Skeeter, I can see how whatever transitions + bottlenecks occurred last Q would make production look insanely good on a sequential basis, but I can't see how shipments went up 65%.
Yeah, they could have increased their share of the market, but assuming they did this was there enough unit demand...alright, they're obviously intertwined and yes memory per PC may have increased, but...I don't know...that just looks insanely high. I don't think the unit demand was abnormally high (if anything, given the news coming out on the OEMs, probably low-to-normal) and I don't see memory per PC just suddenly skyrocketing over the last 3 months (i.e. why would it?). So he seems to be basing it on them taking market share, which is funny, considering how silly his market share claims are to begin with (I'm still trying to figure out how they can have 25% market share and yet his revenue estimates for this quarter seem to imply they are down in the 14%-17% range based on extrapolation from the SIA figures).
The funny thing about how Niles gets number is that to get there he's assuming shipments went up this insane amount and yet he has ASP's per 64Meg equivalent declining? And for 64Mb parts themselves, he has them going from an ASP of 9.00 last Q (which he revised upwards from his 12-23-98 sheet) to 9.05? What's that, a 0.55% increase? I know contracts trail below the spot and maybe there's too much focus on the spot, but that looks really low.
IMHO, the whole thing seems really counterintuitive.
What makes more sense:
A) Reasonable growth in bits shipped + reasonable ASP increase in main product
V.
B) Insanely high sequential growth in bits shipped + almost flat ASPs in main product?
I guess those modifiers sort of give away my bias <g>
Good trading,
Tom |
