An article about why bits are NOT bits

gobble

Well-Known Member
Joined
Sep 23, 2008
Messages
5,348
Points
113
Location
Bangalore
To quote - This article will attempt to dispel the popular notion that "bits is bits." This belief holds that if the ones and zeros in a digital audio system are the same, the sound will be the same: Stereophile: The Jitter Game

Not a proclamation from me, just an interesting thread to follow :)

Regards
 
Jitter results in frequency domain errors, not *just* time domain.

Though it is a clock issue, the overall conversion is affected because the DAC is using the clock to reconstruct the original signal.

The ear can reliably detect timing errors up to about 0.5ms (equivalent to the click you hear when a stylus hits a small speck of dust on a record). This is 500,000 ns, well beyond jitter figures exhibited by the worst systems.

Jitter is completely (read: not maybe) eliminated by clocking the transport and DAC with the same master clock. This requires both to have a word clock input, not easy or cheap to implement. Which is possibly why very few of the big-dollar companies offer it in their mainstream products, reserving it for their highest-end products.

For those with pro audio cards as transport, the SPDIF input can be used to feed a clock signal (with some skullduggery) to the transmitter chip.

Well the jitter being created by a decent audiophile budget CDP would be less than 100 nano secs right? So is it the frequency domain errors that the ear picks up more than the time domain?

Regards
 
Gobble, this is a very old article. I think DAC chips's capabilities and DAC technology has moved forward quite a lot since that time. As I explained in Bangalore, a Voltage Controlled Oscillator can vary the speed of a clock as required after comparison to an external reference. That way jitter can be controlled quite well.

Cheers
 
Gobble, this is a very old article. I think DAC chips's capabilities and DAC technology has moved forward quite a lot since that time. As I explained in Bangalore, a Voltage Controlled Oscillator can vary the speed of a clock as required after comparison to an external reference. That way jitter can be controlled quite well.

Cheers

Ok so a VCO is different from the master clock of yesteryears? I didn't realize these were two different things. Do you know any products that use VCO?

Regards
 
Ok so a VCO is different from the master clock of yesteryears? I didn't realize these were two different things. Do you know any products that use VCO?

Regards

hi gobble,

i am extremely pleased and happy to read about such advances and also happy in the knowledge that today's extreme audiophiles will be moving towards digital storage, retrieval and playback-

it has this import for me - i will, in the near future, be able to buy (second-hand) those exotic transports/dac/power beasts with 35 clocks doing masterful duty for a pittance:yahoo::yahoo::yahoo::yahoo::yahoo:
 
Ok so a VCO is different from the master clock of yesteryears? I didn't realize these were two different things. Do you know any products that use VCO?

Regards

yesteryears used grandfather clocks with big pendulums for adding weight and resonance control. nowadays as has been pointed out, all are digital clocks.

there are also cuckoo clocks. with these one tends to hear a significant difference that normal people can't.

regards
 
Great! I want a DAC with a pendulum clock that I can hang from the wall. One that needs keying with a chaavi so that it is green and does not consume power :D

Cheers
 
Jitter is completely (read: not maybe) eliminated by clocking the transport and DAC with the same master clock. This requires both to have a word clock input, not easy or cheap to implement. Which is possibly why very few of the big-dollar companies offer it in their mainstream products, reserving it for their highest-end products.


as with the dcs elgar three box system. supposedly claimed the world's best.
 
I came across an excellent white paper on external DACs in NAIM's website. Please read the article at http://naim-audio.com/download/Naim_DAC_White_Paper_Aug_2009.pdf. Though they have covered a number of issues that frighten most of us from switching to digital storage, and playing through a DAC, the following specifically refers to Jitter.

Naim White Paper said:
In the Naim DAC the master clock is not recovered from the S/PDIF signal as usual. Instead the audio data is read from S/PDIF, stored in solid-state memory and then clocked back out to the DAC chips using a fixed-frequency local master clock. This eliminates jitter caused by S/PDIF. In essence the memory, master clock and DAC structure behaves in a similar manner to the CD, master clock and DAC structure of a CD player.

If Naim is to be believed and such technology is possible, we can all seriously look at storing and playing good quality music from hard disk.

Cheers
 
What you said is called reclocking and any DAC worth its salt has been doing this since the last ten years. There's nothing new in reclocking and its a known fact that reclocking reduces jitter but it doesn't completely eliminate it.

What can actually reduce jitter is to use I2S where the clock doesn't need to be recovered from the data stream. However I2S can only be used for very short distances.

If SPDIF must be used, the best option still is to get a low jitter source - the best ones are those that use asynchronous USB. Also some sound cards allow an external clock to be input into them when using their SPDIF out. If one has access to a high quality external clock, its best to slave both the sound card and the DAC to this.
 
If SPDIF must be used, the best option still is to get a low jitter source - the best ones are those that use asynchronous USB. Also some sound cards allow an external clock to be input into them when using their SPDIF out. If one has access to a high quality external clock, its best to slave both the sound card and the DAC to this.

Since I2S are non-viable at other other than extremely high end systems, Naim has stuck to S/PDIF and used their own algorithm to remove/reduce jitter. Taking a step away from asynchronous sample rate conversion or the standard VCO, Naim uses 10 fixed frequencies. The DSP senses the incoming signal and chooses one of the ten frequencies to match the clock speed of the source. This way the use of a 'unreliable' voltage to control the oscillator frequency is done away with completely.

Though I am aware that these technologies have been around for some time, I am impressed with the way Naim has taken standard technologies and implemented them to deliver a near perfect signal. Whether is it jitter removal, filters, or over-sampling, Naim seems to have developed, tested and proven to themselves algorithms that can do the DAC conversion extremely well. And in my mind, that bodes well for the future of music.

Cheers
 
Last edited:
A beautiful, well-constructed speaker with class-leading soundstage, imaging and bass that is fast, deep, and precise.
Back
Top