Absolutely these are only first impressions. Choir libraries typically take some work and commitment to get good results, but the conundrum then becomes are you willing to dedicate your time to something that initially left a bad impression?
I say $87 loss, but then I thought about this for a while and I think a better choice of words is in order. It's $87 more than I would have liked to pay but will accept it in good graces knowing how good a developer Soniccouture is. One always hopes for a home run straight out of the gate, but given time I think I could warm up to most of what is contained in this bundle.
There are two issues here. If you get pissed off at a library because it leaves a bad impression on the first run-through, you are going to have to get used to a lot of buyer's remorse and stuff being left on the shelf.
To answer your question: yes I am going to spend time on something that left a bad impression because unless I had a rush of blood to the head and bought something that sounded bad in the demos I'm going to assume that the issue is that how I'm playing or programming the lines is more of an issue than the library until proven otherwise. Unless a library has incredible scripting or it's a percussion instrument, the chances are that it's going to be very sensitive to playing style. For stringed instruments that are a bit off the beaten track (for me at least), such as the nyckelharpas or the sheng, they aren't going to sound right unless played even remotely idiomatically. Play a monophonic melody from a MIDI keyboard with no other manipulation on the sheng: it's going to sound not even vaguely right.
I think the way that SC set up the scripting for the nyckelharpa and the oriental mouth organs isn't ideal for those instruments though somewhat friendlier than something out of EastWest Ra or Silk. OTOH, you do get more control over what comes out vs say the Tarilonte ERA version of the nyckelharpa, which does sound a bit more alive out of the box but suffers from being a bit too consistent on the keyclicks. The clicks I think are attached to specific notes, whereas on the SC they seem to be scripted and triggered separately. SC's issue is that they were a bit too enthusiastic in how many clicks you get in a passage - but the loudness is automatable (and on the ERA version as well IIRC). Net result: approaching the instrument idiomatically is going to net better results but does take time.
With respect to the choir, it did what I expected it to do and with decent results pretty quickly. But I was approaching it very much as something that will do detailed, fairly subdued vocal pads rather than something like an 8Dio choir library. I also don't trust polyphonic legato further than I can throw it on pretty much anything.
On the value element. I've got to say, this is nuts as far as I'm concerned. I don't think I've looked at a bundle deal like that and thought "well, this means each instrument is worth X and the bundle is only worth (n-z)X where z is the number of instruments I don't plan to use so now I'm losing money". As far as I'm concerned, the bundle is worth what I expect to get use from, which could be only a small fraction of the total. As long as that is less than buying those things separately, anything else that turns up in the bundle is just gravy, which at 85% off normal prices not a lot. Even if you're using to buying in SC's sales, it's still something like 40% off, which gives a reasonable margin of error. But I suppose the (n-z)X equation is how we end up with threads on what you *don't* use from Komplete Ultimate.