(Does anyone know what article I'm talking about, and where to find it ?)
#WORD CLOCK RECORDING SERIES#
I'm sorry I can't find the article anymore, but I read one in which an engineer made measurements on a series of modern pro-grade converters, and found out this : most converters distort more when slaved to an external clock, no matter how good that clock is, because it is somehow harder for them to be a good slave than a good master. What I really learned is that it makes sense to pay attention to the details, but ultimately Mr. There are so many choices to make, multiple clocks running asynchonously vs multiple clocks locked together vs a single divided and distributed clock vs heaven knows what. And it will be oodles (that's a tech term) more efficient than the equivalent linear power supply.Īnd clocks? WHOLY CARP, clocks are difficult. It was my first foray into switching power supplies, I still don't like them, but you can design a good, clean power supply with a switcher. Many years ago a friend and I set out to build a personal mixer system that used a novel (at the time) new network audio protocol - AVB. Not always a good vs bad thing, but differences do exist. (the same applies to analog stages, power supplies, pretty much every building block.)Īnd, while I doubt I can hear it any more, there are differences between even well respected devices. Shocking though it may seem, the reference designs are getting worse and worse, I would not bet the farm on a reference design. That said, it is NOT a trivial task to build a good clock circuit. I think today it is as much a matter of convenience as anything else, most clocks are pretty stable already.
Once upon a time it was true that an external clock could help some converters, even if it was the only box in the studio, and especially if there were multiple boxes in the studio. If you need to clock multiple devices to word clock, a standard video distribution box (they're cheap) will work. Every good audio device, whether it's an interface or processor, has a good clock, and I'd rather invest in that.
Today - not in the past, today - I personally wouldn't buy a clock generator. ADATs used a (was it 10-pin?) wire for syncing multiple tape machines, while using lightpipe for audio. Do not use lightpipe for clocking, because a) it's not very reliable, b) it won't sound as good. If you can run wire between two or more devices, either using word clock or just "digital black" clock, instead of using the clock embedded in the digital datastream along with the audio, it will probably sound better Whenever possible, use the device doing the A/D as the clock master I did a lot of playing around with clocks at the time (because I had several digital devices). His argument was what Scoremixer says - that recovering an external clock from a PLL (phase-locked loop) is intrinsically more jittery than using an internal clock. Bob Katz, a mastering engineer, said that you need to shout at the designer of your equipment if it sounds better with an external clock. I recently bought a RME wordclock card for my slave computer and it has noticeably made a difference in the audio that used to sync via ADAT, regarding pops and clicks. I don't notice jitter but I do notice occasional pops or clicks in the audio from devices clocked this way, especially via lightpipe, which has to be run very carefully. But in my experience getting sync via these cables is much less reliable than word clock. I could clock everything to one source using a combination of the right SPDIF, AES, and ADAT cables. a couple of digital mixers, my 2408 cards, my RME card in the slave computer, some digital outboard FX, the TC Electronics Finalizer. I have always had a ton of external gear that needs clock. Interesting article, especially about half-way down the part about "interface-induced jitter":
Some people also say they can hear the difference between 44.1k and 48k. Click to expand.Some people say solid clock improves the audio.
0 kommentar(er)
