Intel Broadwell Chip Release Pushed Back to 1Q 2014
The company has pushed back Broadwell's release into 1Q 2014.
CNET reports that during a call discussing Intel's third-quarter earnings, CEO Brian Krzanich said that the company will begin production on Broadwell in the first quarter of 2014 instead of the final quarter of 2013. He said the delay is caused by a "defect density issue" that impacts the number of usable chips, or yields.
Typically, when defects are discovered, Intel implements a set of fixes and then moves on to mass production. However, in the case of Broadwell, the fixes didn't deliver all the improvements Intel had anticipated. Intel now believes the correct batch of fixes are in place, and the company should go into mass production sometime in the next quarter.
"We have confidence the problem is fixed because we have data it is fixed," he said. "This happens sometimes in development phases like this. That's why we moved it a quarter. [Intel and its PC partners] have a strong desire to get Broadwell to market. If I could, there'd be nothing slowing me down. This is a small blip in the schedule, and we'll continue on from here."
Skylake, the PC chip to follow Broadwell, won't be delayed, he said.
On Tuesday, Intel reported better than expected revenue for its third-quarter earnings, but is keeping a cautious eye on the fourth quarter. Intel reported $3 billion, or 58 cents a share, on revenue of $13.5 billion, flat from the same quarter from a year ago. Wall Street actually expected to see less: 53 cents a share on revenue of $13.46 billion. Intel reported that its data center group was up 12.2 percent from a year ago, and its PC client group was down 3.5 percent.
Intel was likely counting on Broadwell to push its PC client group up in revenue for the fourth quarter and first quarter of 2014, but now the company is seeing a delay. Broadwell is based on the same architecture used in Haswell, allowing PC builders to rip out their Haswell chip for the newer model. "Broadwell and Haswell are pin compatible, so for the most part this will slide into existing systems," Krzanich said.
Intel's Broadwell chip is expected to make devices even faster, thinner and lighter than the previous generation, as well as boost device battery life. The chip is the first to be manufactured using the 14 nm processing technology, reportedly putting the company at least a year ahead of its rivals. Intel is promising to go even smaller, down to 7 nm, allowing the company to pack even more transistors onto each chip, making them more powerful while draining less battery charge or power.
Follow us @tomshardware, on Facebook and on Google+.
Both articles have misleading titles and tomshardware needs to fix it.
That's a pretty bold promise to make. How can he possibly be sure that something *won't* go wrong?
Both articles have misleading titles and tomshardware needs to fix it.
i can guarantee a minor performance improvement of 5-10% hehe
That's a pretty bold promise to make. How can he possibly be sure that something *won't* go wrong?
It's a lot less likely because it will also be made on the same 14nm process.
Both articles have misleading titles and tomshardware needs to fix it.
If you'll notice both of these articles have the same author.
Kevin does this all the time. Embellishes or adds emphasis in places in order to make the article seem more relevant or newsworthy.
He posted the Surface Pro article earlier which was also inaccurate in terms of content and mistakenly listed the Surface 2 as costing less than $100.
This is not the quality journalism that Toms is known for.
FTW!
That's a pretty bold promise to make. How can he possibly be sure that something *won't* go wrong?
It's not the architectural change that is stopping them, it's the process change. Skylake will be made on the 14nm process just like Broadwell
not according to the source of the article... the author of this article is wrong. according to the source, broadwell is a mobile part only, desktop is getting a "haswell refresh" instead.
http://www.xbitlabs.com/news/cpu/display/20131015230058_Intel_Delays_Mass_Production_of_Next_Generation_Microprocessor_by_One_Quarter.html
Well, Intel's CEO said PIN-compatible... and he only said that this MAY enable SOME people and OEMs to upgrade their devices. Lots of conditionals in there so it really sounds like he literally means PINs.
Desktop Haswell uses LGA1150 but mobile Haswell are either BGA or PGA946/947. PIN-compatible = mobile/NUC/AIO/ITX/etc.
AMD has zero to do with this delay. Intel has much bigger fish to fry than them. Broadwell is about closing the gap in the mobile space where ARM dominates, not about 1 upping and pulling ahead of AMD's desktop parts. (Or in that case 2,3, or 4 upping them...) And had you been referring to AMD's up and coming mobile parts, I'd say they have a pretty compelling APU offering where CPU power can be traded off for more GPU power (tablets and such), making the "intentional delay" argument even more unlikely. I would gladly swap my Ivy i5 in my Surface Pro with a next gen AMD APU and pocket the cost savings.
I know that is a joke, but to be serious... 100 picometer is less then the diameter of 1 silicone atom. So, it'll never happen. A silicone crystal has a lattice spacing of 543 picometers. So your 500 picometer dream is here by destroyed as well!
Suffice it to say a sub nanometer process isn't happening, not for silicone, and not for photolithography.
I know that is a joke, but to be serious... 100 picometer is less then the diameter of 1 silicone atom. So, it'll never happen. A silicone crystal has a lattice spacing of 543 picometers. So your 500 picometer dream is here by destroyed as well!
Suffice it to say a sub nanometer process isn't happening, not for silicone, and not for photolithography.
Silicone? Is Intel going to give up on CPU's when they need a sub nanometer process and switch to making breast implants?