Not Quite
Looks like Intel isn’t quite ready yet with Nehalem.
Perhaps we’ve become spoiled, but Intel provided at least Intel-run benchmarks for Conroe and Penryn three to six months before introducing the first products. We have not seen this, simply because the mobos aren’t ready.
It’s probably safe to say that we will see yet another of these “mini-delays” that have become so common from Intel lately. The first Nehalems are scheduled to be out Q4, it’s reasonable at this point to expect them somewhere in mid- to late-Q4 than at the beginning.
The one preview we have shows generally stratospheric increases in performances, but you shouldn’t assume these will be typical or average numbers. Intel is saying Nehalem will have a 20-30% advantage over Penryn, and in the past, such projections have proven to be accurate, albeit at the lower end of the range.
The preview indicates at the least that the CPU is done and that a 20% general performance increase is well within the range of possibility, so until we get a full test on a completed platform; it’s pretty safe to assume 20% for purposes of “Buy now or buy later.”
There seems to be some confusion out there about what these Nehalems are supposed to cost, and whether or not they’ll overclock.
As of now, the Intel pricing roadmaps show Nehalems showing up in Q4, with the cheapest one (no doubt the 2.66GHz) being priced at $316. The 2.93GHz version will probably go into the $530 price bracket, while the 3.2GHz version will be the $1,000 item. Actual prices may end up being higher than that, but there should be a somewhat affordable Nehalem available when they first come out.
Second, the first generation of Nehalems, the Bloomfields, are the Nehalems that are supposed to remain overclockable. It’s the lower-end versions that are supposed to have this rumored overclocking lock.
Then again, it’s always possible these plans can change, especially since the competitor is MIA.
Not At All
So far, an expected attendee at Computex hasn’t shown up yet. No one has seen nor heard of any AMD 45nm chips in any way, shape or form.
This is definitely not good news for Green, especially since they said not long ago that they were supposed to be ramping initial production roughly around now.
The only person who has any recent word on these chips says that these chips “might” show up in October/November.
He’s more than slightly skeptical about this. More importantly, so are the Taiwanese board makers.
Given the total absence of these processors, can you blame them?
Remember, this CPU generation is supposed to be the technologically easy transition. If AMD were doing OK, or even not so OK with these processors, they ought to at least have working (if slow) silicon being demoed behind closed doors.
If they can’t even do that four months before these chips were supposed to ship, this indicates a serious problem and a more serious delay than the relatively minor ones Intel has been having.
Be the first to comment