With one exception, Intel’s press conference at CES was depressing and a lawyer’s signature short of outright lying. It was not just the usual content free and painfully dull hour long slog, they went as far as possible to deceive the press, and it worked.
Mother always said, if you have nothing good to say, don’t say anything at all. We have long ignored that advice, but we will start out with the one good thing that Intel announced at CES, the phones. If it weren’t for those, and we actually heeded the advice our mother never actually told us, this would be a pretty short story. Luckily for us, Intel showed off Lexington.
Lexington is not just the town over from where mother never gave us the aforementioned advice, but also the new low cost Intel phone platform. If you recall CES 2012, Intel introduced their first real phone platform based around the Penwell SoC. The SoC in this platform is just the same thing with more features fused off and more clocks reduced to get to a ‘value’ product, but the silicon is unchanged. This ‘new’ CPU is called the Z2420, and it is the heart of Lexington.
The platform has a 3.5″ touch screen, 7 shot burst mode pictures, and just about everything you would expect from a modern phone platform. Since it is a down-costed model aimed at the third world and emerging markets, it also has a bunch of features added in that we don’t get. FM radio and dual SIMs with dual standby are at the top of the list, followed closely by a MicroSD slot. [Editor wonders why we don’t get FM radio on our phones here in the US, starts to feel like a 3rd world country here.] Last but not least, it has mandatory wireless display functionality because most villagers in rural Africa can’t afford a full cost phone, Intel seems to think they still have a burning desire to display their cheap phone on a large flat panel TV. Ummm… nope, no good explanation for that one.
That said, Intel seems to be doing the right thing here. A low cost phone aimed at the biggest untapped market for smartphones is a good idea, and although they were a bit coy on the cost, no carrier would pick it up unless they were in the right ballpark. Chats with Intel’s Mike Bell convinced SemiAccurate that they do get the market they are jumping in to, and understand it’s unique needs fairly well. Software and use cases are very different in these markets, and those needs are being addressed well, wireless display aside.
One other thing that most people don’t quite understand is that Intel gives carriers a turnkey platform that they can customize at their whim, or not. The not is the important part, Intel gives OEMs a carrier certified device that is plug and play if they don’t want to mess around with it. Load up the logos and splash screens and off you go. In the US and Europe this isn’t a big deal. In a brutally cost conscious market it can save the OEMs a lot of money, and that is what this market is all about. Intel is doing right with this phone platform.
From there, they announced the Clovertrail Plus SoC for tablets, and knighted it the Z2580. The specs would probably underwhelm, but luckily they were not provided. And when Intel moved on to the upcoming Bay Trail 22nm Atom announcement, it was promised as a quad core coming late in 2013 that would have “>2x performance over current generation”. That is not exactly a high bar for Intel to reach for, and rest assured the drivers will still be abjectly broken so software will run badly, faster. Don’t wait in line for this one.
From there the topic shifted to a much sadder area, desktop and mobile. Intel was keen to show off “innovation”, but they utterly failed to do so. In fact, their offerings were so bad that they had to distract and go within a hair of outright lying about their products. Luckily they did so to a room of tame press, and only one lightly called them on it. Money talks and BS gets reprinted as headlines. The truth dies alone in the cold.
What is the problem? Intel talked all about Ivy Bridge chips at 7W, and how it was a pull in of a product. It isn’t. Nor is it a real product, just a binning and castrating of the existing 17W parts to reach a lower wattage. The unfortunate part is that Intel had to actively deceive a room full of press and analysts to get the headlines they wanted, and get them they did.
You know Intel is up to something sleazy when they don’t pre-brief anyone, don’t give out specs, and don’t answer questions. It is a sure sign that the products both suck and don’t live up to the meager promises they are making. The new “7W” Ivys do both, they suck and are most abjectly not 7W parts. No specs were given, no nothing, but that didn’t stop everyone from proclaiming the awesome technical achievement that Intel had just made.
The one journalist who mildly called them on it was Anand of Anandtech fame, and his article on the subject has a nice table listing all the -Y parts that Intel even failed to mention the name of at the CES keynote. The problem is that Intel invented a new power measurement mechanism called SDP, and only mentioned it in the keynote in passing. By that, we don’t mean calling it out as a new measurement and moving on, they just said roughly, “7W, pulled in, we are teh awsum! Go us. 7W, no one else can do a 7W SDP part, and we pulled it in. We are so cool, 7W, yay us!”.
The problem is that the cTDP that they invented to make a part lower wattage than it really was was simply not producing good enough numbers so they made up a new test that was easier. Voila, 7W. Go Intel! Yay team! Unfortunately it is a 13W TDP CPU, a mere 85+% whoopsie. A 13W Ivy Bridge part would be a good thing, but Intel had to go and screw it up to get headlines. And headlines they got, at the cost of their honesty. If they didn’t mention SDP in the keynote, I would call them outright liars, but they did, so on a technicality, they are only purposefully deceiving the press, analysts, and anyone else watching. Technically not lying this time, but they are being abjectly dishonest.
Sadly, this didn’t stop them from congratulating themselves until they got bored of it, then the majority of the press took over from there. Sadly, they didn’t have to do this, instead of purposefully deceiving, they could have just been honest and called it a 13W part. That would have been both accurate and technically correct, plus it would have grabbed headlines just as effectively. Honesty seems right out at Intel now, so they had to make up a complete BS measurement scheme that tracks absolutely nothing remotely real so they can technically not have a senior executive lie on stage. Only technically mind you, the intent to lie was still there. Shame on Intel. Shame on the press for not asking about it and repeating the deception. Shame on everyone but CNET, but you know that story by now.
Then came Haswell, and Intel showed off a reference design. It was thin, but unlike what they claimed it to be, was nothing near innovative. It was just fractionally smaller, fractionally lighter, and there were some grand claims made about it. No details, so given the rest of the presentation, lets just assume they are lying about something, that is the safe bet when Intel speaks now. We will believe they are not lying when we see it, but we will probably be waiting a long time.
Moving on, a few things were tossed out that will shake up the industry. Intel said what we all knew was coming, the 2014 Ultrabook platform will make touch mandatory. Ultrabooks are dying in retail, after a year they barely reached 1/8th the targets Intel set for them in 2012 and 2013 doesn’t look to improve anything. They are a niche, but that is about it, shiny things for the stupid indeed. So what to do about a too high price, crippled feature set, and design rules that preclude fixing the most abject problems including price? Make touch mandatory, that will fix everything! Windows 8 is dropping sales by double digit percentages, so lets play to their broken paradigm.
What could be better? How about mandating McAfee Antivirus, Intel just slipped in that fun fact at the keynote and no one seemed to notice. McAfee seems to be mandatory, I guess Symantec was not losing share fast enough. If you wanted a cheap, crapware free laptop in 2014, Ultrabooks are right out. Intel seems to have a knack for making their computers as unappealing as possible to the consumer, then proclaiming victory as sales slide. Independent observers are understandably mystified, and buyers are not living up to their name in droves. Clue to Intel, making your already awful Ultrabooks suck more isn’t the key to massive sales. No, that was not a joke, we are dead serious on this one.
From there, Intel went downhill. They talked about perceptual computing, face recognition, gesture recognition, voice recognition, and lots of other technologies that have nothing to do with their hardware and platforms. Luckily, each of these licensed technologies will add a little bit to the cost of the system, take a bit off the battery life, and bloat the software image. If you are a fan of crapware that not only is hard to remove but gets in your face all the time, Intel’s 2014 lineup is for you. If you like a clean, fast, glitch-free computing experience, you should be very afraid of the crap Intel is shovelling on the industry.
Then Intel tried in vain to convince the audience that touch was the future. A large, very expensive flat folding touch screen all in one PC was shown off as a way for a family to play Monopoly together. You just spend $1500+ on your AIW PC, tens more on a PC version of Monopoly and off your family goes. You have two hours of battery life in this mode, so hope you have short games, and if you didn’t charge your, I am not kidding about this, e-dice, your whole family night is SoL. How could anyone not love the concept of a battery operated Monopoly game that costs 50x or more what a Monopoly set does? If the future is not Ultrabooks as consumers are indicating, Intel has plans to lower the bar yet again.
In the end Intel had one bright spot, the low cost phone platform. Lexington could be a serious winner, the only question is the margins Intel ends up getting for the chips. The rest was a dismal array of purposeful deception, dancing around lies, and depressing crapware bundling. Intel is bleeding marketshare to tablets, and moving all their engineering effort in to making the user experience abjectly worse for everyone. It is enough to make one lose faith in the few good scraps they were trying to hide. It worked.S|A
Latest posts by Charlie Demerjian (see all)
- You can now buy bare Snapdragon SoCs with 410E and 600E - Sep 28, 2016
- Spin Transfer Technologies talks about their ST-MRAMs - Sep 27, 2016
- ARM adds CMN-600 interconnect and DMC-620 memory controllers - Sep 27, 2016
- Globalfoundries 7nm process isn’t even close to the name - Sep 26, 2016
- ARM upgrades realtime offerings to v8-R and adds Cortex-R52 - Sep 21, 2016