DAC sample rate and bit depth?

[whotheff]

I recently bought Topping E30 II DAC. It has 32Bit/768kHz chips inside. Most of my content is 44.1/16 and some 96/24 and a few 192/24.
I'm wondering: In pure USB DAC mode (no pre-amp) Windows setting affects the number displayed by the DAC (currently 192/24). However, If I play a 44.1 file, DAC displays 44.1.
So my question is:

1. Does Windows sound setting matter, if the DAC switches between sampling rates to match each song's encoding?
2. If I want the best sound quality out of this device, should I select 32/768kHz or a lower number, matching my biggest files (192/24)?

Thanks!

 

[cpt_paranoia]

Don't let Windows mess with your samples; send them direct to the DAC, and play them at the native rate. The DAC will sort out any upsampling it needs in its conversion mechanism.

If you're using the DAC to do volume control, let the DAC do it, not Windows. Hopefully, the DAC driver supports volume control in this way.

32-bit DACs are nonsense; it's just the precision used in the filtering. Even 24 bit dynamic range (144dB) is just on the limit of the noise floor. 32 bit dynamic range is an impossibility.

 

[Yamaki]

1. No
2. No, because it doesn't matter due to the device playing each file in it's native format and specification.

 

[whotheff]

1. No
2. No, because it doesn't matter due to the device playing each file in it's native format and specification.

Last evening I tried pushing it to max frequency possible at 24bit - 384kHz just to check if I can hear any changes. Bass became weaker and sound became more.. "airy". However, I have the feeling sound lost punch.

 

[xdd]

Seems from reading around that higher degrees of oversampling give a less dynamic sound. My experience too which is why I've stuck with NOS all these years.

 

[Evonimos]

I feed 24bit 48kHz signal to CS8412 that drives a TDA1541A.

I know the TDA1541 is 16bit.

However, to my ears (don't ask me how) this bit/sample rate sounds marginally better than 16bit 44kHz.

 

[xdd]

There's possibly a sweetspot where moving the DAC image frequencies further out of the way is more beneficial than the slight reduction in dynamics is harmful.

 

[dewdude]

Upsampling was handy when we had lousy DAC builds and limited connectivity. I have a USB 1.1 DAC that if I want to use 96khz I have to disable input and specifically flip it in to 96khz mode. The problem with this is the TI controller won't do less than 88.2 in 96 mode....so I had to force upsample to a minimum of 88.2. The SB Audigy2 had a special "p16v" module block in it's infrastructure that provided 96/24 decoding (and 192 with cyberlink dvd-audio auth); and it was the same deal. I had the drivers which let me bypass the internal DSP and "wire" the p16v directly to the output...but it had to be 96khz. Some chips ran at a fixed rate on top of this.

None of this is a problem. Modern DACs have improved where the old fringe tricks of maxing out sample rate to avoid conversions no longer matter.