01-01-2012, 02:15 PM
Yes, please don't let my ramblings put anyone off. I've studied audio all my life, and I've never known why I've found it so very absorbing. What I will say is that while it's one of my specialisms, I'm really no expert - for as long as I can remember, I've been at the "more I learn, the less I know" stage. We will never have definite answers to many of the questions - and for me, that's part of the appeal.
I've found that it is best to keep an open mind, but the more science you learn, the better. If nothing else, it allows you to put the claims into perspective. After all, it is possible to measure differences in capacitors, but you need highly specialist measuring equipment, and the changes you might observer are often >100dB down on the test signal. When you learn the the ear has a mechanism called the "masking effect", where similar frequencies are masked if one is louder than another, and that the difference between level might only be 12dB or so, it makes claims of audibility of 100dB separation highly questionable indeed...
One good thing that's come about as a result of audio data rate reduction schemes in recent times is a lot of new research on human hearing. Actually, it's proved that a lot of the research done in the 1930s was surprisingly good. But increased understanding of the hearing systems can only help with these sorts of questions. I've always argued that the hearing system is a great deal more complex than vision - a lot of the cleverness with vision comes at a higher "level" within the brain, but the basic "low level driver" is pretty simple. By comparison, the "driver" sitting between the cochlea and the subconscious parts of the brain is much more advanced. For example, how do we locate sounds? We don't find ourselves weighing up the relative level and phase of the two sets of impulses coming from the cochlea. How can we tell that sound is behind, or above, or below - with only two ears. The "driver" works all that out, by constantly assessing the acoustic properties of the room or space, and by judging the HF loss that our outer ears imposes. It's completely magic, and fun to watch new-born children developing this skill
All the best,
Mark
I've found that it is best to keep an open mind, but the more science you learn, the better. If nothing else, it allows you to put the claims into perspective. After all, it is possible to measure differences in capacitors, but you need highly specialist measuring equipment, and the changes you might observer are often >100dB down on the test signal. When you learn the the ear has a mechanism called the "masking effect", where similar frequencies are masked if one is louder than another, and that the difference between level might only be 12dB or so, it makes claims of audibility of 100dB separation highly questionable indeed...
One good thing that's come about as a result of audio data rate reduction schemes in recent times is a lot of new research on human hearing. Actually, it's proved that a lot of the research done in the 1930s was surprisingly good. But increased understanding of the hearing systems can only help with these sorts of questions. I've always argued that the hearing system is a great deal more complex than vision - a lot of the cleverness with vision comes at a higher "level" within the brain, but the basic "low level driver" is pretty simple. By comparison, the "driver" sitting between the cochlea and the subconscious parts of the brain is much more advanced. For example, how do we locate sounds? We don't find ourselves weighing up the relative level and phase of the two sets of impulses coming from the cochlea. How can we tell that sound is behind, or above, or below - with only two ears. The "driver" works all that out, by constantly assessing the acoustic properties of the room or space, and by judging the HF loss that our outer ears imposes. It's completely magic, and fun to watch new-born children developing this skill
All the best,
Mark
