External Clocking - is it all that?

splitrmx

woodsdenis wrote: »

If I was to give a piece of advice as to how to improve your the sound of your DAW I would say get an external clock first.

I've heard a few different schools of thought on this. Dan Lavry (who makes some amazing converters) says:

"As a rule Internal clock yields less jitter, so use the internal clock when ever possible. There are times when one must use external clocks (such as the case of many channels) When doing so, the jitter goes up and the sound quality suffers a bit. That is a compromise we have to live with. Just do not fall into the false hype claiming that external clock improves your sound."

He also talked on Gearslutz (I'm trying to find the exact quote) about how the clocking part of the audio device is the simplest part to get right, and the actual converters are the more complicated and expensive part, and if you buy a great converter it should already have a decent clock in it.

This would seem to suggest that instead of buying an external world clock, a person wanting to improve the sound when working ITB should just get a better converter.

That said, I've heard other people say that when they use the Apogee Big Ben to clock their Pro Tools converters that it improves the sound.

I'm an amateur home producer myself (who spends more money than is sensible on audio equipment) and if I had the choice I'd get a proper converter before going for an external clock. However more research should be done into each choice.

As for the ITB vs OTB argument, that's one of those "vinyl vs CDs" ones that will rage to the end of time. I own a Mackie 1402 and a Motu 828 MKII (yes, this converter is currently the weakest link in my studio!) and when I stick everything into the Mackie it sounds a bit better than the Motu. They're both low end pieces of gear though.

PaulBrewer

splitrmx wrote: »

if you buy a great converter it should already have a decent clock in it.

That's true, it should. However some don't.

Also most pro-setups would be multi chan so a distributed master clock would be the norm. Some outboard equipment also needs word-clock - so NOT distributing clock isn't an option, even if it is desirable in other aspects.

Of all the things in a studio guys don't want to buy a €1400 box that doesn't make any sound is one!

However, on plugging them up and making them go in a clients studio they rarely return to me.

splitrmx

PaulBrewer wrote: »

That's true, it should. However some don't.

Also most pro-setups would be multi chan so a distributed master clock would be the norm. Some outboard equipment also needs word-clock - so NOT distributing clock isn't an option, even if it is desirable in other aspects

Wouldn't most good converters output world clock that could be chained to multiple devices that needed it using those bnc splitter connectors?

PaulBrewer

splitrmx wrote: »

Wouldn't most good converters output world clock that could be chained to multiple devices that needed it using those bnc splitter connectors?

Looping word clock is considered a no-no. A master clock will have multiple outs i.e. one for each unit.

madtheory

Here we go again... all current converters use a PLL (Phase Locked Loop), which basically means that they re clock, so an external clock makes no difference whatsoever. Also sometimes referred to as "jitter immune". The PT HD boxes have a PLL, for example.

Apogee are trying to sell you a Big Ben, so they go on about how important clocking is, Dan Lavry's info just makes one more aware of what to look for, and is not based on trying to sell you something (his advice applies to all converters).

Five years or so ago, external clocking made sense, but not now.

As for looping clock, it will work perfectly as long as you know where the termination is, and you're using 75ohm cable. Problem is most boxes have the termination built in, so unless you open them up and do a little surgery, you'll have feedback all over the shop.

jtsuited

epic clocking war begins here.

fitz

Split this off from the "Analogue Desks" thread so a proper clocking discussion can rage on in its own little space. Have at it gentlemen.

woodsdenis

splitrmx wrote: »

I've heard a few different schools of thought on this. Dan Lavry (who makes some amazing converters) says:

"As a rule Internal clock yields less jitter, so use the internal clock when ever possible. There are times when one must use external clocks (such as the case of many channels) When doing so, the jitter goes up and the sound quality suffers a bit. That is a compromise we have to live with. Just do not fall into the false hype claiming that external clock improves your sound."

He also talked on Gearslutz (I'm trying to find the exact quote) about how the clocking part of the audio device is the simplest part to get right, and the actual converters are the more complicated and expensive part, and if you buy a great converter it should already have a decent clock in it.

This would seem to suggest that instead of buying an external world clock, a person wanting to improve the sound when working ITB should just get a better converter.

That said, I've heard other people say that when they use the Apogee Big Ben to clock their Pro Tools converters that it improves the sound.

I'm an amateur home producer myself (who spends more money than is sensible on audio equipment) and if I had the choice I'd get a proper converter before going for an external clock. However more research should be done into each choice.

As for the ITB vs OTB argument, that's one of those "vinyl vs CDs" ones that will rage to the end of time. I own a Mackie 1402 and a Motu 828 MKII (yes, this converter is currently the weakest link in my studio!) and when I stick everything into the Mackie it sounds a bit better than the Motu. They're both low end pieces of gear though.

My 2 cents on this. Using an External clock on any Protools system I have heard improves the sound, no question. Do Protools/MOTU interfaces have Lavry grade converters? NO. There is the answer. Dan Lavry is probably right in that if you have his converters there is no need. Do you know how much they cost?

More importantly though is that if you are running multiple Digital devices
you definitely need a Master Word Clock clocking/running everything. This does not apply to PT HD systems where they loop through. I am talking about DAW/Digital Desk/FX Units/Video/Analogue Tape sync etc.

My comments were based on whether an External Clock or a Summing box
should be your first purchase if you had a choice.. I would still go for the External clock.

PaulBrewer

woodsdenis wrote: »

Using an External clock on any Protools system I have heard improves the sound, no question.

That has been my and my customer's experience.

If anyone wants to try a Big Ben for themselves let me know.

PaulBrewer

madtheory wrote: »

Here we go again... all current converters use a PLL (Phase Locked Loop), which basically means that they re clock, so an external clock makes no difference whatsoever. Also sometimes referred to as "jitter immune". The PT HD boxes have a PLL, for example.

Apogee are trying to sell you a Big Ben, so they go on about how important clocking is, Dan Lavry's info just makes one more aware of what to look for, and is not based on trying to sell you something (his advice applies to all converters).

Five years or so ago, external clocking made sense, but not now.

As for looping clock, it will work perfectly as long as you know where the termination is, and you're using 75ohm cable. Problem is most boxes have the termination built in, so unless you open them up and do a little surgery, you'll have feedback all over the shop.

If it had ' no difference whatsoever' it would sound the same ? That clearly isn't the case.

madtheory

Has anyone done a blind AB test in a treated room, and decided that the external clock is better? If the answer is yes, then I'd have to agree. However, all I've ever read is conjecture, or non blind testing. Fact is, the way the converter works means that external clock ends up being synchronised to the internal clock with 100% accuracy, because it's a Phase locked loop. So the difference heard with the external clock is caused by subjectivity.

I contend that if the subjective element is removed, as with a blind AB test, then no difference will be heard.

Actually Paul, I'm tempted to borrow a Big Ben from you to set up my own blind AB test. I might get a few other guys on board, a friend has a super room with a HD2 rig. Would you be willing?

PaulBrewer

madtheory wrote: »

Has anyone done a blind AB test in a treated room, and decided that the external clock is better? If the answer is yes, then I'd have to agree. However, all I've ever read is conjecture, or non blind testing. Fact is, the way the converter works means that external clock ends up being synchronised to the internal clock with 100% accuracy, because it's a Phase locked loop. So the difference heard with the external clock is caused by subjectivity.

I contend that if the subjective element is removed, as with a blind AB test, then no difference will be heard.

Actually Paul, I'm tempted to borrow a Big Ben from you to set up my own blind AB test. I might get a few other guys on board, a friend has a super room with a HD2 rig. Would you be willing?

Of course! You're welcome to a unit.

What room and what converters?

Woodsdenis' background would be at the Highest End of recording for quite a while too so his is one opinion I would listen to. (if not always agree!)

I only have a very slight understanding of the science behind it.

Most of the guys who have purchased them would be more experienced chaps who -
1. Are looking for incremental improvement
2. Have no particular interest in the science behind it.... merely that it's 'better' in their opinion.

The Digi 192 seems to be one unit that raises it's game with the Big Ben as does the SSL Alpha.

Of note to is the fact one must record AND mix to gain maximum benefit (A2D + D2A) of using a clock.

madtheory

Cool, I just emailed a few of the guys around here, we might set up a test. The room is here, it's 192s:
http://www.soundsound.ie/index.php?article=gear
PS that's an old pic, it actually sounds better now

For me, opinion is not science... but that's just my opinion

PaulBrewer wrote: »

The Digi 192 seems to be one unit that raises it's game with the Big Ben.

Of note to is the fact one must record AND mix to gain maximum benefit (A2D + D2A) of using a clock.

RE the 192: is that at 44.1k, or 192kHz?

Your second point has me concerned. If you have a jittery clock, then the difference becomes more apparent with each conversion! Maybe jitter sounds like "better" to some ears? A bit like harmonic distortion does?

woodsdenis

madtheory wrote: »

Has anyone done a blind AB test in a treated room, and decided that the external clock is better? If the answer is yes, then I'd have to agree. However, all I've ever read is conjecture, or non blind testing. Fact is, the way the converter works means that external clock ends up being synchronised to the internal clock with 100% accuracy, because it's a Phase locked loop. So the difference heard with the external clock is caused by subjectivity.

I contend that if the subjective element is removed, as with a blind AB test, then no difference will be heard.

Actually Paul, I'm tempted to borrow a Big Ben from you to set up my own blind AB test. I might get a few other guys on board, a friend has a super room with a HD2 rig. Would you be willing?

Even try using a Digi Sync IO and you will hear the difference. Yes I have heard the difference in a treated control room.

I think you are getting confused with the way that digi syncs more than one of its interfaces. If you are using an external clock to clock a 192 using word clock where is the loop ? There is only one master. We are not talking about Digi loop sync we are talking about Word clock.

Word Clock
Many professional digital audio products—including
open-reel multitrack tape recorders, digital
mixing consoles, and the Tascam DA-88
modular digital multitrack—have Word Clock
(1x sample rate) connectors.
Word Clock allows the DA-88 (and other Word
Clock-compatible devices) to send or receive external
clock information which controls the
sample rate, which in turn (where applicable)
controls the play and record speed.
Using just Word Clock, it is possible to create a
“chain” of digital devices in your studio by picking
one source as the Word Clock master, and
configuring other sources as Word Clock slaves

This is from the Digi Sync IO manual. Granted what we are doing here is
for example controlling the sample rate of external devices from a master.
There is no loop here, one master clock controlling the sample rate/speed
of Protools against an analogue tape machine for example.

What we are discussing is the benifits of having a Master Word Clock for one or all of your digital devices. When using it on a Protools HD3 rig only there is a difference. There are many different brands Apogee/Nanosync/Aardvark. Try it out, if you dont notice a difference ignore this.:D

madtheory

woodsdenis wrote: »

If you are using an external clock to clock a 192 using word clock where is the loop ?

It's not that sort of loop, it's a PLL. This is a way of ensuring that the master oscillator (the one controlling the sampling rate) never drifts. It's inside the converter, part of the circuit. I'm not talking about Digi's way of syncing. PLL is a very common clock design for digital circuits, a.k.a. jitter immune. I think Benchmark were the first to trumpet it, although the concept itself is probably thirty years old.

And yes, I am going to try out the Big Ben

PaulBrewer

madtheory wrote: »

Cool, I just emailed a few of the guys around here, we might set up a test.

RE the 192: is that at 44.1k, or 192kHz?

Your second point has me concerned. If you have a jittery clock, then the difference becomes more apparent with each conversion! Maybe jitter sounds like "better" to some ears? A bit like harmonic distortion does?

Across various sample frequencies as I recall.

Well, as I say I'm no expert.

If the unit's own clock has more jitter that being clocked externally (even though there may be issues with that process as has been suggested) then my thinking was that if you're recording with less overall jitter with clocking externally (if indeed that is the case) so too are you playing back with less jitter.

That, of course, is presuming that Less Jitter does , in fact, sound better!

What room are you suggesting? I know very few 'good' rooms around.

madtheory

PaulBrewer wrote: »

What room are you suggesting? I know very few 'good' rooms around.

I edited my post, while you were editing, see previously for link...

woodsdenis

madtheory wrote: »

It's not that sort of loop, it's a PLL. This is a way of ensuring that the master oscillator (the one controlling the sampling rate) never drifts. It's inside the converter, part of the circuit. I'm not talking about Digi's way of syncing. PLL is a very common clock design for digital circuits, a.k.a. jitter immune. I think Benchmark were the first to trumpet it, although the concept itself is probably thirty years old.

And yes, I am going to try out the Big Ben

Fair enough madtheory but in the real world I am using an interface which isn't jitter immune therefore using a reliable clock improves this.

What audio interfaces for DAWs us this PLL system ?

PaulBrewer

Ah DanDan's room ...... ex of this Parish! Very ex

Cool, just let me know when you need it. Our demo one is out at the moment but should be back in a week or two.

To reiterate the point I think the fairest way is to record AND play back using the clock.

I know in one instance we played back a mix that was done on 192s with a Big Ben and everyone agreed that it now sounded like there was a bunch of frequencies around the 5k area quite prominent.

The mix engineer (Ciaran Byrne in Cauldron) then piped up that that's what he'd been adding to his 'clockless' mix ! Had he not been around that info might have gone unnoticed skewing the outcome.

PaulBrewer

woodsdenis wrote: »

Fair enough madtheory but in the real world I am using an interface which isn't jitter immune therefore using a reliable clock improves this.

What audio interfaces for DAWs us this PLL system ?

I think it's a very interesting area and as Madtheory has alluded to perhaps it's the making more jitter that sounds better !

I personally don't know - however I'd say we've sold maybe a dozen Big Bens and of that dozen about 8 of those are pure music guys i.e. non techy in anyway so it does seem to be a very music driven decision.

PaulBrewer

I've just dropped a line to Apogee and SSL to see have they any thoughts on the subject for us.

woodsdenis

I dont know of anyone who thinks that jitter is a good thing.

However, intrigued by madtheorys post I googled PLL audio and the only thing that came up in about a minute of searching was a section from the Bob Katz book on mastering, who had been externally clocking all his equipment, until he bought a PLL convertor. I am sure they sound fantastic and don't need to be externally clocked but I suspect they cost a great deal more than what the majority of us are using.

Bob Katz's opinion is that all Firewire interfaces should use external clocks.
"Let Firewire carry the Audio not the clock"

I still would like to know what brands of Audio interfaces use this type of PPL converter.

PaulBrewer

woodsdenis wrote: »

I dont know of anyone who thinks that jitter is a good thing.

However, intrigued by madtheorys post I googled PLL audio and the only thing that came up in about a minute of searching was a section from the Bob Katz book on mastering, who had been externally clocking all his equipment, until he bought a PLL convertor. I am sure they sound fantastic and don't need to be externally clocked but I suspect they cost a great deal more than what the majority of us are using.

Bob Katz's opinion is that all Firewire interfaces should use external clocks.
"Let Firewire carry the Audio not the clock"

I still would like to know what brands of Audio interfaces use this type of PPL converter.

Phase Locked Loop it stands for as I recall. It's not a new thing, used in radio tuners etc I believe.

http://en.wikipedia.org/wiki/Phase-locked_loop

PaulBrewer

Actually reading the PLL wiki the implication is that it references to a 'clock standard' i.e. a wordclock master in our case.

Isn't the implication then that the more stable and accurate the reference so to the PLL circuit and hence Jitter?

This would imply jitter is directly related to the master clock's accuracy or am I missing something?

madtheory

Ya, but the master is always the internal oscillator.

woodsdenis wrote: »

What audio interfaces for DAWs us this PLL system ?

All the Digi HD stuff And pretty much any converter built since around 2004, give or take.

I can't keep up with the rate of posting and editing!

Bob Katz's book is a little out of date on the issue, I haven't seen the revised edition so I dunno what he says about it in that. Given the omnipresence of PLLs, I'd say that no current firewire box takes its clock ref off the firewire bus.

Note that Dan Lavry has had some very heated arguments with Apogee folk online, about their claims for their cables and their clocks. Funnily enough, it was always Mr.Lavry that had the science argument, and Apogee the subjective line...

teamdresch

Hehe, clocky/jittery topics are always fun.

I finally picked up a wordclock cable to hook up two pieces of gear instead of clocking them across ADAT.
I'm expecting to convince myself that I can hear a difference

splitrmx

PaulBrewer wrote: »

I've just dropped a line to Apogee and SSL to see have they any thoughts on the subject for us.

Gee I wonder what Apogee will have to say about the subject of whether external clocks improve things or not!

PaulBrewer

madtheory wrote: »

Ya, but the master is always the internal oscillator.

So it doesn't ever reference to an external one? So does that imply in our situation that PLL isn't in use if one uses an external clock source?

madtheory

Re: what Apogee think- see my previous edited post, Lavry v Apogee.

PaulBrewer wrote: »

So it doesn't ever reference to an external one? So does that imply in our situation that PLL isn't in use if one uses an external clock source?

The PLL effectively re clocks whatever you send in.

PaulBrewer

splitrmx wrote: »

Gee I wonder what Apogee will have to say about the subject of whether external clocks improve things or not!

Indeed, I hear ya barking Big Dog !

Obviously they may be 'pro' clocking however it's the science behind it they make have points to make on.

madtheory

PaulBrewer wrote: »

Indeed, I hear ya barking Big Dog !

Obviously they may be 'pro' clocking however it's the science behind it they make have points to make on.

Note that Dan Lavry has had some very heated arguments with Apogee folk online, about their claims for their cables and their clocks. Funnily enough, it was always Mr.Lavry that had the science argument, and Apogee the subjective line...

I think it's also worth noting that jitter distortion is at least as quiet, and probably quieter than, 16 bit dither. So in most studios, there are much bigger issues than clocking...