I feel pretty conflicted about these concerns and negativity.
Retroarch is the reason why I own multiple 3DSs, and the reason why I homebrew-enabled every 3DS I had access to in addition to my own. As far as I'm concerned, the work represented by these unassuming 3dsx and cia files is pretty much the apex of both homebrew achievements and general 3DS use. It's glorious, and honestly, we probably aren't worthy.
However, I can't deny that from about mid March 2016 onwards, regressions started to creep in that effectively broke the cycle of steady improvement, a cycle that might have peaked around November - December 2015.
Over the Spring and Summer of 2016, testing new builds started to feel less urgent and important as it seemed a little like meaningful work had effectively stopped on this project. I guess for me - if I'm brutally honest - I eventually got weary of having to reinstall my old, better performing builds after every time I tested new ones. Perhaps this was a bigger issue because I'd long been used to associating nightlies with improvements. And then one day I decided I wasn't going to test new versions anymore and instead wait for news that they're worth playing with. I was at peace with all of the cores I use (most of them!) and their quirks. In the scheme of things it wasn't a big deal.
In the meanwhile, when someone expressed negativity about, say, dynarec crashing gpsp - especially if it was done rudely (beyond anything excusable as having English as a second language) - I felt myself wince inside and there's no way volunteers and kind-hearted coders would be motivated to work on a project being dictated to by a few entitled brats acting like less-than-perfect performance of homebrew emulator ports actually make their lives worse somehow.
So basically, I see today's situation as one where one or two experimental, mostly cosmetic ideas first got implemented (in leiu of ones more related to emulation performance), and an ensuing negative feedback loop was born that's still gently simmering away 11 months later.
I feel bad about treating 1.3.3 as the final, finished version of Retroarch 3DS but by the same token can't imagine how demoralised devs who worked on porting these programs to the 3DS must feel sometimes. It's a tough platform for sure, and perhaps with a overly tough crowd to cater to as well. At the end of the day it's just video games. I love them with my heart, but that's all they are.