In my day job I build mathematical models (well someone has too!) Recently I've been knee deep in NMR (Nuclear Magnetic Resonance) spectra of polymers. The gist is this : we take a spectra of the polymer process via the NMR every 15 minutes. Each spectra contains 2,000 points. If I print out these spectra, it really is quite hard (impossible) to tell them apart whilst the process is working well, visually. However, mathematically it is possible, and even though the difference are slight they can (and are) used to control the process.
Back in the audio world, some things sound the same, but are different, these differences are slight, but they are there. Some people have better sensors (ears/amps/headphones etc) than others and they can hear difference that others don't hear. Mathematically we can look at the audio data and see that different chips produce different signals, these differences are small, but they are there. Engineering attempts to reduce this 'uncontrolled variabilty' to a minimun, but there is usually a cost curve to use, at what cost is a desired result good enough? Human ears are difficult things to use as sensors, because they are affected by what they are sensing, the time of day, the mood of the person etc.. all of which are also uncontrolled variables; so listening to the same sound on different days will not result in them 'sounding' the same! (My rig always sounds better late at night!) The question is this; which of these two sources of variablity has the biggest effect, variation in the actual signal or variation in the listeners? My philosopy: Minimise the variation that we can (better cables, stable power supply, low jitter, low noise etc) and then whats left we can modify with a glass or two of whiskey. -- BigTony We're Only In It For The Music! www.zappateers.com ------------------------------------------------------------------------ BigTony's Profile: http://forums.slimdevices.com/member.php?userid=10638 View this thread: http://forums.slimdevices.com/showthread.php?t=56712 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/lists/listinfo/audiophiles
