Originally Posted by
bigshot 
A clean volume pot is easy. Every audio product going back to the dawn of radio has had one. If we can put a man note moon, we can put a clean volume pot on a headphone amp
what an epic quote! probably the first time somebody compares volume regulation to moon expeditions 
but seriously, clean from what? probably you do not realize it but any form of volume control, either in the digital (before the d/a conversion) or analog (after conversion, typically resistors) has its drawbacks and the only question is how big is the data loss involved.
Here is the interesting thing here. You look at the charts on that page and reject it because you can't believe one can rate higher than another. I look at the same charts and say to myself, "those two things sound exactly the same". How can we come to such different conclusions?
I've put the numbers behind that test into the context of human audibility. It's a waste of time to go into the minutiae of testing methodology to skew things a few points one way or the other when the scale of the distortion is so infinitesimal. Even if he's wrong by a bit and the amp you like rates a bit better, it still doesn't matter because with these numbers, you're talking about angels dancing on the head of a pin.
Inevitably, this is what happens with people who value their totally non scientific subjective opinion over one that has testing evidence behind it. They question the methodology, argue over gnat hairs' differences and totally lose sight of what is audible. They refuse to let go of their opinion that is based on no testing at all and reject a controlled test because of one meaningless variable or decimal point.
I know what the specs are for human hearing and I apply those to the specs of equipment I am interested in. Usually with digital sound reproduction, there isn't any reason to go any further because the specs of even a $25 WalMart Coby DVD player outstrip the range of human hearing.
No. I reject the charts mainly exactly because they do not allow me to duplicate the experiment they supposedly describe and verify the results myself. Karl Popper, among other things, argued that any results claiming to be scientific have to be falsifiable. How would you potentially confirm or refute empirical results impossible to reproduce because their authors did not care to describe them in enough detail? What does +2 dB mean exactly, what was the SPDIF transport driving the NFB? Did they adjust volume levels and how? Was the operating system and playback software the same or not? Without answering these questions the results are useless, there are far too many interferring variables. And please note that I am not undermining their honesty.
And it does not really help me that I never came upon a cell phone that was claimed to be able to drive demanding 300 ohm headphones. Did you ever read a single credible review of a mobile device comparable to dedicated headphone amps in driving demanding phones? If the Galaxy S is so great, why did not Samsung marketing pick these capabilities and target audiophile market? At the very least I would expect some numbers on headphone output power performance. Then, outputs on HTC Desire and HD2 that I tested are a disaster, they perform poor even on low impedance headphones like Ety HF5. The idea of using HD650 and alike is ridiculous and seems more of a joke. Is Samsung so much better than HTC? Maybe, but I see very little credible evidence so far.
I would rather not seriously comment on the thing about $25 WalMart DVD outstriping the range of human hearing. Let me just ask if you compared the mentioned DVD to any modern, dedicated DAC priced over $500 heard no differences. Please, let us know what the DAC was 
Edited by axw - 6/27/11 at 7:07pm