Hi everyone

I've been a long time subscriber of Stereophile but I just now came to see the forum. Cool stuff!

I was recently re-listening to some stereo recordings made by me and by others, including Peter McGrath's recordings, which are made with the microphones separated about 6.5 inches emulating the distance between our ears. The whole premise being that because the inter-aural distance is preserved, both volume and time arrival differences can be preserved in the recording so that upon listening to them the brain can pin-point all the sound-sources with more realism.

Then I listened to some Nimbus DVD-As of ambisonically recorded orchestral music.

I have this question that bugs me:

If Ambisonics is recorded with a coincident microphone array and hence, it does not account for the time-arrival difference between left and right ear while recording, does that mean that it basically volume-pans only?

I have been trying to find an answer to this question in different places and experts seem to swear that Head Related Transfer Functions (HRTFs) are accounted for, but it escapes me how this works.

Empirically, they sound unbelievably real so I can't believe it's just "refined stereo" imaging only with more accurate precision and in multi-channel...

Can someone shed some light please?

Thank you!

Gabe.