|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
84.160.32.25
In Reply to: Re: Firewire: jitter-resistant or clock-quality-dependant? posted by Gordon Rankin on August 31, 2006 at 07:56:16:
Hi Gordon,
let me first quote myself with the answer for the other poster, taht way, you might get a better idea how my DAC works.
The DAC I am using (RME FireFace 800) is a recording studio quality DAC which accepts adat, spdif, AES, and FireWire via ASIO. Its internal clock is of very high quality and it has very elobarate jitter correction. In the recording studio, you can use it as a master-clock for all other gear.
So no need to upgrade the clock on the DACs side. My question only originated in the thought, that storing audio on my computer, the computer might introduce any kind of "jitter" or other crappiness into the signal, and feed it via FireWire to the DAC - the DACs jitter-and signal-correction mechanism might have to work extremely hard, which I might be able to change, if I use the DAC as some kind of "state of the art" master clock....
From your explanation I see, that I cannot slave my computer. But (sorry, but I don't know about any of those programming/interface things such as controllers): My DAC is from highest quality and the DACs "firewire reception interface controller" (sorry for the strange, non-technichal term) is especially programmed by custom-developed firmware for RME, does THEIR "data reception controller" tell the computer anything about what to do?
Would the computers low-end quality firewire controller, which sends the data, fire some kind of crappy signal to my DAC, so that the DAC has a hard time to put the data into order?Here is some info about the DAC and it's controlling mechanism - maybe, that gives you some additional idea...
Follow Ups:
Actually the RME spec for clock juitter is pretty ropely. 2 nS is massive compared to <100 pS for good CD players. RME is, of course, dealing with some poor computer or studio electronics and unless steady clock can reduce jitter to pS range, is pretty useless.The XOs that RME uses in the ADI2 and PAD2496 is an industry standard unit (Epson) and the power supplies are not great (by high end audio standards).
So, for computer and ordinary studio stuff, RME is head and shouders above most, but definitely not the best. For this, one has to go to dCS or such like, at 3-5 times the price. Even then, a $3-5000 clock unit still improves the sound eg BIg Ben or Verona (I think?).
DCS claims that their $10K external clock will improve the sound of their DACs but that is a pretty suspect statement.It is rather hard to believe that a clock from an external box tranmsitted over a 3 foot cable with 2 connector in the middle ends up a the DAC chip in any better shape than a clock signal hardwired and located closely in the same box. If, however, the external clock truly improves the sound, I can't help to wonder how much better this could sound if they put the same clock inside the DAC.
Studio clocks are meaningful when you have to synchronize multiple pieces of equipment but for a playback chain I fail to see the benefit.
Cheers
I've had a little experience with this recently.Currently I'm running a system that has a low jitter clock in the DAC box (about an inch away from the DAC chips) driving reclocking flops that reclock the data and clocks going to the DAC chip. I also send the clock to a SB3 and feed it in inplace of the crystal. I then tapped off the I2S signals going into the SB internal DAC chip and send it to my DAC. The connection is about 1.5 feet of ribbon cable, no LVDS or any special drivers of any kind.
The interesting thing is that the sound from the analog out of the SB3 got way better when fed the low jitter clock, even though its going over an unterminated cable!
Yep if I put the clock in the SB3 it owuld probably sound even better, but I was very surprised at how big an improvement there was with an external clock. I was ready for the analog outs to sound worse because I expected the clock jitter to be terrible with the cheap and dirty connection I made.
Just read the audio reviews; clocks are very important. Most XOs are not very good, especially in computer audio.I have first hand experience of clocks; the better the jitter performance, the better the power supply, and the better the matching of impedance, the better the sound.
I know that bits are bits believers don't believe the importance of electronics!
This post is made possible by the generous support of people like you and our sponsors: