Nubs are way better. The first we had didn't survive 100K, the new ones got about 400K.
We had hoped for 500K, but 400K is the minimum they survived - and that's 4 - 7 times longer than before.
One thing we have to make clear here:
The stresstest just tests how long they will survive at least before the rubber TEARS (means: Until the nub doesn't center anymore)
It has NOTHING to do with the nub jumping around or not working properly anymore. That issue has been solved with the first of the new prototype.
Most nubs that failed so far were not working properly anymore, they didn't tear. And so far the ones that teared were the ones having problems (jumping, nub dance, etc.) right from the start.
Phew... so far, so good.
Now let's hope for the boards... I still don't really trust that company... but we are already looking for alternatives (and found some promising looking one).
It won't help for Batch 1, but we cannot continue to work like that in the future.
We need reliable partners.
If the board company produced all the boards as promised, I'm more than happy and we don't need an alternative. But it's probably better knowing there are others who can do it, too