USB vs S/PDIF
USB vs S/PDIF
Submitted by dmsmit1 on Sun, 2010-09-12 03:23I just got a DAC1 USB, and am using the USB input. Would there be any difference in sound quality if I use S/PDIF coax input?
The PC is a Dell xps410 running Windows Vista, with Sigmatel "High Definition Audio" card, which is capable of bit depth and sample rates up to 24/192K/, although with USB selected as the playback device it's fixed at 24/96K.
Assuming that the data coming
Assuming that the data coming from the soundcard's SPDIF is bit-transparent (identical to the data contained in the audio file), then there will be no differences in sound.
Unfortunately, we have often found that sound cards alter the audio data, even when using the digital output. Our USB input is capable of streaming the data without alteration.
Therefore, we recommend that you use the USB input. The only time we would recommend using the soundcards digital output is if your computer is having difficulty with consistent USB data streams. If this is not a problem with your computer, simply use the USB.
Best,
Elias
USB vs S/PDIF
That's good to know.
Thanks.
although with USB selected as
although with USB selected as the playback device it's fixed at 24/96K.
This surprises me.
If it is fixed, all your audio is resampled to 96 kHz.
Vista SRC is not bad but it is not the best there is.
Normally you can choose various sample rates.
You might try the WASAPI driver to play the audio at its native sample rate.
http://thewelltemperedcomputer.com/SW/VistaTweak.htm
I’m afraid the Sigmatel has a bad reputation.
As a minimum, disable all enhancements
http://thewelltemperedcomputer.com/SW/Trouble/Dell.htm
USB selected as the playback device it's fixed at 24/96K
I apologize. I must have misread it. I checked again and there are three choices: 24/44.1K, 24/48K, and 24/96K. So if the source material is 16/44.1K, which is the best setting?
Native sample rate.16/44 is
Native sample rate.
16/44 is ok, 24/44 won't have any impact on sound quality (8 bits are padded) but if you use digital volume control you can chop this 8 bits of for 48 dB reduction without resolution loss.
DAC1 is 24 so use it as 24 regardless of the source
Native sample rate
So when the source is 16/44, the closest match available in the settings is 24/44.
I'm not sure if I fully understand your comment about volume control 48 dB reduction without resolution loss, given my limited knowledge. There seems to be some magic here - by means of a very simple operation (padding with 8 bits (lower bits right?)), I now have 24, so I can reduce the volume by 1,2,...8 bits and the resolution of 16 bits is still there. What puzzles me is why wouldn't all products do this, given that it's so simple?