does mixing kill the soundstage?

In modern multi-miked recordings the soundstage is an artificial creation. A talented recording engineer can assemble the individual recorded tracks (typically between 24 and 48) into a coherent sonic image. The juggling of left versus right channel volume and phase manipulation allow the placement of specific instruments within an artificial, digitally created, ambient soundfield. Within any musical genre there are generalized rules that govern the mix with mono compatibility being the highest priority. A skillful engineer can fashion a very convincing aural experience.

Just to add a bit to 61's post, there is always the actual room sound aspect of recording vocals, piano, acoustic guitars, etc. that can help an engineer create a convincing soundstage, even within the artificial environment 61 so well describes. I am about to begin a recording project that involves a "clicktrack", acoustic guitar, guide vocal, percussion, bass, piano, horns, in that order, and I will try to keep as much of the natural room sound as possible even with much overdubbing.
If the actual room sound is not that great, I will add a different room sound as discretely as I can.

Bmpnyc, obiviously if you have a great sounding tracking room, then it can only help the quality of the final mix. It sounds like you'll be recording in only one room, hence your ambient room sound will be consistent from one track to another. Alot of recordings are not this fortunate and efforts must be made suppress the original room sound and to then add a synthetic ambience (basically a reverb, nearly always digital) as a sonic "glue" to bind the divergent sounds together. Good luck w/ your project!

Thanks 61.

Speaking from experience both as an audiophile and having done professional recording engineering both in top notch studios and minimalist classical recording venues, I just had to chime in. First of all, 99% of all pop/rock music is completely artificial. Single source mono microphones are used to record the majority of instruments. In a final mix, the mono signal is panned to the left/right channels to give a sense of "placement". Personally, I think it is more accurate to call most pop recordings "placed" mono than stereo. Still, to make a good sounding mult-track rock recording takes talent and possibly the finest engineer out there is George Massenburg (he also designs his own gear under the name GML and it can be as breathtakingly expensive as high end audio gear). In terms of symphonic recordings, multi-channel micing is utilized mainly by Deutsche Grammaphone. Engineers opt not to record in this fashion because the sound leaves a lot to be desired. Recording a symphony is more about the sound of the orchestra in a hall and capturing the essence of the orchestra as a whole. 48 tracks of mono mics that are mixed together can never represent the sound of an orchestra. A recording that has been made in this fashion can be instantly identified and the phase shifting and incoherrence between the microphones is nothing short of annoying. The finest classical recordings out there often use very few microphones. Chesky does a phenominal job of doing minimalist micing (but the performers aren't the greatest). The older DECCA recordings from the "Golden Age" would be your best bet for symphonic recordings (but use a few more mics but did a great job with it). John Eargle recordings on Delos also use a few more mics but are exceptional. If you are looking for an out of this world acoustic recording, pick up "Meeting By The River" by Ry Cooder and Vishwa Mohan Bhatt. The disc is impressive both because the two of them had never met when they recorded it and because of the calibur of the recording. It is a minimalist 2 channel analog recording that cannot be topped.