Afternoon all! Thought I'd chuck these opinions in to generate a bit of controversy (just for fun, but... )
I would certainly expect a well designed piece of audio gear to exhibit good laboratory-type measurements. However I remain sceptical that such measurements always reflect the real world performance a music lover is seeking - let me give some examples: Low Total Harmonic Distortion measurements are reassuring to a degree, but most people find odd harmonic artifacts much more objectionable than even harmonic ones (tube amplifier lovers seem to actually like a bit of even harmonic intrusion to give a euphoric rather than clinical insight into their recordings). I'm not advocating even harmonic distortion per se, I'm just saying that of two pieces of kit with similar THD measurements, the one with the higher odd harmonic components is likely to sound less musical unless the designer has managed to get the overall THD to a spectacularly low level. When I was a nipper, much stress was placed on the Damping Factor of an amplifier, which if I recall correctly was defined as the ratio between the (presumable relatively stable) output impedance of the power amplifier to the nominal (and often highly volatile) input impedance of the loudspeakers in use, which tended to be given as exactly 8 or 4 ohms somewhat oddly, given the variability of loudspeaker impedance with frequency. This measurement seems to have fallen out of favour, which is probably for the best. In the old days loudspeaker manufacturers would go to extraordinary lengths of driver combinations and elaborate (passive) crossover designs in order to get a relatively flat frequency response from their designs *-when measured in an anechoic chamber-* (charts of these frequency responses were freely published to sell the designs, little mention was made of THD statistics... ). Sadly few users had one of these chambers, so they ended up with a sound principally dictated by the size, shape & contents of their listening rooms, whilst also presenting their power amplifiers with difficult reactive loads and a loss of close control of the actual drivers themselves. It is hardly surprisingly that few of such legacy loudspeakers that have survived sound very musical when compared to a halfway decent modern loudspeaker. When dealing with the merits of PCM sampling rates much stress is placed on the implied upper high frequency limit in accordance with "Nyquist theory" (which was fully in place by the 1950's, long before its subsequent application to digital audio - the Wikipaedia article on "Nyquist-Shannon sampling theorem" is very informative, if not the easiest of reads) & again I fully accept the frequency response upper limit & the need for severe filtering at or below half the sampling rate to prevent the intrusion of audible artifacts not present in the original signal being introduced by the decoding process. However, I'm not entirely convinced that this is the only concern when what we are doing is seeking firstly to encode (ADC in recording studio) and then decode (DAC in your music system) a complex & continuous analogue waveform (let's call it "music") using any fixed quantity of 0's & 1's and ANY sampling frequency. The only real question is whether the reconstruction is sufficiently close to the original to fool your brain. As a analogy, there are many real numbers of mathematical significance that can not be completely represented in the decimal format to any finite number of decimal places - they're called irrational numbers, and there are an infinite number of them between 0 & infinity (infinity is a slippery mathematical concept - some infinities are bigger than others... ), but pi and the square root of 2 will serve as familiar examples. Obviously in practice a sufficiently accurate rational number substitute can be used for calculation purposes - what is sufficiently accurate will depend on what you're doing. What I am saying is that a continuous & dynamic analogue signal, like er, "music" may be harder to successfully capture than many people assume. If all DAC's sound the same, then why should all the hardware in the Squeezebox family from the original Classic onwards not sound the same? If that were the case why have so many members of this forum acquired a Transporter or a Touch? Unless someone is going to say that poorly designed DAC's were put into the earlier models by Sean Adams who seems to me at least to have had a pretty good idea of what he was up to as far as design was concerned... Any computer program is only as good as its designer's a priori understanding of the job it must perform. When I was younger & had better concentration, it used to amuse me to beat my computer at chess in all manner of ways (childish I know, but fun). The issue here was that the programmer, who was almost certainly a much better chess player than me, had been faced with a rather difficult problem - once his 3, 5, 10 or whatever (depending on how long you were prepared to wait for the computer to select its next move) decision tree had been completed, he had to tell the computer how to select the most desirable position for its pieces to be in, by some weighting system or other. In other words, he had to define what a "general" good chess position was. I didn't have this problem, since I was only ever playing the one game in front of me rather than worrying about all possible games and positions that might arise. Also the program didn't itself understand how to win at chess - it was always waiting for me to make a mistake. As long as you gave it enough time, it didn't make tactical errors (giving pieces away cheaply) but it had no concept of strategy, so you could with some effort gain a sufficient positional advantage that it was unable to survive a careful attack. I know that computer programs have got better, & some of them now attempt to self-learn (difficult when you consider how many different possible games of chess there are) &/or use "fuzzy" logic, and of course they now have access to more powerful processors & more memory enabling them to run longer decision trees of all possible moves for a given number of moves ahead more quickly. I also know that my ability to maintain my concentration has reduced with age. So I don't find this pastime as much fun anymore (who likes to lose?) because the chance of me making a mistake before achieving a strategic advantage has increased markedly. It's hard to get the upper hand when you don't have as many pieces left as your opponent, unless you're Paul Morphy! These days I prefer to play backgammon, although I have to turn down the wick on my program since it's actually better than any human player if you give it free rein. The professionals use it to conduct a post-mortem on their games. Of course, there's also an element of luck in backgammon which is not the case in chess or in its more difficult stablemate Go, which is played mostly though not exclusively in the Far East. Edward Lasker who was a chess world champion later devoted himself to Go which he recognised as a greater challenge & he was responsible for bringing it to the attention of the Western world. I believe it now has more active players in the world than chess, although I'm happy to be corrected on this because I can't be bothered to check just now. The fact that super-computers are now capable of beating the world's top human players at both of these games is a truly remarkable demonstration of the power of our technology, and gives me pause every time I re-watch Arnie in the Terminator series of filums... I'm not saying that we're incapable of designing ADC - DAC pairings good enough to fool our brains. I'm just saying that I'm not sure that all implementations currently reach that standard, & by implication I am somewhat sceptical of the claim that "all DAC's sound the same" when dealing with "music" rather than a regularly-shaped test tone, whether it be a square wave or a sine wave, is really the case... FWIW, I'm equally sceptical of the true provenance of many of the "hi-res" recordings currently available & suspect that few of them come close to achieving the potential quality of their purported format - I might get a bit of agreement on this point at least. I've put my firefighter's costume on in preparation, lol Dave :) ------------------------------------------------------------------------ Golden Earring's Profile: http://forums.slimdevices.com/member.php?userid=66646 View this thread: http://forums.slimdevices.com/showthread.php?t=106914 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
