Bristol_Jonesey
That all depends on room size.
 
Working on nearfileds in an an average size bedroom studios @ 83dB SPL is VERY loud, for this reason I use 78dB which is a lot more forgiving on my ears, the wife & neighbours

In Katz's document (part 2) he says he mixes pop songs at 6 db below the standard, which puts it at 77db :
 
"In 1996, we measured that monitor gain, and found it to be 6 dB less than the film-standard for most of the pop music we were mastering. To calibrate a monitor to the film-standard, play a standardized pink noise calibration signal whose amplitude is -20 dB FS RMS, on one channel (loudspeaker) at a time. Adjust the monitor gain to yield 83 dB SPL using a meter with C-weighted, slow response. Call this gain 0 dB, the reference, and you will find the pop-music "standard" monitor gain at 6 dB below this reference.
By now, we've mastered hundreds of pop CDs working at monitor gain 6 dB below the reference, with very satisfied clients."
 
I think everyone is in agreement.

 

A lot of what you are asking about is what I call production.   After you have the song written.... I wouldn't call it "dreaming up a song" because mostly it's hard work, however, then, you need to "listen to what the song is telling you.
 
I like to say that a song writes itself and will tell you what instruments it wants for the best expression of itself.... maybe that's a little too much of a  Zen thing... but that's my general take on writing.  I use the instruments that bring out the best (IMHO) in the song..... whether it's crunchy guitars and a smacking snare or mandolins and steel guitars.....
 
Bitflipper mentioned the mute..... yes...absolutely.... less is more.  Nothing screams "beginner" like having all the instruments playing all the time. That usually results in a jumble of sound where nothing is really distinct, and where you, the listener, have the impossible task of trying to hear a particular instrument's sonic quality from among the many things happening in the mix. The only thing that almost always plays in my music is the bass and drums..... everything else rides in and out on volume envelopes. And when something is not in the spotlight, it's normally totally out. Not always, but most of the time this is true....hard enveloping. Parts switch back and forth like a band of musicians would very likely do it live on stage.
 
Craig Anderton's list is a good one.
 
Most importantly..... the best way to learn these things is by doing them.  Some sites have the tracks available for download, and of course, you should be recording your own as well..... simply mix the tracks as you think they should sound and post the results in the songs forum. Ask for critical feedback on recording quality, production, etc..... folks here have good ears and will give you their opinion on the mix pointing out the good and the bad. It's a great resource to use to learn what you need to know to get better mixing chops.
 
Obviously, there's way more to it than this..... but since it's a huge topic covering much ground.... it's best to simply say, start mixing and post the results and begin the journey. It's a long one, but you will see progress with time and effort. Consider it a life-long journey to better sound.
 
Edit: Listen to the music of the folks here. Most have links in their signature.... also... read my signature. I use envelopes in a huge way for production of the music I write.

The other thing I would mention is that you mix based on the audience and what you want the listener to experience. Let me give you some examples of what I am talking about.
 
I listen to and play mostly bluegrass music. Here the individual performance is up front. When the singer is singing, they are in the forefront. When the mandolin is doing a solo, it gets pushed to the front. A lot of this comes from the old days when they had but a single mic on stage. The band would gather in a semi-circle around the mic. Then as a part would become the focus they move up to the mic. When they were background they would move away from the mic. Doyle Lawson and Quick Silver still do this a lot in their performances (here is a youtube clip - here they use 2 mics but I think it'll give you an idea - https://www.youtube.com/watch?v=NYsd4vm3kC0). Further, bluegrass tends to be more 'honest' in that what you hear on a CD is nearly the same thing you will hear from the group on stage or sitting around a porch playing. Not a lot of effects, and those that are used are subtle.
 
I also do the recording for our church services on Sunday. We have a praise band that includes acoustic guitar, electric guitar, electric bass, electronic drums and keyboard, lead vocalist and three backup singers. When I first started doing the recordings I tried to do it like the music I had been listening to, and needless to say it wasn't working. Nobody was happy with the sound, especially me. I started talking to the leader of the praise band, and we sat together for a number of days talking about what we could and could not do while recording the service. He also lent me some CDs of the groups that had recorded the original songs they were doing. Folks like Chris Tomlin and Casting Crowns. Here is a YouTube video of the Cast Crowns song "Courageous" (https://www.youtube.com/watch?v=pkM-gDcmJeM).
 
If you compare these two styles of music, you can easily see how different they are, in presentation, in instrumentation and in the mixing. After listening to a bunch of these CDs, we were able to significantly improve our mixing of the praise group. We still talk about how things can get better, from both sides of the mic.
 
Lastly, how you record/mix has a big influence on your final product. When I record at home, I can take my time, do multiple takes and select the best ones. I can also try different effects, different mics, move panning around, and do all different sorts of tricks. Not so when doing a 'live' mix. Every Sunday we record to a CD. One take. Two effects (chorus and reverb) and no second chances. Plus we also want the CD (which is also the signal used for our live streaming) to bring across the feel of the live environment. So we include adding in the congregational singing, the hand clapping, etc. Totally different mix from the 'sterile' studio environment.
 
As was said earlier, lots of different things to think about when trying to 'focus' your music. Hope this has offered some ideas to think about.

Thank you all for the insightful replies!
 
This is such a great forum because participation and knowledge are never in a short supply.
 
I especially like the tip about using the mute button and not having everything play all the time. (one of my common problems I admit)
 
It is one thing knowing the particular instruments that the song asks for and another thing knowing where to
(and where not to) put each element.
 
There is also the problem of when a song gets too sparse and at that point adjusting various volumes are not going to fill the void. :)
 
I have a particular question.
 
I know this question is going to reveal how little i know about certain things but this is how i learn.
 
I am confused, I am watching a video seminar on mixing and the instructor said i need to have all of my files in 32 bit floating pointer or I will not be able to achieve depth in the mix.
 
I have recorded my files (in this particular song) in 24 bit 48 khz.
 
Now i am perplexed about another parameter in cakewalk. When I go into edit => Preferences => driver settings I have the box checked 64bit double precision engine...  
 
Do i need to convert all of my project files to 32 bit or what?
 
The instructor said my Intel chip would actually function better with my files at 32bit floating pointer rather than 24 bit.
 
If I were to convert my files to 32bit how can I tell if my files are "floating pointer"?

24/48 is fine.  24 bits is as high as the hardware can capture and process out.  32 bit floating is a bit depth storage. So record at 24 bits and you can save the file as a 32 bit format.  24 bit storage is fine, too, as in the words of Bob Dylan, "You will not die, its not poison."  It is just storing the sound at a bit higher resolution than the working resolution.
 
The 64-bit engine is confusing, esp. as Cakewalk implemented it about the time that 64 bit processors were introduced, but the engine doesn't have anything to do w/ that.  It processes your sound at 64 bit,  a higher resolution.  So when your computer is doing the calculations the truncations (or rounding off errors) are at an even lower level.  Think of Pi (the number).  It goes to infinity the last I heard.  Same w/ sound calculations w/in SONAR, but if you want the computer to get finished doing them you need to stop it somewhere.  The 64-bit engine does that at a higher number so the sound more accurately reflects non-discrete analog.  A lot of older effects "sounded" better since they "upsampled" (to 32 or 64 bit processing) the sound internally.  SONAR can do this for every calculation.  I've never heard any major difference but w/ modern computers so fast it makes no practical difference so why not?  Same as storing your bounced files as 32 bit.  Hard drive space is cheap compared to 5 to 10 years ago.
 
Record at 24 bit (the highest hardware allows), store processed files as 32 bit and process your song w/ the 64 bit engine.  Easy enough to remember.
 
@

Thank you tremendously for answering my question here AT. You explained is so I could understand it also.
 
So you are saying that rendering a 24 bit file as a 32 bit file will give a more accurate representation of the 24 bit file? 
 
If so that is an interesting concept that had not occurred to me. :)

Is a song better naked than too embellished?

When it comes to music i cannot seem to reconcile the thing that music is busy and has many things are going on in it and it is not best to be viewed in a single state. 

RexRed
Is a song better naked than too embellished?

 
Depends on what you are trying to achieve with the recording.    Demo's for record companies and artists need to be less embellished most of the time.... they want to be able to "hear the song" .
 
A simple piano or guitar with a good vocalist demo may be all that's needed when you have a well written song.
 
On the flip side.... a mediocre song, with a full production behind it and harmony vocals ...... there's a saying in the songwriter's world..... You can put lipstick on a pig, and it's still a pig.
 
The secret is to know when you have a well written song and even then, to only use only the minimum instrumentation and production to bring out the best in the song.
 
Publishers and artists don't care about your playing skills on guitar or piano.... the artist needs to hear the song to decide if they would like to record it.  When I do demo's that I know might be headed to Nashville publishers and artists, I tend to leave out the longer solo's that I would love to play. That bit of info came straight from other Nashville writers. Keep it simple and to the point to show the song in it's best light. No long intro's and no fancy playing on solos. Be into the hooky chorus in less than 60 seconds and be done before 3:30 (give or take) If publishing is not your goal....ignore that and do what you want.
 
Example:  http://soundclick.com/share.cfm?id=12024980  This song started out well over 4 minutes in length. It was re-written and edited down to 3:40 total. The instrumentation was pared back to a basic country band sound. Singer is easy to hear. No solos... only turn-arounds between sections.  Intro is about 8 sec... short is best.  Outtro is short. The song was signed by a publisher.

An interesting article on just that topic showed up today at the Atlantic.
 
http://www.theatlantic.com/entertainment/archive/2015/01/in-music-uniformity-sells/384181/