Upgrading clocks
randy robinson -- Sun, 12/27/2009 - 11:23
As jitter is reduced, is there a corresponding improvement in sound or is improvement only achieved when jitter is eliminated?
There are clock upgrades available claiming extremely low jitter. I've seen these available for both CDP and DACs. In which unit does the upgraded clock have the most impact?
All content, design, and layout are Copyright © 1999 - 2011 NextScreen. All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Jitter can never be completely eliminated. Reducing jitter improves the sound.
Regarding your second question, I would expect that an outboard clock would have a greater sonic impact on a DAC compared with a CD player for the simple reason that a DAC requires a digital interface to drive it, which adds jitter. That is true unless the DAC has an internal master clock to which the source must lock. This also assumes that the DAC and CD player have equally good clocking circuits.
What DAC or CD player are you considering using a clock with?
Robert, Does the Berkeley DAC have an Internal Master clock to which the source must lock? Will the Berkeley benefit from an external clock and if so how much? There are clocks out there equal or more the price of the DAC itself. Did you try the dCS clock with anything else other than dCS players/DAC?
Robert,
Have you seen the Antelope "Zodiac" DAC? This puppy retails for $1500 and has an "oven stabilized crystal" clock (like an FM or TV transmitter) to reduce jitter. Antelope is a professional recording products company (although the Zodiac looks very audiophile high-end) who specializes in external rubidium clocks (tres cher). I have heard this DAC and it sounds stupendous, much better than its competition such as the Benchmark or the Bryston. You should check it out.
George Graves
Recently I asked this forum for their thoughts on a top tier SACD/CD player for under $10k and, after much deliberation, I eventually purchased the Esoteric X-05 which I have found to be a substantial improvement over my 8 year old Sony SCD-777ES. The X-05 was $6k...so...now I'm thinking: would adding the Esoteric G-03X Master Clock ($4K) result in a significantly better player?
have added a grimmaudio clock with a lessloss power cable to my D O3 P 03 :result VERY substantial bettering of my system
Robert, sorry for the delay. I was blocked because of an accidental duplicate post. I have a modified Music Hall CD-25.2 and a Bel Canto DAC2. My question is about replacing internal clocks.
JLeeMD - IMO, yes. Adding an Esoteric G-0s to my UX-1 gave it a more 3-dimensional sound and a bit more clarity. Most noticeable when you turn it OFF.
The comment about the clock's effect being most noticable when you turn it off parallels my experience. Once you get used to the sound of a digital front end with a precision clock, it's difficult to listen without the clock.
Are there noticeable differences to be heard by upgrading internal clocks? The upgrades I've seen are around $500 yet the clocks have specs of better than +/- 0.2 ppm.
I don't have any experience with aftermarket modifications to internal clocks. I suspect that a skilled designer could render a large improvement to a mass-market player by upgrading the clock.
Can anyone comment on the sound or performance of using an Esoteric G-03X clock versus Esoteric's much more expensive G-0Rb clock?
If you're considering a rubidium clock for an audio system, you should buy it.
I have not heard the G-03X, only the G-0Rb.
I have also not heard the G-03x, but common sense tells me that a clock that is magnitudes of accuracy better than another is probably also sonically superior.
+/- 0.1ppm - G-03x vs +/- 0.00005ppm.
My advice is that if your gonna be a bear, be a GRIZZLY and get the rubidium, but only if your system/finances can support it. Otherwise, there is a good chance that you really might not hear much of a difference, if any at all, using the G-03x in your system. This is probably a moot point for most systems, but on more resolving systems looking to wring the last possible bit of performance out, the rubidium clock could be considered almost the equivalent of a componenet upgrade.
This is the second or third post in this thread that has confused frequency stability with jitter. They are NOTHING like the same thing. The difference is as stark as the difference between speed and acceleration as there is a dimensional change between the two quantities.
For sure you can't hear the frequency difference between a clock that is off by 50 ppm, let alone fractions of a ppb. And besides, frequency stability this fine is simply not important anyhow. For example, when you're listening to music, are you saying to yourself "hey the middle C on that piano is off by 0.0000001 Hz !!!! And that trumpet is off by twice as much !!!".
Jitter difference between the two MAY be a different story.
How does one tell the difference between jitter and frequency stability?
Frequency stability means simply how the frequency of the oscillator changes over time. You use a frequency counter to make the measurements. At the precision levels of a CD player or DAC, such a device with a simple ovenized quartz crystal will suffice. The state of the art for quartz oscillators has advanced greatly over the past few years. You can get a 200 ppb ovenized oscillator that has 10 ps of jitter at I think about 10 kHz can be bought for under $100 in quantity That is, the player itself may now have a crystal that is as good as that of the test equipment for very little money.
Jitter is defined as the difference in significant instants of a signal from its ideal. It is also not a single number as people seem to think - it is expressed as a frequency spectrum and the number given is merely one point that lies on the spectrum. For a clock signal such as the discussion in this thread, all jitter means is the instant in time that the active edge of the clock crosses from one value to the other as compared to the ideal time for such transition.
Jitter is caused by all sorts of things:
ground bounce
power supply fluctuations (good DACs will use highly bypassed regulated supplies in their digital sections)
crosstalk from other signals in the equipment and indeed devices
impedance mismatches at connectors
and on and on and on
These factors are additive - that means the designer of a high quality clocking system must go after ALL of these in a very aggressive manner.
I just noticed that I didn't really circle back and answer the original question - how does one tell the difference.
Firstly, humans can't tell pitch (a word used in describing in music that simply means frequency) to the levels being discussed here. Anyone who thinks his $50,000 atomic clock at 0.000001 ppb pitch accuracy confers any advantage pitch-wise over a $0.10 quartz 50 ppm crystal is simply fooling him/herself. But the builders of the atomic clocks are grateful for the cash. A blind test would very quickly clear this up (oh, now I stepped in it).
As to what it sounds like to have less or more jitter, I don't know as I've never done a test. You could make a test jig by amplitude modulating the control voltage of a VCXO in the audio frequency range that clocks out the data to the DAC chips so it's not like it's particularly hard to do. It would be an interesting test to do for sure. Some have done reviews of high precision clocks (RH is one) - you could read what they wrote.
Hi, I added the G-03x to my X-01/D2 and there was a noticable improvement. We did some on/off tests for comparison. The clock settings on the 03X can be changed and this can effect the sound for better/worse depending on the type of CD, e.g. SACD, CD, HDCD. I haven't heard the Rb clock but the cost (about $16.5K) was a lot more than I was willing to spend. The -03X was a dealer demo at a real good price. I did notice that Esoteric will upgrade the 03X to Rubidium for about $4k. I might do this in the future but for the time being I'm real happy with the 03X. The clock in the D2 is rated at 3ppm so at .1ppm, the 03X is a nice improvement.
There are precious few reviews on G-03x...could you please elaborate on how the sound improved? I'm eager to hear your comments.
The Esoteric D-03 owners manual has a published jitter spec of 0.5ppm, with the qualifying statement "in mint condition". The G-03x has a published jitter spec of 0.1ppm. Robert Harley's article on the G-ORb stated that this made a definite improvement in the sound of the D-03/P-03 combination. The G-ORb has a jitter spec many times lower, but is price prohibitive. Is the G-03x jitter spec lower by enough for that to make a meaningful improvement to the sound of the D-03/P-03 combination?
John -
That's the first I've heard of Esoteric upgrading the 03x to Rubidium for $4K. Its got me wondering what sort of an "upgrade" it could be without changing cases, etc. Depending upon what they would do, it might be a good deal, as IMO the difference between the 03x and any of the Rubidum clocks is probably greater and more noticeable than the difference you achieved adding it to your D2. Used Esoteric Rubidum clocks do turn up on Audiogon from time to time in the $6500-7500. range. Just a thought....