Sarcasm and loudness aside.
Why is referencing a mix against a commercial song a bad idea?
I honestly was not being sarcastic when I gave my response. It's an awful idea because your average commercial CD today sounds awful. And most of the reason they sound awful is because their RMS is simply pushed too hard in the mentality today that loudness is more important than quality.
Think about it like this; travel back to 1975. You're looking to make your own line of automobiles. Would you really want to use your average Ford, Chevy or Chrysler made back then as the bar-setting standard to compare against for anything other than what you need to do better than? The only standard they would set would be how NOT to set assembly line quality control. The same is true with the production values used in your average (not all) commercial recordings of 2009; unless you want to mix and master another Ford Pinto, Chevy Citation or AMC Matador. Most of them are not setting a standard to be admired and pursued when it comes to compression and loudness.
If I make a song that I vision sitting inbetween track ''x'' and track ''z'' in a dj's set is it bad to use those as a reference point?
It's wrong, IMHO, to use anything but the actual sound of your own mix as the reference. There's three important points here:
First, every production mix, "pro" or amateur, has a point somewhere along the compression/loudness spectrum where it's going to start sonicly falling apart. That point on many commercial mixes is often going to be much higher than it is on most amateur mixes because both the talent and the gear on both sides of the glass is greater than it is for us amateurs. (Even then, they still stupidly push the mixes past their breaking point.) But when you record in a sonicily inferior room to a pro studio, with entry-level gear and amateurish control over the gain structure throughout the recording and mixing process, trying to push things just as hard in mastering is going to wind up pushing the mix even further beyond it's breaking point because it's breaking point is lower.
Which leads to the second point; USE YOUR EARS to determine that breaking point and mix and master to that point or below. Whatever tracks may surround yours in a playlist is irrelevant in that regard; that point will remain the same within your song regardless of what surrounds your song. So how do you handle the playback volume difference? That's point 3:
the difference in volume is IN PLAYBACK. It is, therefore, IN PLAYBACK that it should be addressed. The simple way to do it is by using the volume control. For DJs, their job is to pre-cue the upcoming songs in their mixes and adjust the playback volumes accordingly so they are getting an even playback. That's half the reason they are even there. This is true both of radio DJs and live club DJs. That's their problem, their jurisdiction, and their task to take care of. And they can do it by simple playback gain control that will have no effect upon the sonic quality of your mix. If you try to take care of it on YOUR end, you have to do it by means other than simple gain - i.e. compression, limiting and so forth. Unlike simple gain, those processes are
destructive in that they change more than just the volume, they change the sonic quality. So let them worry about it; it will serve you, them and your mix better that way.
What about those DJs that are phoning it in and won't play over-compressed mixes because they're too damn lazy to do their own job? It's their loss. The great DJs have two advantages over those slack-assed jokers in that they can hear beyond the volume and recognize great mixes regardless of their intrinsic volume, and will recognize that many non-squashed mixes actually sound better. They won't be afraid to play your mix, but rather will go out of their way to pre-cue it and play it, and will wind up with the kind of playlist that puts them a step above the slackers, and helps make them the DJs taht are the most popular and the most in demand.
I was just after some thoughts on how some of you proffessional guys satisfy yourself that a song is finished.
The oft-applicable answer is "when the client says so"; which is the #1 reason why today's mixes are over-squashed. The engineers usually don't want to push them anywhere near that far themselves, but they gotta do what the client wants if they want to stay in business. And often the client couldn't tell the difference between a good mix and a woodpecker banging on a metal tree trunk, they just think "mo' louda' mo' betta" and little else.
But if and when the mastering engineers had their druthers, they usually let the songs on the album tell them what to do (let's not forget that "mastering" a single song is not really mastering, but rather usually needs to be put in context of an album or EP.) As mentioned, each mix has a volume it "wants" to be at and a sound it "wants" to have, and the mastering engineer, left to his/her own devices, will usually use their superior suite gear and superior critical listening skills to find those intrinsic qualities within the mixes and get them to comfortably link and flow together as an album. When they reach that point, they are done.
G.
) not a professional, so I probably need the occasional sanity check more than someone who does do this for a living.