Cd vs flac

wanders

AK Subscriber
Subscriber
I'm sure this topic is not new. If so, I'd appreciate any AK links.

I've ripped a couple of hundred cds to Flac using my office computer and jriver s/w. I copied them onto a 1tb, USB hard drive, play them through my Bluesound Node 2 and sent the digital output, by coax, to the dac in my Marantz Sa8004.

Comparing the Flac files to the donor cd, I hear no difference. But, in a demo session, several audio friends claim to hear differences. (Some say the Flac is superior, some the opposite.)

I can understand audiophiles who have expensive cd systems being reluctant to accept that a modest, non-mechanical system can equal their high-end systems.

But are there any valid reasons for hearing differences?
 
We'll, As mentioned, the original cds were optically read by a computer and Flac/ed through jriver.

Playback of the original cd was through the Marantz sa8004 and its interior dac then to the pre. Playback of the Flac file digital content was from a USB drive to the Marantz dac to the pre.

As said, I don't hear a difference, but people with five figure cd systems say cd is superior.

The tests were not blind as I anticipated the reaction and told them the source in advance.

Thx
 
FLAC is compressed, CD is the uncompressed source. FLAC, while being a lossy format, offers near-lossless sound quality. If you have sufficient storage space, ripped AIF or WAV files can be read by nearly any device and can always be coverted to FLAC whereas FLAC files must be readable by the device that reads it. I find it difficult to believe that FLAC can sound better than AIF/WAV files.
 
I've got 27k worth of flac files, ripped directly from my cds. Now, perhaps it is the Meridian DAC with the Flac files, and the native DAC's in my extensive CD players I am lucky to own, but I prefer the CD every single time.

Recently was listening to the hard drive, a song from Bjork, and it sounded brittle and harsh. Plopped the cd into my recently acquired Denon DCD-1500 II, and I was like, "Ahhh, there it is!"

Your results probably vary.
 
FLAC is compressed, CD is the uncompressed source. FLAC, while being a lossy format, offers near-lossless sound quality. If you have sufficient storage space, ripped AIF or WAV files can be read by nearly any device and can always be coverted to FLAC whereas FLAC files must be readable by the device that reads it. I find it difficult to believe that FLAC can sound better than AIF/WAV files.

You are misinformed. FLAC is a lossless format. That is why they are closer to WAV file size than mp3.

To the OP: what you are really comparing is the difference in your playback systems. The data on the CD is functionally identical to the FLAC.

Edited to add: that is assuming you rip your CD to 44.1khz 16bit FLAC files. Any other FLAC sample/bit rate is a conversion and can introduce artifacts.
 
The reason for the test is basic: I've seen cd systems costing $50-$75 K. Cases milled from solid aluminum, extensive damping, timing, optical sub systems all designed to neutralize intrinsic flaws in (not improve) a cd playback system. Contrast all that with a system that contains a $50 hard drive, $500 file server (streamer - translator) that delivers exactly the same information (digital) into your system (dac, pre, amp, speakers). That kind of thing makes a lot of people nervous.
 
To the OP: what you are really comparing is the difference in your playback systems. The data on the CD is functionally identical to the FLAC.

Edited to add: that is assuming you rip your CD to 44.1khz 16bit FLAC files. Any other FLAC sample/bit rate is a conversion and can introduce artifacts.

As noted, the digital information was delivered to the dac in two methods: 1) from the Marantz sa8004 cd/sacd player, 2) from the USB hard drive, to the Node 2 streamer, to the Marantz dac. From there both were played through the same system: 6SN7 pre, mac 2105 amp, Vandersteen 2ce sig.

All digital info transmitted by Node 2 and received by the dac at 44.1/16.
 
Guess the dilemma is: you bring your $250,000 Ferrari to a race and some kid with a $25,000 Ford Tempo races you go a draw.
 
The reason for the test is basic: I've seen cd systems costing $50-$75 K. Cases milled from solid aluminum, extensive damping, timing, optical sub systems all designed to neutralize intrinsic flaws in (not improve) a cd playback system. Contrast all that with a system that contains a $50 hard drive, $500 file server (streamer - translator) that delivers exactly the same information (digital) into your system (dac, pre, amp, speakers). That kind of thing makes a lot of people nervous.

Please don't forget that besides the digital section of any DAC (stand alone or inside a CDP) there is a analog amplifier. Implementation of this section could produce very different results regarding sound quality. For example, in cheap DACs one gets a IC that does the job and in more carefully designed DACs one gets class A amplification with discrete components, dual mono topology and maybe different choices for the active components (vacuum tubes, MOSFETs, and so on). This analog part of the DAC greatly modifies the outcome signal and has audible effects that may be evident. Besides this, one then may discuss the digital section as well, but perhaps the analog section is the one that yields the more notable differences among DACs (or CDPs).
 
But are there any valid reasons for hearing differences?
As with many components, you really end up listening to the results of its power supply and this is no exception. By looking board level at both the SA8004 and the Bluesound Node, you can easily see differences in design and execution. The Marantz has a toroidal transformer and large filter caps isolated from the circuitry. By comparison, the smallish power supply in the Bluesound lacks those features.

Additionally, if the Node was directly hooked to the 1 TB drive, it was getting noise transmitted from it as well. There are many ways to implement computer audio, but that particular arrangement does not define what's possible.
 
As with many components, you really end up listening to the results of its power supply and this is no exception. By looking board level at both the SA8004 and the Bluesound Node, you can easily see differences in design and execution. The Marantz has a toroidal transformer and large filter caps isolated from the circuitry. By comparison, the smallish power supply in the Bluesound lacks those features.

Additionally, if the Node was directly hooked to the 1 TB drive, it was getting noise transmitted from it as well. There are many ways to implement computer audio, but that particular arrangement does not define what's possible.

This is an interesting concept to explore.

I would like to believe that I could store the digital information from 500-600 CDs on a hard drive half the size of a paper back book, put the physical cds in a box in the closet, and not lose any sound quality.

In this test, all the Node 2 & hard drive had to do was present the digital information to the Marantz dac, which they did electronically with no mechanical effort. The Marantz optical, transport, and damping features were tasked to do the same job - present the digital info to the Marantz dac - but there was mechanical work involved there.
 
This is an interesting concept to explore...I would like to believe that I could store the digital information from 500-600 CDs on a hard drive half the size of a paper back book, put the physical cds in a box in the closet, and not lose any sound quality.
I've effectively done that for years. What I find to be key is isolation from a wide range of noise sources - either computer circuitry or mechanically driven disk drives like the one used.

In the main system, I use a Sonore microRendu, a purpose built streamer which also uses a quad core ARM processor, but little else. That is coupled with an ultra conductor based power supply from Uptone which is galvanically isolated from the AC line. I've found that even its DC cord can make a difference. Mine uses Canare star quad. While I also use a 2.5" disk drive for media storage, it is running in a server located on a different floor of the house - connected via CAT7.

urendu_sm.jpg
 
I've effectively done that for years. What I find to be key is isolation from a wide range of noise sources - either computer circuitry or mechanically driven disk drives like the one used.

I got lots of interference at first, so I moved the Node 2 and hard drive about 20 feet away from the main stack and connected by a Blue Jeans coaxial cable. Lots of noise when the streamer was near the lomc phonostage.
 
I got lots of interference at first, so I moved the Node 2 and hard drive about 20 feet away from the main stack and connected by a Blue Jeans coaxial cable. Lots of noise when the streamer was near the lomc phonostage.
There is still electrically transmitted noise through USB from the drive. Another approach some folks use for handling data storage is employing a distant NAS via ethernet.
 
Additionally, if the Node was directly hooked to the 1 TB drive, it was getting noise transmitted from it as well. There are many ways to implement computer audio, but that particular arrangement does not define what's possible.

The h/d is a Western Digital 1tb, mechanical drive; about $50. The same outfit makes a series of solid state drives ranging from 1tb to about 4tb. Considerably more money. I'm wondering if a solid state drive would address your noise concerns.
 
The h/d is a Western Digital 1tb, mechanical drive; about $50. The same outfit makes a series of solid state drives ranging from 1tb to about 4tb. Considerably more money. I'm wondering if a solid state drive would address your noise concerns.
I use an SSD for the operating system in my computer. Both still generate noise across the USB connection and require more current from the power supply. The approach taken with the LPS-1 and the linear I use with a Raspberry Pi based unit in the garage is to minimize power supply requirements and eliminate connected peripherals.

I will continue to completely isolate storage from the streamers via ethernet or WiFi.
 
The Node 2 claims to provide usb 3.0 power, so there is no additional power supply in the system. H'ever, the Node 2 manual suggests that the usb port should be used for smaller drives. FWIW, I've been adding cds incrementally to see if there is any noticable effect. So far, I've got around 250 cds and 25 hi-res files on the hd. I've not noticed any deterioration in sound or performance in increasing the storage. One blessing is that all of the files are stored on my computer and it's easy enough to put them into another system at a later date..

This file server/streamer idea has been attractive to me for some time. My cd library of 500-600 cds in modest by most standards; even so, at least half of them are in storage, and most of those that are in the listening area don't get played much. Since putting them on the h/d, I'm listening to stuff I've haven't heard in years. I've always resisted placing a computer in the listening area; there is a Bluesound IOS program that uses the iPhone as a controller.
 
The Node 2 claims to provide usb 3.0 power, so there is no additional power supply in the system.
It is of higher capacity, however, than it would need to be without an internal DAC and extra available current. The noise would still be directed from the drive to the source.

I've not noticed any deterioration in sound or performance in increasing the storage.
Disk utilization has no effect on its operation.

II've always resisted placing a computer in the listening area;
Amen!

there is a Bluesound IOS program that uses the iPhone as a controller.
I use the iPeng app on both iPhone and iPad to control the "headless" players. Using LMS, they can also serve as players ,too. :)
 
Your Denon has really brittle sounding DAC chips. Which Meridian DAC? Newer ones sound better. CD and FLAC, on my system, they sound the same and they should as long as all the variables are the same (DAC, AMP, Pre, etc.)

REALLY? Sounds great to my ears. Hmm. And to think, I really like that player. :dunno:

I know you're trying to help, in your own "no sugar in your coffee" way, Bott-ster.

The Meridian is the Director model (I think). When I first hooked it up, I was surprised how much better it sounded than my Dragonfly. If I had to guess, I'd say it is the laptop I'm using, or perhaps that I'm still using Winamp.

Regardless, I've also used a modded PS1, and my Yamaha CDX-1100U, and all sound superior to my digital setup (all ripped FLAC). It is a very small difference, but with those McIntosh XRT20's, you certainly can make it out with minimal effort.
 
Here is my experience:

I have ripped hundereds of CDs to FLAC on my computer. When I send the files vie wifi to my Squeezebox Touch using the digital out into a DAC, the result is better than playing the same CD in CD player and using the digital out into the same DAC. I have used several CD players including heavyweights such as the Denon DCD-3520 and Yamaha CDX-1120. always the same result, a slight preference for the FLAC/Touch vs the CD when using the same type of digital output/input.

The only explanation I have heard that makes sense is that the CD reading is much older technology with error correction working in real time and doing some interpolating that may or may not be 100% correct. The readers in computers are more recent, more accurate and when ripping will go over sections multiple times as necessary to read the disc perfectly.

If audiophiles claim different digital transports can sound different and more advanced ones sound better, it makes sense that a more advanced (even if cheaper and less glamorous) way of reading those discs would yield a better result.
 
Back
Top Bottom