Dozuki Wrote:Actually Low passing a track at 16K wont give you much more headroom. most of the headroom is taken up by the Low end of the spectrum. It really all boils down to voltage. A sub takes a hell of a lot of voltage as well as amperage to get that huge cone in the speaker moving enough to produce sub bass. On the other hand, it takes very little volatage and amperage to make a tweet do its thing, and just a tiny bit too much and you could blow the little sucker. If you think about, the music you are making is directly controlling a sound system. This is why the db scale is used for both voltage AND sound. Basically they are the same thing.
Anyway, point is, the highs take up VERY little headroom. Good needles will go to 20K+ and most fair ones to 18K. Also realize that when a company says their needles cover from 20 - 18K Hz this is typically the F3 point (where the signal is 3db down from the rest of the signal). It doesn't mean that the needle won't play higher frequencies, just that they are 3+ db lower than the rest of the audio spectrum. If you did lowpass your track at 16k you would loose all the 'air' to all the instruments and the track would sound duller, even on vinyl. Most people, depite not hearing above 16k, would also definately say the track sounds different. Same thing happens with bass. You may not hear 20Hz but you can definatly tell a difference with and without it.
Anyway, don't mean to rant.
-D.
I agree with everything youve said there, but at the same time I stand by almost all I said...
But that is my fault for not qualifying my earlier post better.
Firstly, the situation I was talking about is specific.
Specific to making drum and bass records primarily aimed at the dancefloor.
I was incorrect when stating the roll off of most carts is at around 16Khz.
However, remember this is focused on reproduction in clubs. Most clubs still use 500 ALs which roll off at 17Khz, and actually more steeply than by just 3dB. But 3dB is still significant headroom in itself.
Now regarding this 'air' in the high end of the spectrum - I am all too aware of this - For the last six months or so, I've been solely writing, recording and mixing hip hop, where capturing air in the vocal performance has of course been important. For most purposes I run a good vacuum tube condenser mic through an avalon 737 whose eq section actually features a 32Khz position !
My point here of course is that the likelihood is that many of the samples being used may not contain much of that high frequency 'air' from the original recording. If the sample IS rich in frequency content above 16 / 17 Khz then is that content actually part of the original material? Was it added by a piece of noisy equipment in the chain at the time of the original recording? How many generations old is the sample if it was found on the internet for example? How good is the cart you used to sample with? Are your audio cables from the deck high quality, low noise? Have you run the samples through any noise, click, or crackle reduction processing, in hardware or software ?
In these cases - which is common - cutting at 16 / 17 Khz will not be removing any air, actually heard, perceived, or otherwise - but it will give you some extra headroom and clean up your mix. Maybe this effect is very small on a single sample, but over many, we sum the benefits.
Yes, still the benefits will not be as great as cutting sub below a given threshold, for the very reasons you gave, but that point had already been covered and I was attempting to add to the discussion.
And the fact does still remain that as we age, for most of us anyway, our hearing deteriorates, and this mainly happens by loss of our ability to hear the upper end of the frequency spectrum. In these cases we will NOT be able to perceive the difference that you refer to.
Unfortunately, this happens to most people alot younger than you might expect. Fortunately for me, most of my hearing is still intact, bar a very slight dip at around 16.5Khz
Anyway, bottom line is to always remove any frequencies that are genuinly not needed, on a per sample / track basis, and yes - evaluate each sample / track diligently and make the decision about what frequencies you can cut, and what you cannot.
I think this strategy is a good one to address a large part of the problem that is poised in the original question - how to get the tunes loud.
Unfortunately, the preference for loudness in this genre of music causes many mastering engineers (either of their own accord or under direction from the client) to squash the hell out of the tune with agressive brickwall limiting. I find it harder and harder to listen to much d&b for any length of time due to the relentless fatigue cuased by lack of dynamic range.
Id recommend everyone read Bob Katz' Book 'Mastering Audio - The Art and the Science'
Its enlightening....
Merry xmas