[tomwu]

I got a HLLY SMK-II, with opt/usb DAC. It's pretty good pair with my D7000 as far as price concerns.

 

[bowei006]

@everyone

I think kingwa read this thread and heard the criticism.

http://www.audio-gd.com/Pro/Headphoneamp/NFB12/NFB12EN_Specs.htm

The specs page now includes filter and other not really actual amp and dac measurements but graphs are used.

Ive actually seen these graphs on other parts of the site before and thought of pointing it out. Anyway, the graphs are now also on the specs page.

 

[Tilpo]

@everyone
I think kingwa read this thread and heard the criticism.
http://www.audio-gd.com/Pro/Headphoneamp/NFB12/NFB12EN_Specs.htm
The specs page now includes filter and other not really actual amp and dac measurements but graphs are used.
Ive actually seen these graphs on other parts of the site before and thought of pointing it out. Anyway, the graphs are now also on the specs page.

Those graphs used to be there before in a more simple form, but this is definitely interesting.

Interesting. From these graphs I would go for filter 1 or 2, due to the lack of ultrasonic content at 96kHz input. Don't know the technical differences between minimum and linear phase, on the other hand.

 

[stv014]

Quote:

Don't know the technical differences between minimum and linear phase, on the other hand.

 
Linear phase impulse response:

Minimum phase:

 
The linear phase filter has the same delay at all frequencies, and the impulse response is symmetrical (has both pre- and post-ringing). It is the mathematically correct way of converting a digital signal to analog. On the other hand, the minimum phase filter has a frequency dependent phase shift, but no pre-ringing (some claim it sounds better because of that, although this is not really proven), allows for lower latency, and can be implemented as an IIR filter.

 

[Tilpo]

Linear phase impulse response:

Minimum phase:


The linear phase filter has the same delay at all frequencies, and the impulse response is symmetrical (has both pre- and post-ringing). It is the mathematically correct way of converting a digital signal to analog. On the other hand, the minimum phase filter has a frequency dependent phase shift, but no pre-ringing (some claim it sounds better because of that, although this is not really proven), allows for lower latency, and can be implemented as an IIR filter.

Thanks for that explanation.

I didn't know people used anything apart from linear phase in that case, but now I do know. I'll do some more research on the two, but if I'm not mistaken linear filters are more conventional, and will therefore probably provide better sound quality.
Meaning 2x oversampling with linear filter is the way to go when playing 24/96.