Ask the Experts: Intel's Aicha Evans Talks Wireless and Answers Your Questions
by Anand Lal Shimpi on August 15, 2013 9:00 AM EST- Posted in
- Smartphones
- Intel
- Mobile
- Tablets
- wireless
We've followed Intel's CPU team's march towards mobile for years now, but we've seen very little from the wireless group. That's all beginning to change. Earlier this year at MWC, Brian and I had one of our first meetings with the wireless group at Intel. In that meeting we met Aicha Evans, Vice President of Intel's Mobile & Communications Group. Every now and then I get to meet someone who is a pretty awesome combination of smart and passionate, and Aicha definitely fit the bill.
Unlike most of the folks I talk to on a regular basis at Intel, Aicha didn't come from the CPU side of the business. She got her start at Skyworks, an RF/comms company, before eventually being recruited by Intel to help develop their wireless strategy beginning with WiMAX. While most folks tend to have a very center-of-the-universe approach to talking about the part of mobile SoCs they contribute to, Aicha struck me as unique in that she ultimately views her role at Intel as one of enablement. In her own words: everything that computes, connects. She's taken it upon herself to make sure that Intel's computing story isn't held back by the lack of a connecting one. Her role is a challenging one. Anyone looking to do big things at Intel in the computing side of the business has decades of success in CPUs to rest on. Trying to do the same with wireless at Intel comes with no such safety net, and definitely no head start.
That initial meeting, which unfortunately remains off the record (it was a pretty awesome one), got me wondering if it would be possible to get Aicha talking directly to the AnandTech audience. A couple of months ago, while visiting Intel, Aicha agreed to sit down with me at Intel Studios and talk a bit about their wireless strategy on camera. She also agreed to personally answer any questions AT readers might have about Intel wireless or the industry in general.
In order to make the interview a bit easier to digest, Intel split it up into 5 - 7 minute segments. I've embedded them all below, along with a brief synopsis of each video. After watching the segments, be sure to leave any questions you have for Aicha in the comments to this article. Once we get enough, Aicha will look through them and answer them in an upcoming post.
Everything that Computes, Connects
Aicha's story is actually a great one and I'm sad that we didn't get the entire thing on film, but in this first video you get a bit of her story as well as insight into exactly when Intel got the wake up call that it needed to focus on wireless as well as computing in mobile.
Commitment to Mobility
Intel's march towards mobility hasn't been a swift one. The first video talked about when Intel brought Aicha on, now this video takes us through the time when the company started taking wireless seriously.
Future of Mobility
With the background established, here I got Aicha talking about the integration of Infineon Wireless and how things are going with the LTE roadmap.
Addressing the Global Market for LTE
We dive a bit deeper into the LTE discussion, bringing up the challenges that Intel will face and where Aicha believes the company's strengths are.
Wireless Innovation
Here we talk about room for improvement in wireless technologies, and what sort of innovation we can expect in the coming years. The coolest part of the final segment is where Aicha shares a bit about what drives her. Here we also end with a call for reader questions, which Aicha will be answering herself in the coming weeks!
Call for Questions
As I mentioned at the start of the post, Aicha will be reading your comments. If you have anything you'd like to hear her answer about mobile or wireless, leave your questions in the comments below!
46 Comments
View All Comments
elabdump - Saturday, August 17, 2013 - link
Intel is now two years to late, and all what there saying is: "We don't know about the standards."That's crap.
Maybe intel arrives 2015/16 in the market, which OEM wants to change vendor and isa in 2 or 3 years ?
And this happens in a market where todays phones and tablets are just fast enough, and windows looses significance.
Moishe - Saturday, August 17, 2013 - link
Wow... Well done. Very nice videos.drew_afx - Saturday, August 17, 2013 - link
We've seen wireless technology go beyond connecting between computers (like wireless I/O)What would be the technical limit on max bandwidth and min latency on wireless I/O?
Wireless USB/eSATA/HDMI/PCIe integrated onto a SoC?
Since Thunderbolt is still on copper and there is no real demand for more bandwidth, I hope for a wireless solution in the works.
elabdump - Sunday, August 18, 2013 - link
Look for 60GHz Wlan, that makes the things you want. But don't expect 60GHz stuff early from intel.jphamlore - Sunday, August 18, 2013 - link
Intel's LTE efforts are dead in the US no matter what they say. Qualcomm controls the IP needed for backwards compatibility with CDMA, IP Qualcomm has used to force the likes of Nokia to make a $2.3 billion USD settlement in 1998, and eventually Nokia had to switch to using Qualcomm ARM SoCs and Qualcomm LTE baseband chipsets in their phones (running Microsoft Windows Phone as the OS). Very soon after Intel bought Infineon, Apple dumped Infineon for Qualcomm LTE baseband chipsets.Read the following vision of Intel for WiMAX from 2007 very carefully:
http://www.intel.com/content/dam/www/public/us/en/...
What Intel was saying was their plan with WiMAX was to disintermediate the major US carriers, to at best make them dumb pipes. Needless to say, the US carriers did not react well to this. Observe how the MID, Intel's vision of the mobile device of the future, never even got off the ground. The carriers imposed a special tax on Intel's devices with extra charges for tethering.
gilgaroth - Sunday, August 18, 2013 - link
Your statement contradicts the statement of Aicha Evans that Intel will pursue 4G worldwide, not just in the US. Also just because Nokia has to settle with Qualcomm over some particular IP case doesn't mean it's impossible for someone else to develop 4G services. So what are you getting at here?WiMAX didn't make it, so what? Just because one effort failed doesn't mean the next one will. Anyways, in 2007, Intel was still working with 100+ watt processors and really couldn't even compete in low power processing, let alone communications standards or capabilities.
jeffkibuule - Monday, August 19, 2013 - link
CDMA is mostly an afterthought these days. Verizon won't be requiring it for much longer on new handsets (likely next year) since their LTE network already almost overlaps 100% with their 3G CDMA footprint.DanNeely - Monday, August 19, 2013 - link
Do you have a citation for VZW dropping CDMA as a requirement; or is this just speculation?I'm dubious about the claim, because while Verizon is on track to upgrade all of their infrastructure soon there are still a number of areas where their data access is roaming on someone else's 2/3g network. On their data coverage map these areas marked as extended 3g or extended 1x coverage. This is especially true for international roaming where most of VZWs partners are not as far along in their LTE network deployments.
jphamlore - Sunday, August 18, 2013 - link
Sorry, 2008 not 1998.http://articles.latimes.com/2008/oct/17/business/f...
gilgaroth - Sunday, August 18, 2013 - link
Anand and Aicha, thank you. I am a physics graduate student and I have two questions.Firstly, what is the progress on monolithic integration of RF and CPU onto the same die? Has a 4G modem and baseband been fully integrated with a Silvermont class CPU yet? More generally, what is Intel doing to advance the state of the art in wireless technology, like what Justin Rattner and Yorgos Palaskas discussed at IDF 2012?
Secondly, Intel has really done incredible work in enabling the high performance of modern servers and HPC facilities. However, most mobile communications technologies seem very consumer centric. They don't seem to offer the same utility for "serious computing" like for servers and HPC, or engineering software like ANSYS or MATLAB. So my question is how can the work in mobile communications be relevant at the highest levels of performance? Obviously, copper and fiber optics are probably going to be faster for the foreseeable future, so are the Xeon groups ever going to have to talk seriously with the communications guys? It seems like this is a pretty good counterexample to "everything that computes connects."
Thanks.