Just bought Ymir and it's a delight! The singing voices of those wonderful kids definitely have a magic all their own. I've owned the full Mars and Venus choirs for years, and they're good, but I don't recall ever being quite as bowled over by their sound as I am by Ymir.
Here's a quick trick I've just put in place for using Ymir in my (mainly) VSL Dimension orchestra template. I have a very particular spatialisation regime that uses binaural panning and algorithmic reverbs throughout this template, chiefly because I don't have the computing horsepower to use convolution reverb everywhere.
Although all the room mics in Ymir, especially when combined in various ways, sound truly lovely - as we've come to expect from Synchron Stage recordings - in this case I'm using just the left and right "Main" (Decca Tree) mics. I send this Main audio pair to two buses, in each of which I use a binaural panning plugin in which I first extract one side of the Main audio, then pan it binaurally by 60º; so my 2 direct paths are now 120º apart, each originating from one side of the Decca Tree. Then I feed each of these to its own algo reverb and apply maximum width to both, so their stereo reverb coronas spread into the middle and also extend way beyond 60º each side.
The Decca Tree L and R mic feeds are mutually phase-incoherent enough that my phase meter across the sum of the binaurally panned pair hovers very healthily around a mean neutral, while the goniometer shows super width, with or without the reverbs. It's way more natural than pushing the old Mid-Side trick to extremes. This trick just sounds ridiculously good!
[I'm sorry but please don't ask me to spell it all out and serve it up on a plate for you here - I don't think that would be fair to VSL in this forum. But I'm sure those who know what they're doing with audio and have some relevant plugins would know how to get it done their way. And I'm pretty sure they'll find it's magical too.]