For Cymer it seems end of summer not likely, maybe end of year. It's still a big slip which will impact ASML's intended rollout.
Below 10 nm, the 13.5 nm wavelength does not seem viable. For example, 7 nm with 0.5 NA is already fundamentally excluded. The window has essentially closed, and the purchasers of the NXE EUV tools are left with big bags.
The illumination source power is a show stopper for production use of EUV but it's not the only issue. The entire EUV infrastructure including masks and resists is unproven. Added to those problems is the belief by many that at this adoption rate EUV will need to change wavelength to 6.5 nm before it becomes a mainstream tool for manufacturing. There seems no end in sight to the development resources needed to make EUV lithography production worthy.