The title is:"There's No Such Thing As Digital..."


Subtitled: "A Conversation With Charles Hansen, Gordon Rankin and Steve Silberman". It's an interesting read if you're not yet familiar with this particular topic...or have only considered it briefly. I wouldn't call myself a digital expert, but I can see no reason to quibble with it one bit:

www.audiostream.com/content/draft

Enjoy.
ivan_nosnibor
So it's more than meets the eye (ear). It makes me wonder just how long before PC audio is perfected and glad that I've stuck with those shiny discs. :-)

All the best,
Nonoise
I can see lots of reasons to quibble with it. It leaves out a lot of pertinent info about digital. Digital is of course just an analog waveform as all electricity is, however sometimes digital is just digital and the analog aspect is unimportant. Also, digital has attributes that analog does not have, such as reflection effect on jitter, signal integrity, current-path characteristics and rise-time effects etc.

Steve N.
Empirical Audio
Instersting read, all I know is I won't be going back to CDs as my Mac Mini has never sounded better. My Evo clock made a noticable difference to the sound. I'm still not sure what jitter sounds like, but I know I like less of it!
The science mafia over at Head Fi had a field day with this. They love to shred any idea that claims subjective imprvement. This article smacks of USB cable wankery in their opinion. I offered it as some explanation as to why a cable could make a difference in sound.
"07-24-13: Gregfisk
Instersting read, all I know is I won't be going back to CDs as my Mac Mini has never sounded better. My Evo clock made a noticable difference to the sound. I'm still not sure what jitter sounds like, but I know I like less of it!"

I'm not sure I get that. I hear comments all the time stating computer audio sounds better than a CD player. Why? I don't understand how ripping a CD and playing the audio file on a computer would sound better than just using a CD player.
Stopped reading after the first answer in this Q&A therapy for trusty ones. I don't want to use word "gullible"... This industry is truly something. They view us as boobs. OMG, Ones and Zeroes and the problem of voltage.
07-24-13: Zd542
I'm not sure I get that. I hear comments all the time stating computer audio sounds better than a CD player. Why? I don't understand how ripping a CD and playing the audio file on a computer would sound better than just using a CD player.

Playing a CD is not the best way to source a digital data stream intended for music. Any timing errors expressed as inconsistent spaces between the CD's pits enters the digital/analog converter and distorts the shape of the resulting analog waveform compared to the one originally recorded and digitized. Mass market CDs have jitter just from the stamping of the pits into the substrate. Then the disk is played back, usually on a not-very-rigid CD transport, often made of lots of flexible plastic. As the data stream is transmitted to the D/A converter, more jitter can creep in from reading and re-reading data, from varying impedances, etc. Ripping a CD to a computer hard drive eliminates these timing errors. New timing errors can occur from the hard disk read, especially if the music file is fragmented among several blocks scattered around the disk, but if the data is buffered in RAM all timing errors are basically reset to zero.

Compared to most CD players, a home computer can be a very powerful playback system, with a CPU whose cycles are measured in Gigaseconds, several Gigabytes of RAM that can be co-opted to buffer the data stream and stabilize it, thus removing jitter that might have accrued along the way. Finally the computer-sourced music files are sent to the DAC via an asynchronous USB link. USB, unlike S/PDIF, is a bi-directional link. It doesn't need the clock signal that comes down the datastream pike; it can generate its own freshly when it enters the D/A converter, making for a jitter-free conversion. This is a significantly different way to convert a digital music stream from the typical CD player.

In fact, for a very long time I've wondered why high end audio companies kept
resorting to expensive Herculean methods to lower jitter--diecast metal transports, thick faceplates, expensive vibration-reducing footers and damping materials, etc., when all they really needed to do was put a few megabytes of RAM in there to buffer the data stream and re-clock it. That's what the Genesis Digital TIme Lens (DTL) of the mid-'90s did--for $1195.

Founder Arnie Nudell used to cut demo CD-Rs to show off his speakers by playing the CD's data stream through his DTL before burning the CD. The resulting CD sounded better than the original and made his speakers sound better at the shows.

There is a wide variety of music playback computer software out there, and most of it is way cheaper than a CD player. I use Audirvana Plus on my MacBook Pro. With it I can configure Audirvana to freeze email and automatic backups so the CPU is never interrupted from playing music. I can reserve up to 8 GB of RAM to buffer the music files before decoding. This is especially helpful if you keep your music files on a USB hard drive. I also configure it to upconvert Redbook CD files to 88.2Khz and 48Khz files to 96Khz. If I ever get a 192Khz capable DAC I'll upconvert Redbook files to 176.4 Khz and 48 and 96Khz to 192Khz.

All these things can make the original digital file sound better on conversion, and most of these things are not available on CD players under several thousand dollars.
Hello Johnnyb53. Thanks for the overview.

I didn't think there was any such thing as 100% jitter free? I thought all timing sources had jitter? Have you also overlooked the fact that some CDP's even reasonably priced ones, use high speed drives and buffer into memory for CD playback?

Computers were designed for word processing and have some of their own challenges likes noisy fans, cheap and very nasty RFI / EMI emitting power supplies, CPU interruptions etc.

Finally, jitter is a timing error. At what point does that become audible? Can our ears really detect say 400 pico seconds of jitter? If not then even a basic CDP is acceptable.
Kiwi, I suppose if I want to clutter my post with un-ending qualifiers, I should have said "drastically reduces jitter" instead of "eliminates." Still, if you take the time to check the spectrum analyses of several Stereophile test reports, you'll see that the asynchronous USB DACs have very low jitter that is barely visible in the graphs.

Second, I didn't know that reasonably priced CD players were speed-reading and buffering the data first. Can you name some? Since the audio press is making a big deal about the just-released $6,000 Parasound CDP featuring this read-and-buffer feature, it didn't seem that it had gone mainstream yet.

Third, your summary of what computers are for is narrow and dismissive. I landed in Silicon Valley in 1980 and worked in high tech computers from then until the end of 2006. Word processing applications were a relative latecomer. Also, when the industry switched to graphics-based interface and displays, the desktop-publishing apps were some of the most CPU-intensive applications available, right up there with finite element analysis and solids modeling.

Along with that, my MacBook Pro has a 2 Ghz processor and 8 GB RAM, far more processing power than a typical--or even expensive--CD player or DAC. It has an aluminum housing (well-shielded), and no fan. AS I MENTIONED BEFORE, the Audirvana software (and several other packages) can be configured to turn off all CPU interruptions. It's called "hog mode." Look it up.

Once a music data file is buffered in RAM, the clock is reset. It doesn't matter how much jitter was in the stream before, for the moment, bits is just bits. At that point when it enters a new stream to the DAC, it's coming with a fresh and reset clock and is not subject to additional jitter coming from reading a wobbling plastic disc.

Finally, it was Ed Meitner of Museatex who discovered and published about jitter over 20 years ago. At that time listening tests revealed that jitter became audible around 200 ps. This presented a challenge to the industry as the most popular receiving chip at the time was only accurate to 20 ns.

And anyway, at some point arguing the numbers becomes a moot point. With standard CDPs not only did I find myself not enjoying the music, I even noticed that the family got more irritable when the music was playing. With my current computer setup, I can actually enjoy digitally-sourced music. I still play a lot of records and go to live concerts for reference, but the computer-based (especially high-res) digital playback is closing the gap.
Can our ears really detect say 400 pico seconds of jitter? If not then even a basic CDP is acceptable.

We can detect perhaps as much as 50ps of jitter while basic CDP can go as high as few nanoseconds.

The reason why jitter is so audible, in spite of small levels, is because it creates sidebands that have no harmonic relation with root frequency. If jitter comes from 60Hz noise then these sidebands will be +/- 60Hz apart from the root frequency and not that audible, but if jitter is caused by higher frequency resulting sidebands will be further away hence more audible. In reality there is some uncorrelated jitter coming from random noise and correlated jitter caused by particular interference frequencies. Also, instead of one root frequency we have whole bunch of them (music) and jitter turns into hash that is proportional to amplitude of the signal (undetectable without signal).

Computer data has no jitter because it has no timing. Data is stored on hard disk without timing. It goes thru all sorts of buffers before it is send out. It can be send out as data when we have wireless or network based DAC but it can also be converted to asynchronous S/Pdif stream. The very moment of this conversion creates jitter.

I also have quibbles with this article. It seems to concentrate on DAC clock, that is usually not that bad, placing less attention to delivery of the signal. It should state that both are equally important.

Assuming perfect buffering of the CD stream signal has to be delivered to DAC with very short transitions to reduce threshold uncertainty - requiring perfect source/cable/dac characteristic impedance match (to avoid reflections on impedance boundaries) or perfectly quiet system with perfect shield on the cable to avoid noise induced jitter when transitions are slow. Since both are, being system dependent, very difficult to do possible solution is to reclock the signal just before the DAC. I have DAC with reclocking built in and it is very clean sounding but Steve found that external reclocking works better.
"I didn't think there was any such thing as 100% jitter free? I thought all timing sources had jitter?"

There is no such thing. Marketing BS and an outright lie.

"At what point does that become audible? Can our ears really detect say 400 pico seconds of jitter?"

Depends on the system. Resolving systems can easily demonstrate difference between 20psec and 100psec of jitter, and it's not subtle. Resolving means ultra-low noise floor and distortion. Usually no active preamp to achieve this.

"I also have quibbles with this article. It seems to concentrate on DAC clock, that is usually not that bad, placing less attention to delivery of the signal."

That is interesting, given that many DACs don't have an internal clock, except maybe for the Async USB interface master clocks. These are usually the important ones. This is where the jitter starts in a USB system. It will even have an effect on additional reclocking.

"I have DAC with reclocking built in and it is very clean sounding but Steve found that external reclocking works better."

That is primarily due to the separation of power systems, putting the master clock on its own power system, separate from the DAC circuits. If you can do this effectively inside the DAC, that is fine too. Pretty awkward to have two power cords though...

Steve N.
Empirical Audio
That is interesting, given that many DACs don't have an internal clock, except maybe for the Async USB interface master clocks

Steve, you're missing all network DACs including Ethernet, Firewire, Wireless etc. You also forgot about asynchronous reclocking DACs.

Typical DACs contain Phase Locked Loop (PLL) that contains adjustable oscillator and phase detector. Phase detector compares average phase of incoming signal and adjusts internal oscillator (a clock) to match it. DAC is clocked from this adjustable clock and not from the input signal. Quality of this internal clock (jitter) is very important.
"Steve, you're missing all network DACs including Ethernet, Firewire, Wireless etc."

These are usually not called DACs, but music processors or servers etc., but you are correct.

"You also forgot about asynchronous reclocking DACs.

Typical DACs contain Phase Locked Loop (PLL) that contains adjustable oscillator and phase detector."

These may be typical, but I didn't forget about these, I totally ignored them. I don't even consider resampling DACs because the effect of resampling is so damaging to audio quality, even using a high-quality clock, which most don't have. I would never design such a DAC. Once you have this, no matter how good the input source is, the internal clock and resampling wrecks it IME. I have modded more than a dozen such DACs in the past 12 years, so I know how this sounds.

Steve N.
Empirical Audio
You probably think about oversampling and that is different. Pretty much every CDP contains PLL. It can be used straight just to suppress jitter providing stable clock (by means of averaging) or can be used to create oversampling. Such oversampling is done by comparing in phase detector incoming stream to divided down frequency of higher frequency internal clock. Because this division is integer these oversampling DACs always operate on multiples of incoming frequency while upsampling DACs can work on pretty much any non-integer ratio. Usually PLL is inserted somewhere. Even asynchronous rate converter based DACs (upsampling DACs) like my Benchmark have some form of PLL to make signal stable enough for upsampling. Such PLL is fast responding and single stage while most of CDPs have dual stage PLL that operates at different time constants.
So it seems there's more than enough room on the head of this pin for everyone to dance on.

One can go on for great lengths citing this and that but in the end, using my own ears, I've yet to hear a PC system easily trounce a really good CDP. The caveat in my belief is that I've only heard these PC setups at audio shows and later on read how great they were but they didn't sound any better than a great CDP setup.

Yes, it was fun to observe and oh my, look at the convenience and look, there's my entire library on a screen for me to have access to, .....

All the best,
Nonoise
"I've yet to hear a PC system easily trounce a really good CDP."

Really good is really good no matter how you get there.

I've heard lots of really good players for reference and think my current PC music server based system is really good in comparison as well.

It's hard to say that any really good digital setup would trounce another one I think if both are played in the same system. Where significant differences are heard, I suspect personal preferences would be the main determining factor.
"You probably think about oversampling and that is different. Pretty much every CDP contains PLL."

This is why I don't use or recommend CD players anymore. There are a few that are computer-based that might be okay, depending on the clock quality etc.

"Even asynchronous rate converter based DACs (upsampling DACs) like my Benchmark have some form of PLL to make signal stable enough for upsampling."

I probably have 100 modded DAC1's out in the field. I know it well.

"Usually PLL is inserted somewhere."

Not in the best DACs. The jitter of a free-running clock will always beat a PLL.

Steve N.
Empirical Audio
"I've yet to hear a PC system easily trounce a really good CDP."

If you buy the USB DAC du-jour, you will not get there either. It takes a bit of work, including selection of the best computer platform (not a "music server"), the best playback software and ripper and the lowest jitter USB converter or DAC with USB interface. All of these things are doable and there are plenty of recommendations out there.

It is not cheap though. In order to compete with for example the dCS Vivaldi, you have to spend at least $10K on the computer, USB interface, DAC, power supplies, cables and software.

The good news is that spending $5K on a CDP will not get you the SQ of spending the same $5K on computer and DAC IME.

Steve N.
Empirical Audio
Johnnyb53 and Audioengr,

Thanks for your posts. I found them to be very helpful.
The last few posts reveal the perspective as to which is better--how much sound quality do you get for your money? In my case I've never spent a lot for a CD player, maybe $500 tops. I did have an Audio Alchemy transport feeding a $700 SoundStream DAC (designed with help from Krell), but when I later got a CEC CD player, it sounded better. I've heard excellent CD players demoed at high end stores, and I play SACDs and DVD-As on an Oppo.

But I had the laptop for other uses already, and stacks of CDs ready to rip. All I needed was Audirvana Plus for $49 (now up to $79), properly configured, to easily exceed the musical satisfaction of my various digital disk players. And Audirvana just keeps getting better. When I updated to 1.5.2 (now it's at 1.5.4), I noticed a significant and pleasant improvement in the sound quality and how engaging the music is. Not bad for $49.
Nice relevant posts, one and all. As Steve has pointed out, $5K will buy one a killer PC set up and that's just for the PC part. That was close to the total cost of my previous system. All I had to spend was $1600 on a source unit so a SACD demo unit was the way to go for me and even that was put on plastic.

Having said that, I'm getting killer sound out of my system that thrills me to no end. I even feel smug, for what it's worth. I also feel that I'm with about 90% of folk here who only have so much to spend yet still want great sound. Lots or recent posts are about bang for the buck and diminishing returns with opinions needed as to which way to go.

I chose my way and don't begrudge anyone who's willing to blaze the path to better ways to listen but for now, I'm content to take the road most traveled by as I know it well and how much it costs.

All the best,
Nonoise
Good for you Nonoise. We are all here to learn from the experience of others with the common purpose of enjoying the reproduction of music - however that may be achieved.

Perhap both you and Audioengr could provide a "bill of materials" on respective solutions for cited overall investments?

Speaking personally, thus far, I have not been able to a music server and DAC to better my Esoteric K-series CDP for around the same investment. I've tried what many believe to be the best of breed servers out their i.e. Aurender and SotM music servers coupled to the DAC's from Meitner and Antelope Audio - but somehow whilst extremely capable, they really were not musically engaging. They lost something. I don't know why. I wondered whether the USB interface between the Server and DAC is a point of weakness? I also wondered whether the multiple linear power supplies in the Esoteric provided a superior foundation on which to create musical reproduction? I don't know and thats why I read forums. To hear from others.
Kiwi - usually if its not musically engaging, this is jitter and the USB interface is at fault. Upsampling hardware can also be at fault, both for the clocks and the algorithms.

Here are some systems that are musically engaging and not megabux:

iPod or iPhone and:
Pure I20 docking station - $89.00
Synchro-Mesh reclocker - $599.00
Hynes power supply for Synchro-Mesh - $600
BNC cable - $250
Concero DAC - $600
Total = $2138.00

Sonos - $400
Synchro-Mesh reclocker - $599.00
Hynes power supply for Synchro-Mesh - $600
BNC cable - $250
Concero DAC - $600
Total = $2449

used Mac Mini 2009- ~450 with SSD and DRAM
Mac Mini Hynes power supply - $900
USB cable - $400
Off-Ramp 5/Turboclock/S/PDIF Hynes reg - $2249
Hynes supply for Off-Ramp 5 - $600
BNC cable - $250
Concero DAC - $600
Total = $5449.00

used Mac Mini 2009- ~450 with SSD and DRAM
Mac Mini Hynes power supply - $900
USB cable - $400
Off-Ramp 5/Turboclock/HDMI Hynes reg - $2249
Off-Ramp Hynes power supply - $600
HDMI I2S cable - $250 from partsconnexion.com
Wyred 4 Sound DAC2 - $1500
Total = $6949.00

As for the linear regulators: IF they are not really fast-responding, then not so great for digital. The Hynes supplies that I design are extremely fast responding. If they are based on off-the-shelf three terminal regulators, this does not cut it IME.

Steve N.
Empirical Audio
Kiwi,
Those units you cite are very favorably reviewed and considered to be up there with the best around. That Esoteric must be one nice unit.

Steve,
Thanks for the cost breakdown: I"m sure it took a while to compose.
I'm going to bookmark this thread in case I stumble onto better times as a reference should I get the urge to experiment. :-)

All the best,
Nonoise
Many thanks for the detailed reply Steve. Very useful.
After seeing this post, I went back and re-read the article. My conclusion is the same as when I first read it - these are some interesting topics to discuss, but they reach no conclusions and they have no real recommendations on how to turn their discussion into better sounding systems. Of course, that is mostly a function of the interview format. I would have much rather see these 2 guys (and Steve can join in) collaborate on a white paper that goes into all the currently discussed issues for why computer audio can sound different to different people and on different systems and analyzes the importance and size of those issues and how to fix them.

The whole "gray area" issue(which Hansen eventually admits is not really an issue in a well implemented system), the issue of running multiple processes on the computer, the potential of noise on the grounds, jitter (in all its forms), RFI etc. have been discussed over and over again. A lot of progress has been made on all these fronts with aysnc USB, galvanic isolation, attempts to minimize CPU usage, much better jitter control, etc. Unfortunate, it seems that after several years of figuring out many issues and addressing them, we are now stuck more in a discussion phase than in an significant improvement stage. I understand that a lot of the easy improvements have been made and many of the problems left are hard to analyze and hard to fix. Although this interview is getting a lot of attention, I do not think it moves us down the road in any meaningful way.

Personally, I am happy with my PC based system. I also know it can be improved. For me, one major question is how the issues within the PC compare in importance to new DAC technology, to new digital filters, to the use of DSD formats, etc. For now, I will enjoy my music.
Jitter is bad but I am not convinced that it is nearly as widespread an issue with most modern gear as it was even 10 years ago. YEs, there are always ways to do better, but the question is how much matters to be "musical".

I do it with Squeezebox Touch (no longer available unfortunately but a fantastic bargain while it was for <$300), a Wifi connection to a standard $500 Samsung laptop with 2 external USB drives (Wifi eliminates a wired connection from music server computer to player which greatly benefits from a noise and isolation perspective and enables most any computer to be used as music server witn no ill sonic effects) and two separate mhdt DACs connected via not expensive Audioquest Toslink digital wire, a SS COnstantine ($300 used) and tubed Paradisea ($450 used). Its hard for me to imagine how the results could be much better. The Wifi connection/isolation is key. CHoose your DAC to get the sound you like, and just experiment with lengths if needed to make sure the digital connection to the DAC is functioning well.
Mapman,

Your DAC is not only isolated from computer noise but also from all other nonsense like playback program, file format, computer speed, type of hard disk, amount of RAM etc.

The only thing you could do is to insert reclocker before DAC.
Kijanki,

Yes, I agree and have considered, but hard to justify when things are sounding just the way I want them to already. I may give a re-clocking device a try still sometime in that I am certain I could do even better. It took some time and thought to get to where I am currently. I feel fortunate I have managed to achieve excellent results for relatively small investment so far. The rest of the system makes a difference. My OHM speakers are not hard to get sounding musical to the nth degree in comparison to many. My smaller monitors from Dynaudio and Triangle are a bit more challenging in this regard, but they sound quite "musical" as well and are not my references.
"The whole "gray area" issue(which Hansen eventually admits is not really an issue in a well implemented system), the issue of running multiple processes on the computer, the potential of noise on the grounds, jitter (in all its forms), RFI etc. have been discussed over and over again. A lot of progress has been made on all these fronts with async USB, galvanic isolation, attempts to minimize CPU usage, much better jitter control, etc. Unfortunate, it seems that after several years of figuring out many issues and addressing them, we are now stuck more in a discussion phase than in an significant improvement stage."

There have been some breakthroughs. For instance, it is pretty much accepted that the reason that USB cables make a difference in Async interfaces is common-mode noise. This can be addressed with CM filter or galvanic isolation.

Unfortunately, there are still issues with different SQ from different playback software and different computer hardware when using USB. It makes no sense that improving the power supply for the computer would make a difference, but it does, even with Async USB and galvanic USB isolation. These will remain a mystery for a while I think.

One way to address this is to use networked audio streaming. The downside is that you are trapped into the implementation and playback software. It is also usually limited for sample-rates.

Steve N.
Empirical Audio
Squeezebox and similar devices are essentially specialized computers optimized out of the box to various degrees for playing music. It makes sense to use a computer designed specifically for good sound quality in high end audio applications. General purpose computers can be rigged to work well, but they are general purpose out of the box and not necessarily optimized for high fidelity audio, though I have gotten surprisingly decent results as well in the past by just connecting analog headphone output to the hifi using a decent quality stereo rca to dual rca phono Y connector. The one I have used is also from Audioquest. ITs the easiest way to just get started and establish an initial baseline for computer audio that one can work with from there as needed.
Note that with the analog Y connector approach to connecting computer to hifi, it helps to use a very long wire between computer and hifi. All computers are noisy and you want to keep them as far away from your audio gear as possible. The Audioquest Y connector I've used to good effect is 12' long, for example.