Why I cannot hear the difference: DSD256 vs 44/16?

You anyway need to have a volume drop of at least -3 dBFS when you are using upsampling. Otherwise you will trigger the limiter and get forced to lower volume on the fly.

And should have it too even without, because most DACs will heavily clip inter-sample overs.

It is not destructive at all, in fact, setting software volume to 0 dBFS is destructive because it almost guarantees that you’ll have clipping/limiting. When you use Roon’s volume control with HQPlayer, Roon is adjusting volume at HQPlayer side.

In addition, many DACs have lowest THD+N at -10 dBFS level. And using a bit lower level allows you to turn up volume on amplifier, meaning less attenuation and noise there.

So there are many benefits of software volume control. Assuming the implementation you are using is good, and you use proper dither/noise-shaper for PCM outputs. (Doesn’t apply to DSD outputs, because they are noise-shaped always.)

1 Like

Appreciate the input Jussi, thank you! :slight_smile:
But, my ears (?) seem to prefer non upsampled audio (from software based solutions) in most scenarios i can muster. (Which means that software volume control, in my case, only adds insult to injury)

This seem to mean that good mastering and mixing of a file is not even close to 0dBFS, to make DACs perform at their very best also. Is this a contributing factor that many of us seem to dislike heavily compressed (dynamically) and loudness wars casualties perhaps?

Anyways, just my PoV. We all have different paths to audio pleasure! :wink:

Nothing we know about human hearing or digital sampling theory suggests you should be able hear a difference. I wouldn’t worry about it too much.

2 Likes

Corrected that for you. :wink:

1 Like

room & speaker positioning, mastering of the recording, amp resolution, DAC internal implementation, filters used, mood.
Also the comparison & listening methodology may be a reason
Being just upsampling, there shouldn’t be a big difference in any case.

Well, if you prefer the resource constrained hardware version, that is of course fine too.

Yes, unfortunately pretty much all RedBook is driven to 0 dBFS and very frequently into digital clipping.

Yes it is indeed. Some of that can be fixed with suitable upsampling filters and leaving some headroom as part of that process.

1 Like

Interesting. Do you have some more information on this? :wink:
Based on what you know? Because based on what I know certainly there are objectively measurable differences one could hear. I don’t make statements about what any particular person can or cannot hear, when there are clearly objective difference.

Certainly for example apodizing digital filters can fix various problems in lot of RedBook content, especially ones recorded using more recent ADCs.

Information on what? If you’re claiming humans can hear higher frequency sound than 20kHz or greater than 96 - 120 decibels of dynamic range, the onus of proof is entirely on you.

You started making claims, so I expect you to come with a proof too. And no, you don’t even need to go beyond 20 kHz and such dynamic range.

I have learned during time, that the shortest path mostly sounds the best
In my case it was Mac Book Pro with SSD and JRiver Mediacenter installed connected with USB to the DAC

this gives the best possible sound quality
(Pure Audio Project often also uses this setup while demonstrating their loudspeakers on hifi shows)

You also could try installing Roon this way and see if it makes a big difference

This was my view before I started down the digital audio rabbit hole. The best component is the one you can leave out. Keep capacitors out of the signal path. Short connections are better than long ones. For a while I even experimented with passive preamps.

I’m still pretty much of that view for analog gear, although these days I prefer an active preamp. With digital, however, I’ve strayed from the path. In my system I found adding a network device (microRendu) improved SQ over direct connection to a computer. Then I added a separate music server and later a further device between microRendu and DAC enabling I2S connection instead of USB.

But I think it is a good idea to simplify things every now and then to check that any added optional components are still providing a benefit. I don’t trust my ears very much. I think they prefer a rich euphoric sound rather than something truly neutral.

3 Likes

I notice on the screen shot above that it recognizes a Denafrips DAC. How did you do that?