Larry Mal wrote:Right, but persuading one's self that Digital Performer sounds better based on this test only makes you the same as all the myopic people you complain about on Gearslutz, only myopic about different software.
That's what I'm trying to tell folks. Don't like the Gearslutz Pro Tools bullies and their attitude towards Digital Performer, based on nothing except opinion? Don't become them in the opposite direction. That's why this site is better in a lot of ways to that one, because it has a more knowledgeable user base over all. I'm surprised to be reading this here.
Larry in principle, I agree with your sentiments.
My personal estimation of this test is that, although there may well be problems with it, the result probably isn't due to groupthink. There are too many unknowns being tested. If this were A vs B, then possibly or even probably. But I think 5 unknowns is too many for the homogeneity of results; you can't attribute that reliably to a dominant personality signaling "this is best" given hours of random double-blind testing of five different unknown objects.
At the very least, what cannot be explained away is that there was a perceived difference, and it was detectable reproducibly. Regardless of whether it was detectable by all panel members independently or was the result of the amplified opinion of a subset of them, the difference was detectable, repeatedly, under double-blind conditions.
IF there were a problem, the place to look in this case would likely be the signal chain. RG describes:
All installed on 5 equal Macbook Pros running optical O/P to a switcher to two Genelecs.
The devil really is in the details, and the key term is "switcher." The only absolutely foolproof way to be sure there is no difference in the output resulting from circuitry is for there to be nothing
at all between the DAC and the speaker output that could be distinguishing. Even parallel sets of analog components going from the DACS to the output could be dispositive. The only absolutely-free-even-theoretically-from-any-analog-artifacts chain would involve having the test administrator walk up and plug a single optical cable into the output of each Macbook in series. Of course, since the goal is not to be able to see the switching, this would make the testing very laborious.
This is exacerbated by the fact that human beings are incredibly sensitive to exactly the kinds of L/R comparisons that we refer to when we talk about stereo imaging and spatiality in audio. Oddly enough: back to Gearslutz, where Paul Frindle, one of the designers of the SSL E and G consoles, talks about that sensitivity and how it was revealed under remarkable circumstances:
http://www.gearslutz.com/board/music-co ... 5.html#144" onclick="window.open(this.href);return false;" onclick="window.open(this.href);return false;
Having said all of this, there are definitely reasons to think that DAWS might output different signals from the same input. Most people here know that DP uses 32-bit fp precision throughout the mixing path and 64-bit precision during the mastering stage. Most DAWS do something analogous. That means that getting to 24 bit output requires wordlength reduction, which requires dither. It is well established that different dithering strategies do result in different sounds in very low amplitude portions of the audio. Whether that, or variations in analog components in the chain, or some other math-related difference, could account for the perceived difference during the test, I don't feel credentialed to say.
Nothing, however, is going to reduce my enjoyment of the monkeywrench this has thrown into the DP-hating goonhouse over at g-slutz.
ok jeez, i've spent way too much time on motunation for the past couple of days...