Is 96khz better than 48khz for multitrack digital recording?

[Dan Lavry]

Originally Posted by Mike Derrick

Dan,

I assume when you talk about the clocking issue you are refering to the use of a separate clock unit, different from a soundcard<-->converter.

I was wondering if it makes any difference whether a soundcard is the master for the clock or the converter is the master for the clock?

I am talking about the use of a seperate clock unit.
For best conversion performance, the clock circuit should be near the AD circuit, inside the same chassis or location. When you use a seperate clock unit outside, away from the converter circuitry, your setup will end up with more jitter.

But do not confuse "conversion jitter", with "data transfer jitter". For conversion, the clock has to have real tiny amount of jitter. Jitter on the converter will alter the data and reduce the sound quality (more distortions and noise, less transparancy). But jitter during say transfer of data from AD to a computer can be relativly high with no negative outcome.

When doing the AD conversion, early and late clocks alter the data fom ideal sampling to less then ideal sampling. But once the conversion is done, the data is "cast in stone", it is made of 0's and 1's and transfering a sequence of say 10010010101... does not require precision timing. Data transfer is not much different then sending information over the web. It is not much different then most digital operations, where the objective is to keep the 0's and 1's patterns in tact.

Regards
Dan Lavry

[Gabriel Sousa]

Originally Posted by Dan Lavry

... True 20bits is rare. My Gold AD is 21 real bits. The lavryBlack AD is 19.5 bits. ...
Regards
Dan Lavry

hello Dan Lavry, and about your lavry blue, how many real bits ? i have one .

[Gabriel Sousa]

Originally Posted by Dan Lavry

There is such a thing as optimal sample rate, and as I stated many times I think 60KHz would be great, but we do not have such a standard. ...

Regards
Dan Lavry

can we all sign a paper to make ir standard ????

[Dan Lavry]

Originally Posted by Gabriel Sousa

can we all sign a paper to make ir standard ????

At the start of digital audio, there were a number of engineers that wished for 60KHz sample rate. There were a number of reasons for that. During those days, we did not yet know about oversampling AD’s, so one of the difficulties of digital audio had to do with anti alias filters (an analog filter in front of the AD). Everyone wanted to have digital audio cover at least the “HI-FI” range, namely 20-20KHz.

At 60KHz rate, the filter transition band is around 10KHz (60KHZ/2 - 20KHz). At 44.1KHz sampling rate the transition rate is at around 2.05KHz (44.1KHZ/2 - 20KHz), making the analog filter about 5 times “more difficult” – about 5 times the circuitry for the same performance, not to mention worse phase response…
But all that did not matter.

A single large manufacturer can “force a standard”. I heard that the decision to go to 44.1KHz had to do with the desire to fit Beethoven symphony number 9 on a single CD. I do not know if the story is correct. It is possible that the CD format of the time could not read and write reliably at 60KHz. There may be other reasons I am not aware of.

At the “end of the day”, it is all about what the large masses will accept, and that of course has a lot to do with what the larger equipment makers decide to do. Some of it is good, some of it is bad… But once you have a standard accepted in the market place, it is just so much more difficult to change…

Regards

Dan Lavry
Lavry Engineering - Unsurpassed Excellence

[Toasterboy]

I just recorded a piano on Radar, at 32kHz 16 bit, and at 96kHz 24 bit and I just can't tell the difference. If anyone wants to hear samples, just let me know how to post them!!

[Gabriel Sousa]

Originally Posted by Dan Lavry

... True 20bits is rare. My Gold AD is 21 real bits. The lavryBlack AD is 19.5 bits. ...
Regards
Dan Lavry

i'm still curious, and about the real bits of lavry blue ...

[peeder]

Originally Posted by Gabriel Sousa

i'm still curious, and about the real bits of lavry blue ...

dynamic range/6, I believe is the equation.

[dangoudie]

Originally Posted by TornadoTed

I think a lot has to do with how trained your ear is.

I don't necessarily disagree but have you never twiddled an EQ knob a bit and thought it made a difference and then realised that it wasn't even activated?

I think that fact that people can 'hear' a difference when they know which is which is pretty unreliable and inaccurate. Therefore a blind test can be the only reliable test. But even then, people are LOOKING for differences and on'e mind's ear can be easily fooled.

I work at 24/44.1 but I almost exclusively work with rock so some people have said they prefer that anyway.

The explanation of the HF distortion that is relocated with higher sampling rates makes a lot of sense and it's the only realistic difference that I can give full credit to. Psychoacoustic are so hard to prove so...

[The Studio RI]

Originally Posted by Fletcher

\
FWIW I don't like 48k sampling that much because it's a noticeable sonic improvement [in the 'audiophile' dept.] to 44.1k and should you stay in the box it will require a 'sample rate conversion' to make a CD... which I have always seemed to find more destructive to the audio than the higher sampling rate is beneficial.

As always... YMMV

Second...

[soundseeker]

I want to record at 96k on a 2.2ghz macbook pro. Ive had problems. Is it possible or not? Someone at the mac store said ts not possible without a mac pro. Digital village said Iys possibe with a 72000rpm external firewire drive.

What should I look to do?