I agree with Frank, we should stay on topic... As long as we are talking about converters and conversions, I believe we are still on topic because we are discussing choices I made in the library...
I am responding specifically to this.
Be careful, web discussions are limited to sampling; they do not introduce discussions regarding future sound processing... Sampled sounds at 44.1 kHz or 192 kHz do not produce appreciable differences if one limits oneself to playing them back; Nyquist teaches us this, and I believe there is no arguing that point. In any case, there are also many other documents showing the behavior of converters when other types of signals (square wave, sawtooth, triangle wave, and sine wave) are sent as input. Obviously, for the sine wave there are no problems, everything is based on that, but when a square wave is sent, one can evaluate the transient response, and in that case, the sampling difference is clearly noticeable!
Here is an interesting article, which is not made of words alone but also of demonstrations with an oscilloscope in hand, and I prefer to rely more on experiments than on chatter.
http://www.remusic.it/IT/Le-vere-misure-del-digitale-prima-parte-008c4400
http://www.remusic.it/IT/Le-vere-misure-del-digitale-seconda-parte-986b6a00
Regarding the fact that 192 kHz samplings sound "worse," well, I think that remains to be seen because it depends on the quality of the converter, and in that sense, we need to agree on the meaning of "sounding worse." Because if sounding worse means that a DAC manages to make you hear more of the bull***t done during the Mix phase, then for me, that sounds better. I can assure you (because I heard it with my own ears) that the quality of the conversion machinery changes the sound, and in a very appreciable way too. Just a few days ago in the studio, I listened to the same sample coming out of an Apogee Rosetta and an 888, which sounded much worse! Both are AD/DA 192/24... Obviously, the sound then changes due to component issues such as the quality of the filters, but the point I made regarding DA is obviously equally valid for AD.
Otherwise, all AD converters would sound great, whereas we know that, if they are poor, they are responsible for being the real bottleneck. In any case, setting aside the transient response—on which 192 converters behave much better, and on which I think we can both agree without much trouble—I would like to see comparison experiments with the same sound sampled at different sampling frequencies, not necessarily 192; 88.2 would suffice for me to minimize the delta relative to quantization noise, applying the same filters to then resample and verify inconsistencies. In any case, from the same gentlemen you mentioned, I have also read articles in which they claim the same thing regarding the number of bits, and this is not true at all! The difference between 16-bit and 24-bit exists and can be heard, even easily by ear.