Feeds

Intel aims for mobo, processor FUD

No user serviceable parts inside - Seattle do nicely

  • alert
  • submit to reddit

The essential guide to IT transformation

Remember the Intel Overdrive processors? This was Chipzilla's attempt to flog user-upgradeable processors to provide a straightforward power boost to ageing systems. A year or so back, Intel chopped the whole range claiming that hardly anyone upgraded anyway, they just went out and bought an entire new system. Now Chipzilla is going one better and making it well-high impossible for anyone to upgrade anything by the simple expedient of making the whole process so fraught with uncertainty and doubt that only a masochist would try. UK reader Rob James writes complaining he bought a Socket 370 mobo because he, like the world+dog, spotted Slot 1 boarding the train for the gulag. Now he finds that although Coppermine P3 chips come in 370-pin form, that doesn't mean they'll actually work in just any old 370 mobo. Don't take it personally, Rob, it isn't just you that's been caught out. Like all mobo makers, Intel takes the view that over-engineering is anathema. If it can save a whopping 50 cents a board by cutting back on unnecessary components such as beefy Voltage Regulator Mechanisms (VRMs), rest assured it will do so. The only problem with this approach is that when new, faster, hungrier processors appear, your cut-down mobo may not have sufficient grunt to drive it and treat you to an impromptu, expensive (and brief) light and smoke show. The upmarket and expensive Xeon variants of the PIII come with plug in VRMs allowing a sensible upgrade path. At the desktop level, the things are soldered down to the board -- no user serviceable parts inside. Hardwired VRMs are VoltageID (VID) programmable so the processor can dial up the voltage it wants, but there are limits to each level of VRM spec beyond which they cannot go. When Chipzilla launched the first Katmai PIIIs at the beginning of the year, they came in 450 and 500MHz flavours. At the time, Intel's flagship BX mobo for desktops was the Seattle (SE440BX). This suffered from weedy VRM syndrome and could only support the 450MHz version. The Seattle 2 (SE440BX-2) then appeared, with all the missing bits soldered back on to bring it up to spec. Everything was hunky dory until Coppermine appeared on the scene. Although the new 0.18 micron chips use a lower nominal voltage (1.65V compared with Katmai's 2.05V), they require VRM spec 8.4 which Seattle 2 can't meet. Enter then the Seattle 2V (SE440BX-2V) which has an uprated VRM (and also chucks out one of the PCI slots for some weird reason). Details on this latest reincarnation of the Seattle are a little hard to come by as Intel doesn't list it anywhere on its Web site. The motherboard support forum is the only place it's mentioned. So now there are three different Seattles out there (four actually, as there is also an OEM-only variant, the Seattle 3, which is a Seattle 2 with an extra PCI slot), all of them with different maximum processor speeds. What chance does Joe and Josephine Public have of correctly identifying which Seattle they have and which processors are supported? The end result of all this is that Intel is effectively making the whole process of upgrading into a lottery in much the same way it has always tried to restrict overclocking. Once it was just overclockers that Chipzilla wanted to stomp on, now it would seem respectable folks who simply want to plug in a faster CPU are being dissuaded by the sheer mind-numbing complexity of it all. And this confusion is only for Slot 1 processors. Socket 370 processors have the added benefit of changed pinouts to ensure total incompatibility. There are two possible explanations here: 1. Intel doesn't give a stuff about its users 2. Intel doesn't have a clue what it's doing Which could it be? ®

Boost IT visibility and business value

More from The Register

next story
6 Obvious Reasons Why Facebook Will Ban This Article (Thank God)
Clampdown on clickbait ... and El Reg is OK with this
Banking apps: Handy, can grab all your money... and RIDDLED with coding flaws
Yep, that one place you'd hoped you wouldn't find 'em
No, thank you. I will not code for the Caliphate
Some assignments, even the Bongster decline must
Barnes & Noble: Swallow a Samsung Nook tablet, please ... pretty please
Novelslab finally on sale with ($199 - $20) price tag
Ballmer leaves Microsoft board to spend more time with his b-balls
From Clippy to Clippers: Hi, I see you're running an NBA team now ...
Video of US journalist 'beheading' pulled from social media
Yanked footage featured British-accented attacker and US journo James Foley
Primetime precrime? Minority Report TV series 'being developed'
I have to know. I have to find out what happened to my life
Broadband slow and expensive? Blame Telstra says CloudFlare
Won't peer, will gouge for Internet transit
Netflix swallows yet another bitter pill, inks peering deal with TWC
Net neutrality crusader once again pays up for priority access
prev story

Whitepapers

Best practices for enterprise data
Discussing how technology providers have innovated in order to solve new challenges, creating a new framework for enterprise data.
Implementing global e-invoicing with guaranteed legal certainty
Explaining the role local tax compliance plays in successful supply chain management and e-business and how leading global brands are addressing this.
Advanced data protection for your virtualized environments
Find a natural fit for optimizing protection for the often resource-constrained data protection process found in virtual environments.
How modern custom applications can spur business growth
Learn how to create, deploy and manage custom applications without consuming or expanding the need for scarce, expensive IT resources.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?