Jump to content
IGNORED

Reclockers


Recommended Posts

AFAIK, there are only a few products that do this: Empirical Audio's Pace Car ($1200+), Apogee Electronics Big Ben (~$1100), Antelope Audio Isochrone DA ($1100), and Grimm Audio CC1 ($1700).

 

Am I missing any?

 

The Pace Car is probably the most flexible, with quite a variety of configurable inputs. But with options (at least to fit my needs), it's also the most expensive.

 

The Grimm does XLR; the Antelope adds RCA S/PDIF; the Apogee does all that and Toslink too (I think -- finding out info on the Big Ben is a little tough).

 

I've seen some unofficial reviews of the Antelope, "official" reviews of the Empirical, but nothing on the Grimm or the Apogee.

 

Anyone have anything to report along those lines? Positive or negative?

 

 

Link to comment
Share on other sites

The Brainstorm DCD-8 is also a reclocker w/ firewire, toslink, rca spdif and aes.

 

In reading unofficial reviews, the pacecar is supposed to offer a more realistic ie. "you're there" sound than the apogee.

 

In terms of a clock, the Antelope has been reported over at gearslutz to be better than the apogee.

 

Link to comment
Share on other sites

Hi Scot - I'm interested in reclockers as well. They have great potential, especially when a noisy source is used. Do you know if the Isochrone DA is really reclocking? I heard over the weekend that when it's used to pretty much pass an AES signal that it isn't really reclocking. This could be totally off so I'd like to get to the bottom of it soon.

 

I think we will see many more reclocking devices in 2010.

 

Founder of Audiophile Style

Announcing The Audiophile Style Podcast

Link to comment
Share on other sites

Ultimately reclocking in the DAC is technically a far better solution than any external reclocker. No matter how clean the regenerated clock produced by the reclocker, retransmitting the cleaned clock via SPDIF or AES/EBU reintroduces a jitter component to the clock and you end up with a dirty clock once again. You might end up with a cleaner clock than your source outputs but you still have have a jittered clock.

 

Link to comment
Share on other sites

Here is all the write-up gives up: "Additional Thru outputs transmit a rebuffered version of the original signal, allowing converters and other audio sources to be used normally, in addition to their new role as master clock."

(http://www.antelopeaudio.com/en/products_iso_da.html)

 

I guess the question is what does "rebuffered" mean? Is it a FIFO?

 

Link to comment
Share on other sites

I agree, you will still have a SPDIF signal, but I've heard of people using two reclockers so that the second one doesn't have to work as hard. Similarly, if the DAC is reclocking as well, then it won't work as hard to attenuate the jitter.

 

Link to comment
Share on other sites

On a simple application of Ohm's law, or a filter analogy, some percent of IN will always be passed to OUT. Therefore, I agree with the perspective that, all other things being equal, two reclockers working in series should provide greater benefit than just one.

 

On this principle, I use power conditioners (regenerators) in series to good effect, and good ones, too (Accuphase models). Less distortion in = less out.

 

Link to comment
Share on other sites

The problem is that in retransmitting the SPDIF you reintroduce jitter into the data, and that presents a limit to how far you can reduce jitter level using external reclockers regardless of how many you chain.

 

Link to comment
Share on other sites

Maybe it's just me, but I'm confused why someone would pay nearly $2k ($700-800 for Lynx card, and $1100 to $1200 for a reclocker) to optimize a legacy DAC - excepting the incomparables like the Alpha and its peers, if it has any - when the Weiss DAC2, or the Ayre QB-9, etc. are available as new slightly above that price.

 

Okay, let's put that opinion aside, and talk about what might be on offer from reclockers.

 

I can easily agree with Chris, when he says that we will see more of them in 2010. I'd bet my most expensive Firewire cable on it. :)

 

OTOH, we have an apparent contraindication (about which more later) of reclockers vis-a-vis asynchronous interfaces - which allow initial clocking (NOT reclocking) of the signal with the fixed oscillator in the DAC. This is the basis of my confusion noted above.

 

And then we have something called 'noise isolation', apparently (to the layperson) provided simply(?) by inserting another device into the playback chain.

 

This was referred to in the Antelope DA review thread, and again above my Chris - re his 'noisy signal' comment.

 

Perhaps we should delve into what 'noises' are being isolated - and more importantly, how? What is really being 'isolated' if/when there are still electrical/physical connections between the source, and the reclocker, and the DAC? Is it simply the noisy clock signal duplexed into the data signal, OR something else?

 

If this is truly effective, maybe inserting the cheapest reclocker into the playback chain would improve even an async interface?? I'll bet against that one, but it would speak to the isolation point perhaps.

 

Just curious,

clay

 

 

 

Link to comment
Share on other sites

i agree with most of your thoughts but... a lot of people have very good legacy dacs and they do not want to change it, just because there is just another - maybe even better - interface out there. as already discussed in many, many threads, its not just about jitter or how good an interface is. it is even unclear how much jitter we can actually hear!

 

so in my opinion it makes much sense to NOT change a good legacy dac - the core algorithm for changing digital to analog signal - but to change the computer interface itself. the computer interfaces are changing relatively quickly anyway. usb 3.0 is just coming out. will you change your dac every time there is a better way of transporting the digital signal? if we accept this, it makes much sense to have a very good async interface (afi1, gonrdon´s new async usb interace or dcs u-clock) synced to the dac. even if you are thinking about buying a new dac today, the choices are somewhat limited if you just take the async interface dacs into account. and would you believe an ayre would better Mani´s Weiss AFI1->Pacific Microsonics Model Two system?

 

i would also like to learn more about how to electronically isolate the computer from the dac - if necessary.

 

 

 

Link to comment
Share on other sites

 

well said,

 

"so in my opinion it makes much sense to NOT change a good legacy dac"

 

thanks for sharing your opinion.

 

"would you believe an ayre would better Mani´s Weiss AFI1->Pacific Microsonics Model Two system?"

 

No, which is why I noted an exception for the 'incomparable' DACs, in my post. It's not difficult to understand spending $2k on optimizing an Alpha, Model Two, etc. and others I'm sure I'm forgetting. I was referring to the vast majority of legacy DACs which are not in that price/quality range.

 

"will you change your dac every time there is a better way of transporting the digital signal? "

 

Probably not, but I've no need to, I use Firewire primarily, but also have an Async USB DAC. Firewire 3200 will be backwards compatible (even offering 3200 bandwidth as I understand it) over existing Firewire 800 cables/inputs, and USB 3.0 will offer (next to) nothing for the audiophile, but will of course, also be downwards compatible.

 

"if you are thinking about buying a new dac today, the choices are somewhat limited if you just take the async interface dacs into account"

 

That's a very relevant point, especially for those that want 'traditional' audiophile brands, as opposed to Weiss, Metric Halo, Lavry, etc.

 

Ironically, the only pro audio brand that has 'become' an accepted audiophile brand is one I would never purchase/recommend due to its 'house sound', i.e. Benchmark. Give me a MH, Weiss, Lavry, Apogee any day of the week.

 

cheers,

clay

 

 

 

 

Link to comment
Share on other sites

PJ

 

Why is the does the re-clocker to DAC (via S/PDIF or AES) introduce more jitter than when re-clocking occurs within the DAC? And how is this level of jitter impacted by the use of other interfaces (e.g., I2S) between the reclocker / DAC?

 

Regards

APS

 

Link to comment
Share on other sites

Jitter only becomes an issue at the point of conversion from Digital to Analog (or vice versa if you are talking about an ADC). So the goal is get a clean clock and data feed where it counts most - the DAC chip. This is best achieved with a low jitter clock source located as close as possible to the DAC chip.

 

The SPDIF interface whether, COAX, TOSLINK or AES, introduces jitter due to a whole range of factors - including but not limited to driver and receiver implementation, impedance mismatches, and intrinsic jitter in the SPDIF receiver. Even the best commercially available SPDiF receivers claim to have around 50ps intrinsic jitter and only filter incoming jitter above 1khz. This all sets a limit to how far you can reduce jitter in a signal sent across any type of spdif link, and means regardless of how clean your external reclocker the level of jitter will always be dependant on the SPDIF link.

 

There is a lot of detail on the problems on clock transmission via SPDIF in Colin Dunn and Malcolm Hawksford's 1992 AES paper ( http://www.scalatech.co.uk/papers/aes93.pdf )

 

In contrast precision VCXO's and XO's have an jitter specifications in the low single digit ps range, dependant on proper attention to power supply, grounding, etc, etc. Feeding a DAC chip directly from a precision clock source like this potentially gives you jitter levels an order of magnitude lower than the intrinsic jitter of a receiver like the DIR9001. Unfortunately it's difficult to extract the theoretical performance from even these low jitter VXCO's. The necessity of syncing with the incoming clock means that regardless of the technique used some jitter will be transferred to the clean clock. How much depends on how effective the filtering of your PLL is.

 

I2S is not immune to jitter either and requires careful implementation and reclocking naer the DAC chip for best performance.

 

cheers

Paul

 

 

Link to comment
Share on other sites

I hate to get off topic here but I can resist mentioning this. I once told Ashley James from AVi that I was reading Malcolm Hawksford's work to educate myself on a specific topic in digital audio. (Note: This was when Ashley and I had a large disagreement about DACs.) Ashley's response to me was this ->

 

"Malcolm's problem is that he is an enthusiastic amateur who has never really had any ideas of his come to a workable conclusion."

 

I disagree with Ashley on this one as well.

 

Founder of Audiophile Style

Announcing The Audiophile Style Podcast

Link to comment
Share on other sites

@Thomas

 

In the case of ASRC you might have a point - jitter is embedded in the data as part of the upsampling process if certain (apparently reliable) sources are to be believed. But at that point you no longer have an issue with clock jitter - instead you have altered the audio data transmitted to the DAC.

 

see: http://www.diyaudio.com/forums/digital-source/28814-asynchronous-sample-rate-conversion-2.html#post342727

 

For clock related jitter the point where it makes a difference is at conversion to analog. The only thing that matters is the clock used to move data into the dac and trigger conversion. How you achieve a clean clock at that point is a matter for debate.

 

cheers

Paul

 

Link to comment
Share on other sites

  • 1 year later...

I have used various ones for years and I have not found a 'best' as the sonic outcomes actually depends on the relocking circuits inside the dac. Slaving to a clock is not really a solution either, as the clock reception circuit and cabling also come into play.

 

Of the units you mentioned, I am not convinced by the Empirical Audio stuff. The literature isn't transparent, Steve uses components like superclocks that I have not found to be particularly 'great', and to convince me to spend $1200 + expensive optiopns, I need to know what they all do and I will not rely on 6 Moons or similar reviews.

 

The Apogee is a flexible unit. You can improve the internal 5V power supply very easily, and it sounds as good with many dacs as other relockers I have tried. It is particularly useful if you have a computer single wire output and wish to use 2 wires for dacs such as the dCS. However, I don't know that it sounds any better (just different) than using the 'old pro' dCS972 upsampler/relocker. Clock signals from its output into various dacs are still affected by cabling and the clock output connectors are actually 50R BNCs.

 

The Antelope Gold dac, despite the clock, has not recieved good reviews from HiFi News or Stereophile, and I have heard demos with their Atomic clcok feeding an Olive that sounded very ordinary. Otherwise, I have no experience.

 

Grimm - no idea.

 

To sum up, my experience is that the receiving circuit of a dac is very important and that a relocker is mandatory for PC spdif/aes outputs.

 

As for slaving, there is a Sound on Sound article which concluded that external clocks are there for synchronisation purposes and not sonic purposes. I have not had their exposure but I broadly agree with this.

 

 

 

fmak

Link to comment
Share on other sites

'cause of lack of deamnd?

 

It is actually not easy to have a clock/reclocker that outputs nice symmetrical square waves as they should do. Most don't have tight mark-space ratios and most have overshoots and non zero crossovers. The flip flops they use are so poor that one output can be different (and sound different) from the next. Some actually contain kinks due to reflections in the output wavforms.

 

The only dac/relocker combo with published output graphs I know of was the old dCS pro units, where waveforms are ideal compared to many others. In contrast, the dCS consumer version of the 972 is poor, like the rest of the field.

 

fmak

Link to comment
Share on other sites

There are not many dacs with clock inputs to hook the external clocks up to. In fact the new models coming out this year avoid clock inputs.

The Marantz SA-11S2 player is the only unit I have owned with a clock input. Made a 50 ohm terminator to keep the open input from degrading the internal clock.

Clocks can be better. The Jet clock technology is well accepted and is used on units with a wide spread of price points.

 

George

 

 

2012 Mac Mini, i5 - 2.5 GHz, 16 GB RAM. SSD,  PM/PV software, Focusrite Clarett 4Pre 4 channel interface. Daysequerra M4.0X Broadcast monitor., My_Ref Evolution rev a , Klipsch La Scala II, Blue Sky Sub 12

Clarett used as ADC for vinyl rips.

Corning Optical Thunderbolt cable used to connect computer to 4Pre. Dac fed by iFi iPower and Noise Trapper isolation transformer. 

Link to comment
Share on other sites

from M2tech. It can drive their EVO USB2-SPDIF/AES/I2S interface, which could be connected their new EVO DAC using I2S connection. Very affordable and elegant, minimal jitter solution for everybody using computer audio as source.

 

Link to comment
Share on other sites

It looks good. The numbers calculate to a jitter of around 200 fS (better than the MSB Femtosecond clock) if they are real which they may be.

 

Not fun with a mixed sample rate playlist however.

 

Demian Martin

auraliti http://www.auraliti.com

Constellation Audio http://www.constellationaudio.com

NuForce http://www.nuforce.com

Monster Cable http://www.monstercable.com

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share



×
×
  • Create New...