Hello,
 

My question is...
Is the production of a song always a 2 step process of mixdown and then mastering?
It seems just before mixdown, it all sounds so great, but when the first mix is played back, it becomes obvious to me that it needs tweaking.
Does something happen, or is something lost in the mixdown process that subsequently requires this final adjustment?
 
I'd would really appreciate your thoughts on this matter.
 
TIA
Adrian
(Sonar X1 producer)

I know what you are talking about. The project seems to sound one way but the stereo mix is not as good. I believe this is due to a mistake in the export settings used. Also some effects (plugins) may need to be rendered in real time. Not many though. 
 
One thing that might help is Mix Recall. Try different mixes using it and export each. 

there are a few things to be considered.
sometimes we mix down from 24 bit to 16, 92k to 41k.
The very speakers we listen through in sonar,
are those the same speakers you use to listen to your final mix on or are you listening on computer speakers ?
 
this is not a bad question by any means though. its actually a very good one.
I find through trial and error, listening in different atmosphere's helps a lot. (the car, the home, the DAW and computer speakers as well. comparing it to other mixes from professional artists/producers.
 
 this is something I am fighting with at this very moment, this is something im continuously fighting with.
better and better mixes, better and better masters. so its okay not to be content with your mix downs, but not okay to be unhappy about them. it just means you are working that much harder at getting more advanced and training the ear that much more.
so to answer your question... I have no clue
 
but I do find the best place to start is levels. too much or not enough. to find that place where they are just right so when you do master there is enough room to expand, but not so much where you have to raise the floor large amounts.
im finding more and more to keep things simple. I used to way over mix my material. EQ, compression, FX.
now I find that by using a high pass filter, a low pass filter and tailoring the middle for emphasis is the way to go.
sure some things need sculpting, but not everything does. almost like im over doing it on this comment, I do the same thing while mixing and tend to bump up frequency's thus the floor raises to high. then im fighting levels to keep at unity.
typically if your mixes don't sound like it does in sonar, its the levels at which you exported and as john said, the settings you used to export.

Mixing is about finding the best balance among a collection of individual tracks. It doesn't matter if, for example, all the instruments are lacking highs somewhat if they all lack highs, because you can still balance them.
 
Mastering is about working with a single entity rather than a collection of tracks.
 
It's entirely possible to come up with mixes that need very little tweaking. But, you need to anticipate what you want the final sound to be, and build that into each track of the mix. Sometimes it's more effective to concern yourself only with the balance during the mix, and do the overall processing while mastering.

If you saying your exports don't seem to sound as good as your Sonar playback,, then you need to figure out why. Import the song back into Sonar and compare. 
 
But that said I have always mastered my songs in a Wave editor because that's what a Wave editor does best. Sonar can do it,  but it is missing some important tools like RMS peak averaging.
 
It's also dead simple to top and tail or add a fade out in a Wave editor. 
But most of what I need is to get that song at the perfect level that matches everything else.
My SOnar expoprts always sound identical to the playback within Sonar. 
I use the default export settings with one change and that's to 16 bit for CD's. ( I work @ 44.1 for now. 
 
They have talked about including better wave editing, hasn't happened yet, but I'm still hoping for it. My Wave editor will last for a long time yet but all software costs money and needed upgrading every so many years. I'm hoping Cakewalk will come through before that time for me. 
 

John
One thing that might help is Mix Recall. Try different mixes using it and export each. 

John, I thought Mix Recall was not available in X1. 

My mixes sound as good or bad as I make them, and the wave file I export sounds the same.
I import the mixed wave file into a new project and Master it there with some Eq and/or Compression/Limiting.
Now, it is often the case that after letting my ears reset I do hear things in the mixed wave file I missed and I have to go back to the mix for some editing or whatever.
 
I recently sent a mix off to a local mastering engineer.  When I got it back I heard a wayward consonant in the second chorus.  Sure enough, when I went back to mix and listened closely, there it was.

pinguinotuerto

John
One thing that might help is Mix Recall. Try different mixes using it and export each. 

John, I thought Mix Recall was not available in X1. 

True. I didn't realize he is on X1. Now I feel sorry for him.  

This is a round and round that ends when you've had enough, "release" it and move on to the next. That said emastered.com is free right now.

Seems I interpreted the question differently from others - that it wasn't about bouncing or exporting changing the sound, but about the difference between listening to a project for balance and listening for delivery.