Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Adats and clocks - are external clocks snake oil ? (Read 3369 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Adats and clocks - are external clocks snake oil ?

If you have four adats coming off say an rme raydat card conventional wisdom says the adats should be set to slave  and us the rme as a master.  Others dispute this. Anyone hear have an op ion on this - are external clocks snake oil ?

Adats and clocks - are external clocks snake oil ?

Reply #1 –
Quote
are external clocks snake oil?
For sound quality, snake oil.

Pros use a master clock for a couple of reasons - First, you can synchronize all of your equipment down to the exact sample.    For example, if for some reason you are recording with more than one interface at the same time and they each have their own clock, they will drift out-of-sync over some period of time.

Second, a master clock is usually very accurate.  Sometimes people record a track with one device (such as a USB mic) and playback/monitor on a different device (such as a regular soundcard).  They play-back the 1st track while recording a 2nd track.  If the playback-time is not "exact" because the clock is different, when the two tracks are later mixed together,  they will drift out of sync by the end of the song.  (This problem is unusual, but it sometimes happens with consumer soundcards).  Or if you are doing video work, the audio & video might drift apart by the end of the movie.  This is a bit more critical since a movie is longer than a song and a small time difference can cause a big problem.   

Sometimes a cheap soundcard can be so bad that the pitch is off and a musician can't play in-tune with the backing track.

Adats and clocks - are external clocks snake oil ?

Reply #2 –
Quote from: augustine on
If you have four adats coming off say an rme raydat card conventional wisdom says the adats should be set to slave  and us the rme as a master.  Others dispute this. Anyone hear have an op ion on this - are external clocks snake oil ?

You need a common clock for all devices in such a system. The most natural choice in the setup you describe is to have the raydat card source the clock, because it has connections to all other devices.

That's not a matter of snake oil, it is a matter of a correct system setup. Not having all devices in sync results in clicks.

Of course, it would also be possible to have a dedicated clock source feeding all devices, including the raydat. That requires more wiring and hence is interesting for example when the setup needs to be integrated into a larger system that has a "house clock".

Could you elaborate what the arguments are of those who "dispute this"? Do they offer any?

Adats and clocks - are external clocks snake oil ?

Reply #3 –
Most claims about sound improvement using expensive master clocks are purely anecdotal and measurements usually show the opposite - some degradation.

So yes, if someone tells you about improvements in clarity and imaging because of their $3000 rubidium clock it is snake oil.
"I hear it when I see it."

Adats and clocks - are external clocks snake oil ?

Reply #4 –
Quote from: augustine on
... are external clocks snake oil ?

Quote
The obvious conclusion is that in simple digital audio setups a master clock is usually unnecessary, although it remains critical that multiple digital devices are clocked sensibly. In more elaborate digital audio systems, a master clock can make the task of slaving multiple units much easier and neater, and allow the system to operate more reliably. In systems where digital audio is being used in synchronism with video, an appropriate master clock is absolutely essential. But in any of these cases, the use of a master clock will not improve the audio quality achieved by the converters in any technical sense — and the most expensive clocks fare no better in this regard than the least expensive. The only relevant criteria for purchase is whether the clock provides the facilities, inputs and outputs required, and is designed sufficiently well to conform with AES11 Grade 1 standard.
~ Does Your Studio Need A Digital Master Clock?
Quis custodiet ipsos custodes?  ;~)

Adats and clocks - are external clocks snake oil ?

Reply #5 –
Quote from: 2tec on
Quote from: augustine on
... are external clocks snake oil ?

Quote
The obvious conclusion is that in simple digital audio setups a master clock is usually unnecessary, although it remains critical that multiple digital devices are clocked sensibly. In more elaborate digital audio systems, a master clock can make the task of slaving multiple units much easier and neater, and allow the system to operate more reliably. In systems where digital audio is being used in synchronism with video, an appropriate master clock is absolutely essential. But in any of these cases, the use of a master clock will not improve the audio quality achieved by the converters in any technical sense — and the most expensive clocks fare no better in this regard than the least expensive. The only relevant criteria for purchase is whether the clock provides the facilities, inputs and outputs required, and is designed sufficiently well to conform with AES11 Grade 1 standard.
~ Does Your Studio Need A Digital Master Clock?



Hi yes this is one of the articles I've been reading. My query is mainly to do with systems of around 32 channels or more and whether therr is an audible advantage in using 1 32 channel madi converter over say, 4 "8 channel" adat converters (with internal or external clocks) . This article seems to indicate it's a boon to use a clock. However I was reading another that I can't find now unfortunately which seemed to imply that even on large multichannel systems you are better off leaving the devices (adat etc) to their own internal clocks rather than using an external clock. Others working in dsp on forums agree but what are the facts ?

https://www.gearslutz.com/board/high-end/99...swears-why.html

Please take note of the guys response further down the page where cites two engineering articles. Astonishingly he was banned from gearslutz.. Bear in mind I am talking about 4 seperate "8 channel" adat units not a single one.  He seems to be saying the cables etc of the seperate units etc to external clock make more inacuracies and you'd be better off relying on the internal clocks of the seperate units. This brings us to the next question, would you be better off with a madi box which will do 32 channels in one box to reduce jitter ? Others say jitter is generally inaudible anyway and 4 8 channel adat units is a lot cheaper than 1 32 channel madi converter (and card)....

Adats and clocks - are external clocks snake oil ?

Reply #6 –
It's not quite clear to me if you're talking about a recording or playback situation. IIRC you're into ambisonics, so I guess it's the latter. In that case the RME will be clock master. With short adat cable runs I see little benefit in external clocking.
FWIW in the 90's Neve recommended unmodulated (digital silence) AES/EBU connections to sync elements of their digital Capricorn console. What external sync options does the adat offer?

Adats and clocks - are external clocks snake oil ?

Reply #7 –
Quote from: augustine on
My query is mainly to do with systems of around 32 channels or more and whether therr is an audible advantage in using 1 32 channel madi converter over say, 4 "8 channel" adat converters (with internal or external clocks) .

If you have separate converter boxes in a common system, all contributing to a single "project", you need them to be synchronized, otherwise you get clock drift between the boxes, and clicks are the result. There are several different ways how this can be done, but letting every box run on its own internal clock is not an option, unless you employ sampling rate converters. If you want the internal clock of a converter box to drive the sampling (for whatever reason), you will have to have all channels in a single box.

In other words, there is an audible advantage of using a single converter box with 32 channels over using 4 boxes of 8 channels each, if you are clocking them internally. The difference will be obvious: clicks vs. no clicks.

If you are clocking the boxes externally, i.e. you synchronize them properly, either way you shouldn't get clicks, and you basically experience the quality level of the respective boxes. The better one will be better, albeit not necessarily audibly better. Reasons other than jitter will dominate here.

Quote
However I was reading another that I can't find now unfortunately which seemed to imply that even on large multichannel systems you are better off leaving the devices (adat etc) to their own internal clocks rather than using an external clock. Others working in dsp on forums agree but what are the facts ?

You need to solve the synchronization problem first, before you worry about jitter. Having multiple converter boxes each running on their internal clock is not a valid synchronization model and will give you grief. How any alleged jitter advantage would even matter in such a situation is beyond me.

Quote
This brings us to the next question, would you be better off with a madi box which will do 32 channels in one box to reduce jitter ? Others say jitter is generally inaudible anyway and 4 8 channel adat units is a lot cheaper than 1 32 channel madi converter (and card)....

The jitter performance will be a lot more dependent on the quality of the devices, not so much the cabling. Hence there are no general tips to be had here. In your setup, having 4 ADAT boxes connected to a common RME card is simple enough. A reasonable ADAT converter box ought to be able to regenerate a sufficiently good converter clock from an incoming ADAT signal. If it can't, it is crap. Moreover, the optical ADAT connection allows galvanic separation between the PC and the audio hardware (provided there is no additional connection necessary), which can go a long way towards keeping the crap out of the audio system.

Moreover, the audibility of jitter effects is greatly exaggerated. In practice you need quite a lot of jitter to make it audible. Orders of magnitude more than what many people seem to believe.

Hence my tip: Don't worry about jitter, do worry about synchronization. There can only be a single clock master in a system (unless you employ sampling rate converters). Think about which one of your boxes is going to fulfill this role. The choice is often obvious once you think about it.

Adats and clocks - are external clocks snake oil ?

Reply #8 –
Great  replies thanks for taking the time.

In case your curious ambisonics yes but also v bap , wave field synthesis and 3 d audio in general. Though I guess this applies to any kind of multichannel recording of bands, orchestras etc as well.
 

Adats and clocks - are external clocks snake oil ?

Reply #9 –
Quote from: augustine on
If you have four adats coming off say an rme raydat card conventional wisdom says the adats should be set to slave  and us the rme as a master.  Others dispute this. Anyone hear have an op ion on this - are external clocks snake oil ?


A fair number of pro audio interfaces have provisions for running a number off them off of the same clock.  Some have several different procedures, all of which can work well.

The procedures for doing this varies from product to product.

External clocks are at least a convenience. Many audio interfaces have means for having a common clock that do not require external clock.

The performance of a group of interfaces is dependent on the common clock.

There is a lot of fantasy and fiction about clocking and jitter. Audible jitter usually takes screwing up.

This is IME very good advice:

Hence my tip: Don't worry about jitter, do worry about synchronization. There can only be a single clock master in a system (unless you employ sampling rate converters). Think about which one of your boxes is going to fulfill this role. The choice is often obvious once you think about it.