[time-nuts] Clocks for Audio gear
actast at hotmail.com
Thu May 10 17:49:37 UTC 2012
Great dialog. The Time Nuts form can be very humbling and often has me questioning my own knowledge base. This thread is no exception. My take on the effect of jitter is when an d toA converter is reproducing a pure sine wave even a single point of slight jitter will show up in the FFT as false analog content. In a complex wave it can easily change the relationship between a fundamental and harmonics. This is a major problem in speaker crossovers design where drivers with different masses can only be aligned at one freq. It has been my experience this relationship between fundmentals and harmonics is noticeably audible at levels that are difficult to measure. To me, it is this difficulty to mathematically define certain aspects of audible distortion that makes audio design fascinating. (It also makes some voodoo science and snake oil products difficult to disprove, especially to those lacking an electronics or physics background) Will timing improvements affect sound quality, because of the complexity of human hearing it seems to me still worthy of investigation. Time will tell.
> From: albertson.chris at gmail.com
> Date: Thu, 10 May 2012 09:48:21 -0700
> To: time-nuts at febo.com
> Subject: Re: [time-nuts] Clocks for Audio gear
> On Thu, May 10, 2012 at 12:57 AM, MailLists <lists at medesign.ro> wrote:
> > Hearing tests showed the ability to discern jitter above a few hundred
> > nanoseconds rms.
> > http://amorgignitamorem.nl/Audio/Jitter/Detection%20threshold%20for%20distortions%20due%20to%20jitter%20on%20digital%20audio%2026_50.pdf
> > Others claim the ability to detect jitter in the picoseconds range...
> If we are to believe the above paper,then those guys who claim to hear
> pS jitter are wrong. Likely they can also here is a fuse is is place
> in the holder "backwards".
> So by the above, no now can hear 250 nS of jitter. I really doubt
> any decent system other then the most low-cost consumer level junk has
> jitter at the 250 nS level. Even a TTL "can oscillator" is better
> than that.
> A TTL can that is marked "4.096 MHz" costs about $2 and will make a
> square wave with a period of very close to 250 nS. Then they divide
> this down to the sample rate of 96KHz. In order to see a 250 nS
> jitter in the 96K signal the TTL can would have to "skip a beat".
> 250 nS is is a huge error and you don't get there with digital noise
> Chris Albertson
> Redondo Beach, California
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts