dcollins wrote on Tue, 01 March 2005 17:06 |
Ronny wrote on Mon, 28 February 2005 11:49 |
Although it's one of the better ways, RMS is not a 100% precise way of determining perceived gain. It's highly material dependent. For RMS analysis to be more accurate in determining perceived level, a ratio between amplitude and time must be calculated. What this means is that many RMS points would need to be calculated, not just one RMS reading on each song from top to tail.
|
First of all, you have to take into account the ears sensitivity to midrange and the equal-loudness contours. The straight RMS energy doesn't tell you how loud it sounds..... And why would it?
As always, I advise people to actively avoid looking at numbers when listening is all you need!
Now who will be the first to debunk the "RMS power" value?
DC
|
I'd be the first to agree with you that RMS doesn't give a perfect indication of loudness and good point about loudness contour, however 99.9% of musical instrumental fundamental frequencies reside in the mid-range contour of the Fletcher-Munson curve. Low being the B string on a 5 or 6 string bass at 31dB and the high C note on an 88 key piano or piccolo at 4.2k, so the vast majority of energy that you are going to encounter with typical music, already resides in the mid's and high mids. I think the main reason that folks get cajoled into thinking the RMS average is a good indication of perceived level is because they think that it represents a constant between peak and lowest level and it does to some degree, but across the whole spectrum and including freq's that are attenuated from human hearing, as you've pointed out. Regardless, what determines loudness is the gain that is heard before and after and what the ear is used to hearing before the evaluation. For example I have friends that live in the country and I can't hear their tv set when we I'm visiting. Their ears have become accustomed to the low level ambient sounds of the country and their listening volumes definitely reflect this, whereas the folks from the city have become accustomed to sirens, trucks passing and generally a much higher ambient background and their definition of loud is more dB's than the country folks. Loudness will always be subjective and completely relative to the listener, that's why it's hard to have a loudness meter, one size does not fit all, as we are dealing with psychoacoustics. It's the same with the smashed music, people get used to it sounding that way and they become accustomed to it, so when you try to leave dynamics in say for example some heavy metal, the young artists miss the pancake sound and complain. The hypercompression has actually become a requirement for that sound to be valid, to some folks.
Sometimes I think that we are better off not trying to police sound, because what sounds too compressed and ruined to you or I, might not sound that way to others. I'd rather people have the choice to squash or not, rather than dictate to everyone what we think sounds best. The biggest problem at this time is not that the engineers (even the mafia guys) don't recognize the problem, but the fact that most of the consumers that buy music aren't aware that the music would sound better with some of the dynamics left intact. Is it really that big of a problem? Maybe, maybe not, but it certainly has limited our choices of what cd's we are going to buy in the future. I didn't buy U2's lastest cd, because I heard lot's of ME's and AE's complaining about how squashed it was. I'm pretty sure that if I bought it, I'd be disappointed too, so I spend my time auditioning other new releases. The audiophile community is less than 1% of record sales, with the main purchasing power going to folks that don't know anything about hypercompression. The major's don't rely on our purchases and will continue to sell pancaked music, until the majority of consumers complain. That will only happen through education. It won't happen calling out stun gun ME's and it won't help addressing A&R types with our greivances, because they know exactly what they are doing. Sales rules, it always has, it always will and if we are going to gain any ground in ending the loudness wars, record sales will be the final battleground.