Market watcher warns of 'Nehalem' slippage
Customer expectations high, Intel told
Intel's next-gen processor architecture, 'Nehalem', is running behind schedule - or will be seen as such when the first of the chips ship - a US market analyst has claimed.
According to American Technology Research's Doug Freedman, "Intel's Nehalem server parts will initially launch in single-socket only". That, he told investors, "will be viewed by customers... as a delayed ramp of the company's second-generation 45nm server architecture".
Intel's single-socket Nehalem is 'Bloomfield'; the part for multi-processor machines is 'Gainestown'. The chip giant has so far only said that Nehalem chips will ship in Q4, and has specified which particular implementations of the Nehalem architecture will debut first.
Still, Freedman's comment concerns customer expectations rather than precise shipping dates. Buyers want to get their hands on Nehalems for multi-socket machines, and if they don't get them in Q4, they'll be disappointed, he reckons.
Intel's most recent roadmap update showed Bloomfield on track for a Q4 introduction on the desktop. However, its successor, 'Lynnfield', was nudged back from H1 2009 to H2 that year.
All this is good news for AMD - if it can take advantage of its rival's revised timeline. ''We believe the Nehalem launch plan opens the door for AMD's 'Shanghai' server chip - the 45nm shrink of 'Barcelona' - in H2 2008/H1 2009, should AMD be able to execute to its present schedule,'' Freedman wrote.
First Nehalem slip, now withdrawn benchmarks
Interesting to see that Intel's previously top VMmark benchmark score has mysteriously been withdrawn because the "result used a configuration that was not fully disclosed and….thus violates VMmark publication requirements." http://www.vmware.com/products/vmmark/results/withdrawn.html
Looks like AMD is back on top with Barcelona.
Sadly AMD is in trouble
Nehalem scales to 8 cores, that is very difficult for AMD to swallow. Nehalem is a very significant development, as it catches up with the theoretical technological lead that AMD had, and as such no more has. Unless AMD can deliver like hell and execute flawlessly Intel will be the clear winner here.
Now Intel will have both the mass market momentum and technological lead. AMD had considerable trouble bringing a quad core to the market... No I am not happy about this, because two good chip companies would definately be better than one. Without AMD we most probably still wouldn't have 64bit computing on our desktops but vague hype about the Itanic.
How does having Intel fail help anyone ?
I just don't get your reasoning. Wouldn't be best for both companies to do well ?