As a follow-up to our previous coverage (basic details won't be repeated here), Taiwanese system manufacturers have learned that Intel will start using performance-indicating model numbers on their 90 nm CPUs starting in the second quarter. Borrowing a page from BMW, the company will use series numbers to reflect models, with subsequent numbers reflecting, ostensibly, performance.
Intel is expected to label each chip with a serial number after its family name, said the sources. For instance, the Pentium 4 (Prescott) 2.8GHz chip would be marked as Pentium 4 520, the 3.0GHz as the Pentium 4 530, and the 3.2GHz as the Pentium 4 540, the sources suggested. The 700-series would cover Intel's mobile chips like Pentium M (Dothan) processors, while the 300-series would include the company's entry-level chips like desktop-use Celeron processors, the sources said.
In theory, the model number should prevent future confusion when revisions come out... if, and this is a big if, Intel counts things such as Prescott as a different series from, say, Northwood. CNet's coverage has some additional detail. The first thing that caught my eye about this is the rather prominent numbering system given to to the Pentium M branch. Most consumers would think a 700-series is better than a 500-series. Is this yet another sign of the impending death of the P4's NetBurst heritage?
Another thing about this approach is that it doesn't give competitors an easy way to get in on Intel's coat tails, which many analysts have said that Intel desperately wanted to avoid. By wrangling model numbers around their own designs (3-, 5-, 7-), they leave competitors such as AMD in a tight spot. No longer will 3200+ necessarily tell you how AMD would like the average consumer to think about their processor's performance as compared to, say, the Pentium 4 540 (notice that 520 might have cued 3.2, but it doesn't). Is this is a win for the consumer? One thing is for certain: even if this is a backing-away from using megahertz to denote a processor's performance, it's rather arbitrary naming scheme that does nothing to enable the consumer to make decisions in a fashion any more trustworthy than before. In fact, some would argue that this makes it more difficult to make comparisons with such abstracted data.
Posted February 1, 2007 @ 12:32 PM, by
Nate Anderson Bryan Lee, one of the key executives overseeing the Zune player, is out as Microsoft gives former Xbox guru J Allard total control of the music player. Full Story
Posted February 1, 2007 @ 12:28 PM, by
Jacqui Cheng Seagate announced its Digital Audio Video Experience (DAVE) this week: a portable hard drive that Seagate hopes will become a mobile content server for use with mobile phones and other devices. Full Story
Posted February 1, 2007 @ 12:06 PM, by
Jon Stokes What do hijabs and unshaven facial hair have to do with Web 2.0? Quite a bit, as Flickr is finding out. Full Story
Posted February 1, 2007 @ 11:40 AM, by
Eric Bangeman Google continues its torrid revenue growth with another big quarter. Are its multitude of partnerships and acquisitions paying off, or is it all about the advertising? Full Story
Posted February 1, 2007 @ 10:21 AM, by
Nate Anderson As the BitTorrent file transfer protocol gains both popularity and legal uses, ISPs are faced with an increasingly difficult decision about whether to throttle it or not. Network neutrality agreements can make this even more complicated. Full Story
Posted February 1, 2007 @ 10:15 AM, by
Eric Bangeman Michael Dell returns to the CEO position at the company he founded in the wake of sliding market share and disappointing earnings. Will his presence in the driver's seat be enough to bring back the good old days? Full Story
Posted January 31, 2007 @ 5:38 PM, by
Jacqui Cheng Recent data from Internet traffic management company Ellacoya says that only five percent of users generate nearly half of all Internet traffic. However, the other 95 percent are what count when deciding what new services to provide. Full Story
