The reason NG have been held to a certain standard is because of their USP, which was set from the early days of 2018-2019 when they began delivering accurate moulds with very high attention to detail and accuracy, along with an active engagement during the proofing process with the collectorate, which was something new. It was a big deal. This was the formative period of their existence and that set the tone and expectation; this was the 'buy-in' and the L-1011 mould was the cherry on the cake for many of us, a much-wanted 'grail and the first time that it was delivered as a definitive subject in 1:400. I remember this early period vividly, and NG's USP was a very clever one with good use of PR.
This, then, resulted in the 'high pedestal' status, in turn creating an unrealistic expectation of perfection at every level, so resulting in a highly critical response when things have gone wrong here and there; not just with loose parts but with details on wheel hubs and the accuracy of artwork etc. So yes, I guess in that regard it's not been fair and it's a bit of a pandora's box when offering such a USP in 1:400 - a catch 22.
Case in point, look at all the detail differences NG have offered with the L-1011, right down to the engine variant. They know the devil is in the detail, which collectors want, so no wonder the level of scrutiny is higher. The devil really is in the detail!
In past debates on this subject I've always maintained that without meaningful statistics, there's nothing to prove, regardless on which side of the fence we might be standing regarding QC. Of all the NG models I have collected since 2018, I've only had 2 problem models - 1 having a loose stabiliser with stripped paint, resulting in a return/replacement. The figures have been equally low for Aeroclassics and others; so no when it really come down to it, I can't complain either.
A big positive for NG this year worth mentioning has been resolving the difficulties encountered with the electroplating process for their L-1011 series moulds, which was enough to halt the process for an extended period, due to a high failure rate %.