The reason I said 2014 is that was Intel's deployment of their last successful new node, 14nm. While we don't know what was happening internally at that point, externally everything looked fine - it wasn't until 2016 when Intel began missing publicly declared deadlines for 10nm that publicly issues began to be known. Nodes typically take 3-5 years to develop, but I don't know when Intel first began to realize their node was troubled or when the key decisions that caused the problems were made.
TSMC used EUV for 7nm+. The original TSMC N7 didn't use EUV, it was a normal finfet node (I've seen reference that TSMC had an internal only 10nm EUV but N7+ was their first commercial EUV).
From what I understand part of what caused the problem for Intel was that Intel wanted to use an experimental cobalt interconnect layer and combine it with more complex lithography and basically failed to combine them and it was too late to reverse course by the time they realized it (Intel 4 which had been their 7nm node dropped the cobalt). Basically had Intel gone with a more conservative 10nm, they could have had a working, competitive node to TSMC all the way through finfet, their non-EUV finfet wouldn't have been so far behind TSMC's initial EUV nodes, and they might've made the transition to EUV faster as well.
And of course Intel also suffered from a host of design problems that were only tangentially related to their manufacturing issues that I mentioned in my initial post - the iPhone was 2008 and it's not like Nvidia has always used the latest and greatest node for their AI products (Blackwell being a recent case in point). And then there's Intel's bad bet on Xeon Phi versus using GPUs (to be fair Xeon Phi was a cool idea it just didn't work out for multiple reasons).
EUV and manufacturing in general were a huge part of their troubles, don't get me wrong, but overall Intel's fall from kinghood was multifactorial. A lot of things went wrong for them all around the same time.
EDIT: I see that your edited your post while I was typing but I think managed to address your questions, if I didn't let me know, although you've probably hit the extent of my knowledge on this topic, so if you have more questions someone more versed like
@Cmaier or
@mr_roboto or others here would probably be better suited to reply.
.
