Mastering Approach - Which method.... ??

Export your mix first as a stereo wav. Then import that into a new project and add your mastering chain, dither and export. That's the best way.

I think the general approach is to first make the mix as good as possible in itself, then at the mastering stage you focus on doing other overall adjustments. The issue is more from a perspective of making a set of songs consistent with each other, so that you can have en EP or album play song by song without radical changes in loudness, sonic image etc. To make it uniform as a collection of songs. If you make to many "mastering" type of decisions in the mix, it is harder to work in mastering modus in the aftermath. I think the main challenge with the mastering process it to make songs sound uniform sonically and for them to sound decent on any sound system ... from your 10$ phone to your 10000$ HiFi system. 

bitflipper's analysis was right on the money. 
 
The only thing I would add is to do the final mastering with "fresh ears".  After doing parts and tweaking to get the mix you want to export, it is always good to let it sit or a few days if you have the luxury of time.  Now, when you listen to that stereo product only, you will get a fresh perspective of where you want to be and, of course, where you are.   My own process has been to do a final mix in Cakewalk by BandLab and export that stereo mix to a designated folder.  From there I access it in Sound Forge which has been my preference mainly because I have used it from day one and my mastering plugins from Cakewalk can easily be ported over there for use.  (I put those plug-ins in a separate designated folder called Mastering in Sound Forge).  I sometimes will go the extra step of porting a mix into an app like Lurrsen or something like that...that is, of course, if I am not sending it out to a mastering house.  If you are indeed sending it to a mastering house, make sure you keep compression-limiting at a minimum and check with their requirements before sending it out.
 
I do suggest, however, that you read as much as you can about mastering because in the end only you truly understand the concept that you are aiming for with a song(s).  There is so much out there now, why not learn as much as you can.  Bottomline:  Take a few of your favorite pro-released songs and use them as "ear references".
 
While it is indeed true that a good mastering engineer is a gem to be appreciated, it does not mean that you can't learn and who knows, maybe you will someday be that mastering engineer.
 

I always have and always will :
Write/compose= anything that can capture audio ( Cakewalk/Tascam DR40/Cell phone) 
Track/Record/overdub/edit = Multi track recorder/ DAW ( Cakewalk) 
Mix=Multi track/DAW  ( Cakewalk ) 
Master= 2 track/Wave editor  ( WaveLab ) 
Replication=Burn CD/ post MP3  ( Nero/ Gold Wave/  Internet ) 
 
I will always master in a Wave editor because that is what that software is made for. It is optimized to work with a stereo wave file. Workflow is superior ( for me) and all the proper tools are on hand. 
And as said if this is an album of songs it is much easier to produce a balance between songs using the tools.  

There are a lot of good points here.  One thing I have noticed when I mix and master:  as soon as I put the limiter in the signal chain, no matter which method, the sound always changes.  Typically it's the mids and vocals that change the most, tone and volume.  I work mostly with hip hop, so high levels are important.
 
So lately, I've been adding mastering processing to a submix buss before the master buss, and it seems to make the entire process faster and more inclusive.  I know that's not the traditional way so don't bash me.  However, cpu performance is my limiting factor here and sometimes I can get myself into trouble with so much processing on the master buss.  Also, this is typically for single song projects and not an entire album.  As others have said, final mastering for a commercial album release has other considerations you can't do with a single song.
 
 

Bristol_Jonesey
Another approach with a surprisingly wide-held support is to actually mix right "into" your mastering chain from the very outset.
 
Admittedly  there are probably as many critics of this method as there are supporters - but it's worth trying out

I think this is super rare.  Having both Mastering Grade Compression + EQ (liner phase) along with a final limiter is very taxing on a system along with the full mix and other effect on tracks and buses.
 
Mixing into a bus compressor and things like that is common, but it isn't' the same as really mixing into a full mastering chain isn't super common in the professional realm.  

Thanks guys for the many replies and helpful insight, all great points.
 
Here's an idea I thought of, what do you think of this approach ?.....  (Disclaimer, I understand this isn't optimal for a body of art comprised of multiple songs, it would be more applicable for individual song releases):
 
1) Mix a project and export to WAV
2) Within the same project create another track (label it, say, Mastering Track) and point it at the mixed WAV file (not physically importing it into the project, just pointing at it).
3) Use the Mastering Track FX bin to add tools and master.
4) Highlight just that track and export to create a mastered file.
 
With this approach you can easily swap back and forth to listen between the mastered track and the mix tracks all within the project.  And if you make changes to the mix, they are automatically picked up by the Mastering Track since it is always pointing to the mix file name.  This approach also allows for better CPU management if that ever becomes a constraint.
 
Thoughts ??    
 
Sonic

SonicExplorer
Thanks guys for the many replies and helpful insight, all great points.
 
Here's an idea I thought of, what do you think of this approach ?.....  (Disclaimer, I understand this isn't optimal for a body of art comprised of multiple songs, it would be more applicable for individual song releases):
 
1) Mix a project and export to WAV
2) Within the same project create another track (label it, say, Mastering Track) and point it at the mixed WAV file (not physically importing it into the project, just pointing at it).
3) Use the Mastering Track FX bin to add tools and master.
4) Highlight just that track and export to create a mastered file.
 
With this approach you can easily swap back and forth to listen between the mastered track and the mix tracks all within the project.  And if you make changes to the mix, they are automatically picked up by the Mastering Track since it is always pointing to the mix file name.  This approach also allows for better CPU management if that ever becomes a constraint.
 
Thoughts ??    
 
Sonic

Like I said above, there are no rules. You can do what you like, but I think you are kind of making things more difficult than they need to be by doing what you propose in the long run. If you are happy doing it however you want to do it, then do it that way, but you asked for advice, and people with long-term experience advised you on the best method to use. They didn't advise that method because it was how things were done in the analogue days, but because they have learned how to best organise things for efficiency and recoverability in the future. The way the majority of people have prescribed is the best method for productivity and future compatibility. You can use another method, but you may find somewhere down the line that you run into problems, and then you will think, "Ah, that's why they said to do it that way!" 

To the post regarding uneven levels song to song. This is why you should use a Wave editor because they have the tools to do this. 
Here's an example of a wave editor ( acoustica 7)  analyzing a song. No guess work involved here.
https://acondigital.com/products/acoustica-audio-editor/  Try the demo, 
Note this song is a backing track and split Bass left/  Drums, keys right so therefore uneven. 

 
 
To the OP, You idea is creative but overly complicated. And your not going to hear changes to a mix until you re-export the song. 
 
What I do is export the mix., keep Sonar open
Open the file in the Wave editor and Analyze it. 
This information is critical and each song needs the be very close to the same readings. 
Even if it's a one off, I keep to my standards
I also top and tail another important task, 
I often find something is not perfect so I return to Sonar, fix that and re export. Once the check list of requirements is met with the analyzing I burn a CD and like Jeff so wisely has said, live with it for while making notes as you listen to the album everywhere possible. 
 
I will admit that one thing that has sped up the process is using the Brickwall limiter on my Master buss. I have it set to -0.40 as you can see above, it works.  

Pesonally, I find it helps to keep the distinct processes seperate: so I have a file for recording, I export stems into a new file for mixing, and finally import the mixdowns into a new file for mastering.  When I try to do everything in one project file I find I lose focus on the current task, and get distracted by posibilities up and down the chain.  I have worked in a single file from recording to master, but the results ain't as good.  Having said that, It might well depend on the sort of music you make, if it's all soft synths with little audio recorded in real time then the prehaps the 3 stage process is overkill?

My approach may not be what the "masters" do, but it works well for me and old style "traditional" country music.
I get great comments and returning customers.
 
I record all the tracks.
Do what is needed for each individual track
Bounce all tracks to a new track (mixdown)
"Master" the bounced track with Izotope Ozone 5 or 6
Export the mastered track as a 16 bit 44.1Khz wav file. 
Burn audio CD's with Nero BurningRom application
 

SonicExplorer
Well, my idea won't work after all, just tried it.  Sonar "locks" the mix file being referenced by the Mastering Track, so you can't export a new mix.
 
What's the consensus on placement of the mastering EQ (provided we are talking extremely minimal tweaks)? Should the EQ be last or rafter before the limiter?
 
     Sonic

EQ before limiter. Limiting should be the last process and EQ should be the first because cutting or boosting frequencies will affect the overall level.

For quick results, take a look at AAMS Auto Audio Mastering: http://www.curioza.com/ - there's a free and paid for version.
 
I tried out the free version on a couple of tracks and the results were about the same as I got from LANDR.
 
There's hundreds of genre presets, so it's likely that given a bit of trial and error you'll get something close to what you're looking for.
 
Like all auto-mastering software, it makes generalised decisions so it wont have the various tweaks you might want in at specific points in some tracks. I did however find it useful as an additional reference track when doing the final master.
 
M.