Both pieces are quite a pleasant. No issues with the mix - though I've never worked with classical music! Was it done using mic'd recordings of real instruments, pure MIDI or real instrument->MIDI?

What we had was the miked up piano recording we used as the guide then we created the instrument tracks which are all VSTi based. We use this technique to do our modern stage tracks - take the original, then go through pulling out each line. Much easier to do to be honest, as with this, I'd copy what Grant played on say a string sound, but then Grant would turn up and say he'd imagined that on a flute, and the versions I linked to are no majorly out of date because he turned up one day and asked where the bassoon was at the start - and somewhere it had got lost, and I'd not even noticed. We have a project on the go at the moment that needs this kind of stuff, so this was really an experiment to see how we would do it - but what's very clear is that we need to do the decision making together - He'd miss something I'd notice or vice versa.

The majority of the sounds are from the Spitfire Audio Albion sampler, with some Garritan and a quite old Colossus sampler which I had to bodge to use with cubase 9 as it's 64bit only now, and the Colossus sampler is 32 bit.

Maybe bring the voice and music closer in volume? I found myself turning up to hear the music and mix, then quickly back down during the speech. It sort of "broke the spell."

I've done it again - well, one of them. Grant found a mistake in the waltz, so I'm fixing that one - is this any better - there are a couple of missing instruments found and replaced.

MUCH better!

I suspect the first might have not been the right one - it made me jump when I played it - I've started moving things about between computers via iCloud, but have been a bit tardy with numbering so I wonder if I messed up = thanks for the comment.

paulears, post: 456707, member: 47782 wrote:
The majority of the sounds are from the Spitfire Audio Albion sampler, with some Garritan and a quite old Colossus sampler which I had to bodge to use with cubase 9 as it's 64bit only now, and the Colossus sampler is 32 bit.

Paulears, I know you're more knowledgeable into this stuff than I am.
But did you try to remove all the reverb of every VSTi and put all instruments to the same reverb to glue the tracks together ?

The biggest problem was matching them, because while on the non-Albion samples and VSTi's I did use an identical reverb, and the instruments just had a touch less or more to find the distance, the Albion samples were very tricky to match because of the multiple mic positions - not sure if you're familiar with it, but the instruments were recorded at Air Lyndhurst - with a Decca Tree, plus outriggers and close mics and each can be blended separately - so pretty tricky, and I'm not totally happy yet. The biggest problem is solo woodwind - clarinets really. Albion isn't good for solo instruments just ensembles, and my other old favourite, Colossus is also a bit lacking. Flutes are OK, I can do those, but clarinets just don't quite work for me. I actually have an old clarinet and although not in this one, I sometimes play in troublesome phrases on this - in this example I lived with it.

Normally I use one instance of a reverb, but here I had to use one per section - so I could drop out the woodwind, strings and brass in the groups keeping the levels locked together. Easier to just run multiples.

paulears, post: 456803, member: 47782 wrote: the Albion samples were very tricky to match because of the multiple mic positions - not sure if you're familiar with it, but the instruments were recorded at Air Lyndhurst - with a Decca Tree, plus outriggers and close mics and each can be blended separately - so pretty tricky,

Oh I see... I always thought that removing all reverbs from my vsti and choosing one on my own gave me better results.. but of course with this multiple mics recording there is bound to be room reverb that you can't remove.. Hard work.. I'm sure you'll get it right for your taste !

The close mics are quite dry, but th decca tree is so good you would never recreate that so the effort seems to be to make the other sources sound like that. In fact, the options would make a good example of mic technique that in such a good recording space makes so much sense. When we try to explain mic technique in non-nice spaces it usually fails. Perhaps I should produce something?