Upload & Sell: Off
| p.3 #1 · RAW compression in D200 is a POS |
Identical scene and photometric exposure, otherwise a comparison is void, meaningless and totally without statistic merit or validity.
That's why ABX testing exists, mainly within audio, but also when you compare human perception evaluation of image quality in a scientifically valid way. We used to do that when comparing different screen pattern types for print.
It's quite hilarious how few "audiophiles" actually hear the difference between a good quality mp3 , a high-rate AAC or a raw uncompressed signal, even on a 3000$ set of heads on a good headphone amplifier. They're usually very sure of themselves in the beginning, but after a few ABX rounds (when they still have the same hit/miss rate as an average random generator) confidence always tumbles.
Without derailing the subject too much, I can't help but point out that reliance on double-blind testing for music playback equipment should be treated with caution for the mere fact that listening to and enjoying music is a different perceptive activity than listening to or measuring sound. The primary issue is that the enjoyment of music is an entirely subjective phenomenon.
With audio equipment, some changes within a system are immediately perceptible while others may not manifest themselves so obviously. Headphone listening also complicates matters because it is far removed from the reality of listening to music in real space, regardless of the equipment's cost or sophistication.
One of the first casualties of lossy audio compression is the reduction of unique phase information between channels, which provides aural cues that are much more relevant when listening to music in an actual room via loudspeakers.
I have learned over the past 20 years to withhold judgement for any particular component until I've had a chance to listen with it for a considerable length of time. Only until then can I really appreciate whether it does justice to the way in which I enjoy music. While sound can be measured and quantified according to the abilities of test equipment, music cannot be measured or quantified in any significantly meaningful way. I don't have magic ears that are guaranteed to make an informed judgement based on short listening sessions, so there's really nothing peculiar or hilarious about not being able to form an opinion based on an ABX test. But allow me to use a given system or process for an appropriate length of time, and I'll be able to say whether it's an improvement.