I don't think AMD has had any more problems than Intel at ramping up clock speeds on 90nm. Transistor density is a pain in the butt all things considered. My take on it is that AMD noticed the main problem of thermal density and let their chips stay in development a bit longer to try to fix it rather than just push them to the market with bigger heatsinks. As a result the 90nm parts seem to run almost as cool as the 130nm parts. (There isn't a significant change in the guidelines for heatsinks for the new 90nm Opterons that I've seen so far!
Don't get me wrong, I'm sure it was a tricky transistion for AMD and Intel both. I personally think AMD has handled it more responsibly, if more slowly. But, that's my take. And in terms of AMD being dubiously secretive, I had no doubts that the 252 was a 90nm part - but maybe I knew because I am browsing around this forum all the time. I don't think it is any more dubiously secretive than ATI making 9800 Pro cards with the R360 chips and not telling the people that bought them that they really have an XT. Point being, there is probably a sensible reason, you just don't know why yet.
As for the 90nm problems, there are a few reviews out there that also suggest that the similarly clocked 90nm pieces outperform their 130nm brethren. (ie. a 130nm 246 is slower than a 90nm 246.) I don't know as it hasn't been much talked about beyond those reviews (the increase was 1% or so anyway so no big deal), but that suggests that the problems are maybe less than you were lead to believe.
Iwill Zmax DP SFF Barebones kit... (300W PSU, 8 USB, 1 FW, etc...)
Dual AMD Opteron 275's | 2 Gigs (2 Dimms 1024Mb) Patriot RAM 3-3-3-8
WD Caviar 80 Gig HDD - System
WD Raptor 74 Gig HDD
ATI Radeon 9800 Pro 256
NEC ND-3520A Dual Layer DVD±RW | Mitsumi Floppy/Card Reader combo