I am using a computer as a source for my music and wondered if a bit perfect setup (under windows) is also needed in case the DAC has built in buffer and re-clocking?
The two aspect you mention are different properties, both in need of being in control to reach the most correct reproduction of the recorded information.
Bit perfectness is about not changing a 1 to a 0 aww.
buffering/re-clocking is about trying to take care of timing aspects and more specifically, seeing to that each word carrying music data is converted in the exact wanted moment in time.
So one is about timing and one is about information. Both need to be cared for to make a good DAC.
//
Bit perfectness is about not changing a 1 to a 0 aww.
buffering/re-clocking is about trying to take care of timing aspects and more specifically, seeing to that each word carrying music data is converted in the exact wanted moment in time.
So one is about timing and one is about information. Both need to be cared for to make a good DAC.
//
Only a really *awful* system would fail to be "bit perfect"(an immensely stupid term, IMO). Getting anything close to "timing perfect", in a genuine sense, is the difficult & *critical* aspect, sadly the aspect that most people, esp. computer sound gear designers, are all but completely ignorant about.
Well, there's always a way around it. At the moment I am using foobar2000 with Kernel Streaming plugin. There's also OpenELEC (recently most of their dev team left and created LibreELEC) which is dedicated Linux release for Media only.
Timing is only critical at the point of conversion, hence re-clocking PLLs etc. in modern DAC chips, spitting out a bit of audio data from any computer based source is not exactly taxing...
What is a "computer sound gear designer", DACs tend to be designed by EE's who are quite aware of timing with a digital signal transfer...
What is a "computer sound gear designer", DACs tend to be designed by EE's who are quite aware of timing with a digital signal transfer...
I think most of us know that re-clocking on DAC is not enough. The timing is still greatly affected all the way from the source.
Timing is only an issue at the point of conversion on any ADC or DAC... probably all modern DAC devices can re-clock well enough these days, whatever your average audiophile may think, digital audio is pretty trivial these days. Would suggest some reading from reliable engineering based sources on this site there are some excellent ones from people who really understand what is going on, learn from them not some myth monger...
Thanks Ken!
stephensank, agree, but many who uses Microsoft PC as a source will have their bit-flow manipulated by the operating system if they don't care to make the right configurations of the PC. But for normal CD players and Squeezbox users, the bit correctness is not a problem.
The chance of a 1 turning into a 0 in say in; RAM, processor, ethernet cable, router, switch, NIC, SSD, hard disc and in a correctly configured OS, is near to zero.
All digital EQ and digital filters change the bits of course but still, it would at least be good to present the original information to these functions to start off as well as possible.
Double agree 🙂 As for timing using a computer based system and if indeed a digital word would be delayed say 0,1 ms (which is an eternity) in; RAM, processor, router, switch, NIC, SSD, hard disc and in a correctly configured OS, this would have zero impact as it would either be taken care of any buffer on the path or manifest itself as a distinct audible click. This error must not be confused with the timing aspect discussed above which is in the order of 10th's of pico seconds deviation from ideal moment which is deemed to be necessary for a pleasurable audio reproduction when converting digital words (16 bits etc).
//
- Status
- Not open for further replies.