Why DAC64 is transport/cable dependent

Originally posted by merlin
So we get drift and varying sizes of buffers:confused: We don't get these issues with other Dacs, regardless of whether they use PLL at their inputs. Sorry but I don't see anything clever in that, in addition, using a very high precision off board clock with clock link inputs on transport and Dac will offer similar benefits, without the attendant failings.
Click to expand...

It might well be that, although there is some gentle drift using the buffer, this does not sound as bad to human ears as jitter induced by trying to synchronise clocks:confused: For a visual analogy; I'm thinking of how horrid digital artifacts look in a picture, even if quite mild but how the same picture slightly out of focus doesn't seem so offensive (to me at least)
 
Originally posted by merlin
So we get drift and varying sizes of buffers:confused: We don't get these issues with other Dacs, regardless of whether they use PLL at their inputs. Sorry but I don't see anything clever in that, in addition, using a very high precision off board clock with clock link inputs on transport and Dac will offer similar benefits, without the attendant failings.
Click to expand...
Not really merlin. The buffer allows the DAC to average the clock over the buffer period. Even if you have a relatively high amount of jitter in the input, the averaging will more than likely eliminate most of it.

Imagine that the gaps between pulses from the transport were as follows over the buffer period (the numbers are purely arbitrary for ease of example):

100, 98, 102, 95, 97, 105, 103

The average of those gaps is 100 so the replay clock in the DAC would be exactly as in a DAC with no buffer receiving the following pulse gaps:

100, 100, 100, 100, 100, 100, 100

ie, the buffer has totally eliminated the variance in the gaps and effectively eliminated jitter. Now of course a 4s buffer won't be able to eliminate much lower frequency variations but these affect the sound much less anyway.

I don't see what "attendant failings" this method has. Sure, it means you can't get two DAC64s to be synced to each other but, really, the number of people who might want that is vanishingly small. Those who do would more than likely be using professional gear that caters for all that kind of stuff anyway.

Michael.
 
Michael,

my point is that if they are drifting in relation to each other over a period of time, then one is playing the music at the wrong speed or pitch shifting in relation to the source:eek:

That's why I consider it a failing. The only reason that I can think of for this behavior is that, although they are both referencing from the same clock's output, they are coming up with a different average to each other:confused:
 
End of the day boys it's actualy more suspectable to transport and cable selection than a lot of other dacs', maybe this is why the MKII has a clock link ability, that will be utilised when the matching transport is released soon.WM
 
Originally posted by wadia-miester
End of the day boys it's actualy more suspectable to transport and cable selection than a lot of other dacs', maybe this is why the MKII has a clock link ability, that will be utilised when the matching transport is released soon.WM
Click to expand...

I agree there WM. And why would they add a clock link if the buffer actually worked as described:confused:
 
The variation you are refrerring to (did you observe it?) is probably so miniscule that it's undetectable. I know I'd much prefer that than the problems that jitter causes. If you were listening to a studio master tape I very much doubt you'd be able to hear a speed difference of +-0.1%, especially if it happened very slowly which is likely to be the case with the DAC64.

What really matters is whether the DAC64 sounds good and to my ears (and a not insignificant number of other peoples ears) it's the best sound to come from 16/44.1 CD that exists, bar none :)

Michael.
 
Originally posted by merlin
I agree there WM. And why would they add a clock link if the buffer actually worked as described:confused:
Click to expand...
AFAIK the MkII doesn't have clock link ability. Their new transport (not yet released but it will be at the Heathrow show) does but that may be for compatability with other DACs. The 2 BNC inputs on the MkII are to enable 192kHz inputs, not for clock linking.

Michael.
 
Also, I disagree with WM about it being more susceptible to transports/cables than other DACs. IME it's much less susceptible, but susceptible nevertheless.

Michael.
 
Originally posted by michaelab
AFAIK the MkII doesn't have clock link ability. Their new transport (not yet released but it will be at the Heathrow show) does but that may be for compatability with other DACs. The 2 BNC inputs on the MkII are to enable 192kHz inputs, not for clock linking.

Michael.
Click to expand...

I know the 2 outboard BNC's are for dual 96khz inputs, but Haven't then got the internals there, just no hole in the side yet?, or have I've been at the cider too much:D
 
I think you've been at the cider too much Tone :D No "spare hole" or anything. Of course they could always reprogram the internals (as it's all done in FPGAs) but AFAIK they aren't about to do that. Some questions for the Chord rep for the Heathrow show I feel :D

Michael.
 
Originally posted by michaelab

What really matters is whether the DAC64 sounds good and to my ears (and a not insignificant number of other peoples ears) it's the best sound to come from 16/44.1 CD that exists, bar none :)

Michael.
Click to expand...

Sorry Michal...but when I did a back to back with the Wadia 301 the Wadia won by a country mile (and the 861 is a further league ahead). And of those at Bottleneck's hearing it against Lee's CAL DAC, the CAL had as many adherents as did the Chord. You may like it, and I'm not saying it's bad, but there's plenty of competition!

I'd also suggest you have another look at the DAC64 documentation. IIRC (saw a tech report from them somewhere) they made a big point about using the buffer for enabling very long FIR filters to clean up the anti-aliasing filter - so the buffer's not all/primarily for jitter elimination.

And to the other Mike (Merlin).
Your 2-DAC experiment just happened to be the perfect experiment to measure "long term" drift (just like Lissajous figures on a 'scope - perfect for picking up variations at the sub-Hz level between two signals even in the 100s of MHz). FIFOs+PLLs will have very low frequency drift, and the lower the frequency the wider the peak difference between the two signals may get as the trade-off for keeping the "pitch" differences smaller is that they are held for very much longer times. The point though is that these "pitch" differences are far less audible in normal usage than those in the audio band produced by other DACs without such a long buffer - which you'd find very difficult to pick out by that experiment.
 
Thanks for that Graham old chap:)

I must admit, I think I had fallen into the current forum trend for theorising and questioning on an objective level:(

It's the subjective impression that carries more weight when listening to music I suspect, and will try to stick with those findings in future.

I like the DAC64, bouncyand lively to these ears, and I beleive that you need to spend more to better it.
 
Graham - I was only being provocative :) Of course you will never get even agreement that any bit of hifi kit is "the best, bar none".

For the price I paid I'm extremely happy with it and, as Merlin says, I think you'd need to pay quite a lot more than it's RRP to get better, if indeed, to my ears, there is better ;)

I know two people here in Portugal who have auditioned pretty much every digital source they could lay their hands on upto £6000 and they both thought the DAC64 blew everything else out of the water. An editor of an online hifi mag here has also just got one and prefers it to anything (dCS, Wadia, Perp Tech you name it) he's heard. All that doesn't prove anything though - it's just personal preferences.

One thing I would love to hear though is the CAL Sigma Tube DAC for comparison.

Michael.
 

Latest posts

Back
Top