I'm sure Mike will be along shortly to answer in person, but in the meantime I'll shsare my own views on this.
[quote="tomdot"]It then goes onto explain how we never hear a true stereo picture anyway.
Stereo is an illusion -- it is a clever way of fooling our ears (or the ears of most of us -- some people really can't hear stereo! Their brains are too clever to be fooled!
) into thinking that there are discrete sources of sound laid out before us between the speakers.
So why must we look to pan to achieve a balance?
Because the inherent action of the pan control results in small changes of levelof the panmned signal to each channel -- and thus upsets the critical balance slightly relatice to a pure mono balance. Consequently, a suitable compromise must be found, which will involve minotr alterations of level relative to the original mono mix.
For example, when recording a mono point source such as an electric guitar wouldn't it be better to leave it up the middle and get it sitting right alongside everything else (as per the quotes above) before panning it later. You would be safe in the knowledge that sounds work in mono before spreading them out to achieve more separation if required.
That is certainly my preferred approach -- born from years of working for a mono-concious broadcaster! To me, the hardest part of making a mix work is sorting out the overlapping spectrum issues, where sources with similar spectral content overlay one another and thus confuse or mask each other. Fundamental issues like this are easily ignored and overlooked if you work in stereo from the outset -- and then become major pains in the behind to sort out after you've already spent hours building your mix.
So, I always start in mono, and sort out the arrangements, EQ and balances to make the thing work reasonably well in mono. After that, I pan things to where I want them to be and fine-tune the balance as necessary, and then check mono again to ensure it still works acceptably well.
It is noticably harder to build a mix in mono (or it appears to be much easier to mix in stereo, if you prefer), which is why few force themselves to work this way. The same issues arise when working in surround -- the additional spatial imaging makes it even easier to create fantastic sounding mixes in surround... but which don't work well at all in stereo or mono!
As an aside to this, I am looking at a mix where I have been trying to balance in the way described and encountering a drop in volume from wider panned instruments when listening in mono.
Yep, that's the inherent compromise involved in stereo-mono compatibility. Central sound sources will always appear to be 6dB louder than widely panned sources of the same source level. There is no way around that. That's why you will need to readjust the balance slightly, and decide on the best compromise mix that satisfies in stereo and is still aceptable in mono.
I'm undecided whether this is a drop in volume there or an increase in volume from central sources?
Either. It depends on what you consider to be your reference point, and how the panning laws are configured. Some attenuate the edegs, some boost the centre... But the same 6dB variation between edge and centre will always exist.
I understand that it then goes onto say that you should judge more filtering later on so I may be jumping the gun, but still I'd like to know how this advice relates to other general advice I've read and whether this is accurate or not.
Most source recordings contain a degree of subsonic rubbish which is detrimental to the final mix, even if not directly audible, so high-pass filtering of each source is generally a good thing. But clearly, you don't want to remove important fundamental frequencies, and you don't want the inherent phase-shift associated with high-order high-pass filters to have an audible affect on the source, either.
Consequently, where you set the HPF turnover has to be judged on an individual basis, removing only the irrelevant and unhelpful LF content, without detracting from the source contribution. And that can really only be judged as the mix is being constructed.
Sometimes it is beneficial to be quite heavy-handed with the HPF. For example, heavily cutting the bottom end out of an acoustic rhythm guitar might sound horribly savage when you audition the tarck on its own, but it might well make it work far better in the mix, avoiding muddying up the lower midrange of the overall mix.
So there are no hard and fast rules, and it has to be judged by ear. However, as a rule of thumb, I normally high-pass filter everything at between 30Hz and 75Hz or so when recording and tracking simply because there is nothing but rumble and rubbish down there anyway (the lower setting being used for bass, kick drum and other sources with lowish fundamentals, and the higher setting for everything else). And then I would typically introduce a second high-pass filter at a higher frequency as I start to build the mix, to remove unwanted LF spill or unhelpful lower mid that is cluttering the mix.