Audacity, iZotope, and Adobe Audition 3) might not be able to support these other formats. Any thoughts?

Afaik neither Audacity nor Audition support either FLAC or ALAC. I believe iZotope supports ALAC but not FLAC but for broadest compatibility with these programs WAV is probably safest. If hard disk space is tight however you might as well compress; since they are lossless formats you can always restore a file to WAV if you need to work on it at any time in the future.

Given the current lack of support for hi rez ALAC outside of iTunes/Quicktime (afaik all other decoders are based on Hammerton's code) I would recommend FLAC over ALAC for broadest compatibility. I don't think iPods support 24/96 audio anyway (does anyone here know?) so ALAC doesn't appear to offer any compelling advantages.

Hi jazzfan,

Just a couple of comments.

Quote:
FLAC does offer at least one compelling advantage when it comes to ripping a CD to FLAC which is that both EAC and dBpowerAMP work with AccurateRip, the online database used to help ensure that the contents of the CD have been extracted accurately. iTunes and ALAC do not support accuraterip.

AccurateRip is just a database of checksums, it is not tied to any particular codec. dBpoweramp supports AccurateRip so AccurateRip can be used when ripping to any file format that dBpoweramp supports, which is lots (including ALAC).

Quote:
ALAC does support DRM, just as Windows Media lossless does.

I guess we're just going to have to agree to disagree here. ALAC is a Codec just like FLAC (very similar indeed in fact) and just like FLAC it could be wrapped in a DRM wrapper but hasn't been. Since it is a proprietary protocol of course Apple could add DRM if they chose to. Hey, it's software, you can make it do whatever you want! However I can assure you that there is no DRM on any ALAC file currently in existence, FairPlay has only been used with AAC.

To disprove that conclusively you would have to produce a DRM-protected ALAC file which will be tough, because there aren't any. However I am willing to consider any other proof points you can offer!

Definitely.