R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [3]  All   Go Down

Author Topic: Digidesign White Paper  (Read 19434 times)

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Digidesign White Paper
« Reply #30 on: October 23, 2006, 06:49:36 pm »

Max wrote on Fri, 20 October 2006 23:57

While it may be true that Dan did not edit any of Lucas' posts....


You said:
“There are things we tried to express that Dan Lavry and others purposely omit in these clocking arguments.”

Not so. What you propose here is a new argument, and I am responding to it. This is the first time I have heard this argument (why?)

The last argument we heard from engineering stated:

(DEC 27 2004)
“We do CLAIM:
1. The Big Ben is an excellent low jitter solution if you need a master clock
2. If you have a clock to a very jittery source, you’ll get better results most of the time by having the Big Ben cleaning it up first
We haven’t claimed anything else.

Lucas van der Mee
Sr. design Engineer
Apogee Electronics”


Soon after the above was posted, I considered the argument settled.

Now we have a new argument claiming that one can have some “special jitter” to make things sound better in most converters. Let’s examine it:

Different AD and DA’s (IC’s) respond to jitter in different ways. The audible effect of jitter on the reproduced signal depends on the choice of IC (modulator over-sampling ratio, modulator number of bits, and more. There is literature out which discusses this in depth).

Next, the “inserted jitter” needs to go through a PLL on its way to the converter IC. PLL’s (and thus their reaction to jitter) also vary a lot from design to design (number of PLL stages, corner frequency…) .

The argument says:  “It is very easy to create a scenario whereby two signals can be induced with jitter, one of which has significantly more jitter than the other, but sounds audibly more ACCURATE to the ear.”

Even those that wish to argue that jitter “improves” sound, must restrict their arguments to altering specific signals. The argument is wrongly extended to a claim that jitter can improve any signal without any knowledge of the signal itself. The music is a wave that changes all the time, and the clock box has no access to that signal. It does not even know what is playing.

What is being “blurred” here is the difference between subjective opinions and scientific facts. Also blurred are comments regarding the impact of jitter on specific known signals, and the impact on music (any signal), all leading to the use of the word “Accurate”.  Accurate according to what definition?
And all that is predicated on the wrong assumption that most AD’s will react to the “special” jitter similarly.

But the sales guy says:

1. “It is certainly possible to design an external clock in such a way that the increased amount of jitter that results will have less audible effect…”

It is especially easy to understand that this premise is wrong when you take into account the fact that the clock box does not get to see the music signal. Additional this “special jitter” will go through various types of hardware that will be reacting to this “special jitter” differently in every case.

The new argument does not state “we figured how to do something”. It says “it is certainly possible”. It does not say “we implemented it”.

And then you said:

2. “For this reason one can not claim that an internal clock will always be more accurate…and the testimony we have been seeing on the market and in our own testing… certainly attest to that.”

Why are you talking about what one may claim or not claim? We are talking about what you have inside your product, by design (hardware, firmware, software…)  

And here is your previous argument: (Proper Word clock implementation thread, Oct 27 2004)

“The problem with discussing the how and why of Apogee's advanced science in this field is that there are things we are doing that we do not want to advertise to our competitors…”

This suggests a scenario such as:

There was jitter research project resulting in some certain type of jitter (intentional timing error in capturing the signal), that when inserted by way of external clock circuitry into most converters, makes them sound better. The insertion of such intentional jitter improves any audio signals across the board; it does not require any analysis of the audio signal itself.  
The “special jitter” was implemented in such a way that inserting the same “intentional timing errors” would improve sound regardless of and without any knowledge of:
1. The music being played
2. The type of converter IC
3. The clock signal path implementation (PLL)
4. Other sources of jitter

The Engineer was stating that the box will remove jitter (from sources external to the clock), which is a very reasonable clock maker’s goal. The salesman is talking about their “advanced science” resulting in adding more jitter of a certain type to do the impossible.

Should we disregard the original engineer’s statement of DEC 27 2004 or your statement-

“I am the spokesperson… and anything I say is wholly endorsed by engineering, making the distinction between myself and anyone in our engineering department moot.”

Dan Lavry
Logged
Pages: 1 2 [3]  All   Go Up