I figure 4Kn, once it is introduced (Edit: introduced into the consumer space, I mean. It's already in the enterprise space), will spread likewise, and in a few years, there will only be "legacy" drives (new old stock) available for use with existing Windows 7 PCs.
The problem here is comparing 4Kn with 512e. They're
not comparable.
1a) Moving from 512n to 512e brings major benefits and efficiencies to the drive and the data structures on the drive. Most importantly, it helps a little with areal density.
1b) Moving from 512e to 4Kn means that the the drive manufacturers can dispense with a (very thin and fairly trivial) emulation layer in the firmware. The benefits of switching from 512e to 4Kn are
very small relative to the benefits of going from 512n to 512e. (And yes, I do consider 512e Advanced Format to be a good thing.) One is about improved efficiency of how data is stored and organized, and the other is about axing a few subroutines in a drive's firmware.
2a) Moving from 512n to 512e is a relatively easy and smooth transition. The only hitch is that XP's partition manager aligned partitions on track boundaries, which corresponds to 63 sectors and causes misalignments. It's not a big deal, since you can easily work around it by realigning the partitions or using different software to create the partitions. NTFS uses 4K clusters by default (because that matches with the RAM page size), so XP's disk access patterns were
already a naturally good fit for 4K physical sectors (assuming the partition wasn't track-aligned).
2b) In contrast, the costs of going from 512e to 4Kn is much higher, since there aren't workarounds like with the partition problem, and the consequences are much worse (drive not working at all with OS vs. a mere performance hit from poor alignment). Going 4Kn would entail
losing a significant chunk of the market. No hard drive manufacturer in their right mind would do that, especially since, as pointed out earlier, the benefits of going 4Kn are so very small and insignificant.
I expect to see 512e for another decade, if not more. There is simply no good reason to go 4Kn, and lots of reasons to not go 4Kn. The enterprise storage market has always been very different (hey, why isn't SAS bleeding from enterprise to consumer?

), and there's very little reason to believe that the (limited) deployment of 4Kn there means anything about 4Kn in the consumer space.