HQ Player

[PeterSt]

2 hours ago, rickca said:

I would also like to invite comments from @PeterSt and @mansr.  I will fall off my chair if you guys agree on anything. ?

 

Rick, you can relax and sit back. 

 

These days I tend to change my stance or technical merit (would I ba capable anyway), and I think I am not alone on this :

 

As you may know o.a. I use the non-ringing Arc Prediction as part of the filter as a whole, or stand alone (as how the original Arc Prediction has been set up). What happens in practice with this, is this and I try to be as objective as possible because I can be regarded a consumer of "a" product, which is only coincidentally my own :

 

Frequently I engage one of the filters which imply a better frequency domain (less distortion there) which I can do for two separate reasons (separate because on-day-this and the-other-day-that) :

1. I have the hunch to desire for more high frequency (less roll off, depending on the filter chosen);

2. I might have the idea that the zero-ringing filter (Arc Prediction) sounds too harsh.

 

Ad 2.

I already know that this is only because I know how the filter behaves and the higher the frequency (audible spectrum) the more THD. So this is theory.

 

Ad 1.

That too is theory, although it usually audible helps right away.

This is dangerous because the now less rolling off part is in a such high frequency reange that I am sure that I am not able to receive it anyway. However, this is merely about the square response of lower frequencies, like my these days some times mentioned "electric butterfly" sounds (could be SandyK's moths).

 

I think since the G3 incarnation of our DAC this never ever holds any more. It is 100% always right at the next day's listening session that from the ever so slightly more ringing filter, I perceive sheer distortion. It is only that I can't determine what distortion that exactly is, but it should be about the mixing of again the higher frequencies because of ringing (it's actually smearing). So I put back native Arc Prediction (totally zero ringing and still only 0.00063% THD+N at 1KHz for 24 bits - don't ask about 20KHz ;-)) and all is super (really super) smooth right away.

Directly or more indirectly the iFi guy tells similar (I think).

 

Notice that a difference will be that I don't apply "just less filtering of less taps etc." but that the filtering means is different to begin with because Interpolative. That next whatever masking could be in order because of higher frequency THD of that, would be something for blahblah commercial stories I can't make up (but which could exist and be true). My personal commercial story could be : ringing can't be masked and the "THD" coming from that has to be of another kind (it buzzes to begin with).

 

Any other story for mine !

 

 

 

[PeterSt]

Re: NOS1a and HQPlayer on Win 10;

Shreekant, That post and a few more underneath it.