If correct, then it's a good thing. Sure it means a process is further away but there are two options. Continue 'developing' a process which actually doesn't work or start making good decisions like killing off a node which doesn't work and make a node that does work.
So spend 2 more years trying to make 10nm work, or put that 2 years towards making 7nm work sooner.
Funnily enough a 7nm using EUV would be much easier than a 10nm without it. Though unless you want extremely slow output you really can't make a full EUV chip. Currently EUV wafer output is far slower, roughly 4 times longer, so the more layers you make with EUV over quad patterning the lower your wafer output overall. It's a mix between doing critical layers and the rest on quad patterning. So the reality is almost everyone's EUV processes are really a mix of quad patterning and EUV.
Most rumours put cobalt in the metal layers, the 34nm fin pitch, the 36nm metal pitch, the contact gate being different and having cobalt also as reasons for the process not working. So the likely reality is a bigger node, cobalt removed, going back to the old gate design and being no where near as aggressive on the fin/metal pitch, maybe moving to more Samsung 10nm, so 44-46 metal pitch should all make a fairly simple and quick to get out process and as said it would be more like a 11-12nm process with most of the supposed improvements Intel were making for 10nm being removed.
I wouldn't be surprised if they started to work towards a 7nm without cobalt being used, with EUV on critical layers as basically Samsung/TSMC are all doing, but based on the fact that Samsung is starting to ship that this year and TSMC will start shipping EUV chips next year, Intel is going to end up miles behind.
The other major issue is the restructure, they got rid of the guy who failed hard leading this process nodes production. But they've also structured it in such a way that the manufacturing arm is now separately managed from the rest of the company....... such that, it's ready to be sold in the maybe not too distance future.
I'm predicting TSMC or Samsung gaining a few fabs in the near future.
Also worth noting that Intel has seemingly stopped working with Micron in terms of developing future joint memory products. They are no longer working with Micron to develop Xpoint (maybe dead, or still potentially developing it further themselves while maybe Micro is the one who wants out).
