x
anonymous 28 January 2002

In the new issue of Tape Op, Bob Katz of Digital Domain states that "99 times out of 100, the INTERNAL clock in a converter will produce lower jitter and thus better sound than if you clock that converter externally from the most stable wordclock generator in the world! The reason: A crystal oscillator is much stabler than a phase locked loop. And when you force a converter to slave, you are replacing its internal crystal with a much less stable phased lock loop. The only exception would be a converter whose internal oscillator is so poorly designed that it is not as stable as its PLL." (Quoted from Tape Op)

I'm trying to reconcile the above with various authoritative-sounding postings and comments (some of them by those who sell external clocks, some by those who use them) that purport to describe how much an external clock like the Lucid improves the sound of even non-bargain basement converters.

Any one care to weigh in?

I'm not talking about WC distribution to multiple places. I'm simply talking about any sonic advantage gained from external clocking vs a converter's "built-in" clocking.

Comments

Ang1970 Tue, 01/29/2002 - 01:00

From Bob Katz's website:

[[url=http://[/URL]="http://www.digido.c…"]It is much easier to build a low-jitter PLL for a wordclock signal than for an AES/EBU signal.[/]="http://www.digido.c…"]It is much easier to build a low-jitter PLL for a wordclock signal than for an AES/EBU signal.[/]

This would indicate that not all PLL's are made the same. I suspect Bob may have been exaggerating when he said 99% in the TapeOp article. IMO, what needs to be analyzed is the differences between PLL's - different methods, different designs, and quality of construction.

Guest Thu, 01/31/2002 - 01:58

The built in clock on an Apogee unit I have is better soiunding than an external clock device (Rosendahl) I bought, however Digidesign converters are improved by the Rosendahl and I have heard a friends Motu 2408 II improved by his Rosetta. So... how about them apples?
Jon Attack from Capital Studios Paris has a LOT ear time experience with this issue, perhaps he can comment.. Jon?

:)

anonymous Thu, 01/31/2002 - 15:22

I don't think there's a simple answer to this.

You can make a good or a bad crystal clock in a converter, and a good or bad PLL or slaving clock. There is also the question of whether what is a good clock on its own, survives the surrounding circuitry and interconnects inside and outside the box.

I would expect a good xtal to outperform an average PLL, but I would also expect a good PLL to out-perform an average crystal.

In addition, a good PLL can actually de-jitter incoming data, but bad cabling can ruin all the gains of a good clocking system.

In my opinion, there are three main things people get wrong when doing converters: analog design, clocking and power supplies, possibly in that order. My opinion is that they are the things that make a converter sound good (or not), with the analog design defining the character of the sound as much as in any other analog device.

By comparison the digital bit is (relatively) easy.

--Richard Elen

anonymous Thu, 01/31/2002 - 15:31

Originally posted by atlasproaudio:
Explain this...My converters in the Tascam MX2424 sound better than the Apogee by itself when clocked from the Apogee....

I don't know which Apogee you have, but we do try hard with our clocks. It can be difficult to keep a clean clocking signal inside a complex piece of equipment - computers are the worst, I think. So the Tascam's own clock may be being compromised by its environment.

I would normally expect the external clock from the Apogee to improve the performance of a unit's own converters under such circumstances.

However I am of course disconcerted that our clock makes the Tascam converters sound better than ours :( I wonder how you'd describe the difference?

--Richard Elen
@ Apogee

Scott Gould Thu, 01/31/2002 - 21:23

There's no doubt that converters & clocks sound different, and people have various preferences; but I have yet to see a definitive consesnsus. Whether converter 'A' is better than converter 'B' (with it's own clock or another) would seem to be subjective, depending on the material being recorded/monitored and the prefernce of the engineer/producer involved. What seems clear to me is that it behooves anyone with pretensions of professionalism to have choices of converters & clocks available. If you don't like the sound you're getting, change something. Just like you would with mics or pre-amps. I use clocks & converters from Lucid, Apogee and Crane Song & I'll monitor both pre and post converter - changing things until it seems 'right' for the material. (Sometimes for a more "lo-fi" sound I'll even use the converters in my d8b or MOTU828). I try to keep myself informed (lots of reading) but by far the most usefull tool in my box is my ears. It doesn't take much research to find out who the players are in the clock/converter market - but you have to LISTEN to them, don't just buy the specs/endorsements/current buzz/expert pronouncements. I use this approach to all equipment decisions, and in 30+ years I've bought mighty few lemons. Oh, and don't be afraid to break the rules, or buck the trends. :D
Peace
Scott

anonymous Fri, 02/01/2002 - 11:49

Originally posted by DigitMus:
...you have to LISTEN to them, don't just buy the specs/endorsements/current buzz/expert pronouncements....Oh, and don't be afraid to break the rules, or buck the trends...

I couldn't agree more. Whenever you're considering spending a significant amount of money on audio gear, your ears give the best advice. If YOU can't hear the difference between one converter and another, say, or YOU prefer a particular software's ergonomics, it doesn't matter what someone else's opinion is, however well-respected they are. You have to judge for yourself.

I just wish we could come up with a way to let potential purchasers audition equipment more effectively and do real comparisons before purchase.

--Richard Elen

anonymous Sat, 02/02/2002 - 01:26

Yes...
The advice offered by Richard,Atlaspro et al (we'll ignore Jules because he's got tin ears as a result of producing too many full-on geetar bands :) ) is the most worthy....in the final analysis..trust your own ears...
However...it can be reassuring to be able to depend on something occasionaly, and, in the modern studio environment, you have to have a master clock to reference all your digital kit.....
So, it's all very well arguing about internal crystals v PLLs...but...the bottom line is all the tackle has to communicate with each other.....
Anything which can't do that goes in the bin...old time studio law...
So..yes..one can hear the difference (in a good room) between a digidesign 888-24 running on its own internal clock versus being clocked by a USD (about one tenth of the jitter).....
Apogee may claim that the crystal in an AD8k has less jitter than a USD, but that means jack-shit if you have to put your rig on-line.......
OK...people find their own preferred ways of doing things, but I've always found that the Aardsync gives a mighty solid clock, with a multitude of options, and everything sounds cool.......magic...
...now...let's make some music!

Guest Sat, 02/02/2002 - 13:57

Clock 'shoot outs" are a PITA.
From:

Apogee
Cranesong
Digidesign
Rosendahl

I chose Apogee, Cranesong was kind of a joint winner, I felt I could hear the bass drum better with the Apogee clock, so I stuck with that, the Cranesong had other cool qualities the Apogee lacked but I had to pick one..

Scott (DigitMus) has got it bad but the lord of all clock crazys on RO is Jon Attack, I will see if I can get him to justify what I refer to as his wordclock "fetish".

:D

Guest Sat, 02/02/2002 - 14:03

Clock 'shoot outs" are a PITA.
From:

Apogee
Cranesong
Digidesign
Rosendahl

I chose Apogee, Cranesong was kind of a joint winner, I felt I could hear the bass drum better with the Apogee clock, so I stuck with that, the Cranesong had other cool qualities the Apogee lacked but I had to pick one..

Scott (DigitMus) has got it bad but the lord of all clock crazys on RO is Jon Attack, I will see if I can get him to justify what I refer to as his wordclock "fetish".

:D

anonymous Tue, 02/05/2002 - 08:42

I have just done a little "shootout" myself! Actually two! For anyone who's interested, here's what I heard:

In my project studio: A/D converter was Lucid AD9624. I recorded a 16-track keyboard sequence (to eliminate differences in "performance") one track at a time to hard disk (24 Bit/44.1 kHz) using 3 clocking situations.

1) Lucid AD clocked internally.
2) Lucid AD clocked to Lucid GENx6.
3) Lucid AD clocked to Rosendahl Nanosync.

The differences are fairly subtle. External clocking doesn't make a ridiculously HUGE difference to my ears with my gear, but it does make a difference. It's not like the jump say from a Mackie to a Neve! More like the difference between a Mackie and an A&H or Soundcraft Ghost!! :) In any case, the results were similar, although more pronounced due to the monitors I suspect. The "master" clocked with the GENx6 had "extended", "present", "tight" high end, although on these monitors (sorry don't know what they are), it sounded extremely "clean" compared the high end on the Apogee. And the bottom with the GENx6 sounded "skinnier" but more "focused" as well. With the GENx6, it was as if the kick drum had been mixed higher than the bass. With the Apogee, it was the opposite, as if the bass had been mixed higher than the kick drum. The bottom end with the Apogee internal clock also sounded a lot "wider". But as I mentioned the top end with the Apogee internal clock was significantly different. By comparison, the GENx6 track almost made the top end on the Apogee internal track sound as though it had been attenuated! I brought a disc of this home, and my impressions remained the same. Interestingly, I preferred the GENx6 "master", one of the mastering engineers preferred the Apogee clock "master" (he's a big bottom end guy!), and the third mastering engineer said he could hear the difference, but wasn't sure which he liked better!

BTW, I also had a disc mastered that night. Out of interest, we mastered the disc clocked to the GENx6. Now that some time has passed and I've had the opportunity to compare work I've had mastered there before with this project, I think the GENx6 master is the best sounding of all the discs that I've had mastered there (probably about 10-15 albums).

Also, BTW, the mastering "shootout" happened first. I don't think it colored my impressions of the project studio "shootout", but it certainly could have! :)

In any case, those are the results of my very unscientific "shootout"! I hope that someone finds this useful, or at least interesting!! And most certainly, YMMV!!! :)

Brent

Bob Olhsson Thu, 02/07/2002 - 13:12

An important thing to understand about jitter is that less jitter at the "wrong" frequency could sound worse than more jitter that has a more pleasing frequency signature.

Another important thing to understand is that while clock jitter impacts an audio signal as it is recorded, when you play it back, jitter only impacts your monitoring unless you are recording the analog signal from your converter. If you make a digital mix, transfer or bounce, the only clock you'll ever hear is that of the final CD player. (This could explain part of why people prefer analog mixing to digital!)

Finally, A to D jitter interacts with D to A jitter producing sum and difference frequencies which means it is a moving target and you can't assume a recording you like played back with one clock is going to sound that way with another.

x

User login