Microsoft is trading at $38 and was solid performer during 2013. Intel is still trading in a $22-$25 range. Are you sure it's really a Microsoft problem?
Intel has a lot of problems - that's kinda what makes it an interesting stock, believe it or not. However, it's the resolution of those problems that drives a stock up. Bottom line is that Intel needs to start growing its sales at least at a 3-5% CAGR for the foreseeable future for it to justify a $30+ stock price, and longs who care about the long-term future of the company should be asking tons and tons of questions.
You just want to bully people into believing that INTC is being unduly manipulated because you either are truly unaware of what's going on or because you are so knee-deep in options that expire in under a year that anything that could potentially drive down the share price throws you into a frenzy.
How much storage space on a phone would full Windows take? Remember that 16GB is not an uncommon configuration for a phone, so for real "productivity" this would be highly restrictive. A 64GB premium device with a Broadwel/Skylake inside would mean business, but it'd be quite expensive.
Will people pay? Will Microsoft really pursue this? Only time will tell, but I'd imagine it's a bit farther off than you think.
When will there be this full Windows smartphone? What you posted was an ARM based Windows Phone/Android product since only QCOM chips run Windows Phone.
One step closer to a world that doesn't need X86. Intel really should make peace with Microsoft and support Windows Phone - even if the near-term financial return isn't particularly compelling. I don't expect it for Merrifield/Moorefield, but I do expect it with Broxton which will have Intel's Gen GPU.
"That is amazing since ARMy has been saying for years that INTC would never get into mobile because they use too much battery power. "
This was always a dumb argument from the ARM folks. People confused the fact that Intel built high performance micro-architectures targeted at 15W+ envelopes with an inherent disadvantage to X86.
" INTC now uses less battery power than ARMy for the same processing power in the range of power needed for mobile products. THAT IS HUGE!!! They are admitting that they were full of BS before."
Yep. This is mostly thanks to Intel's superior process, although I would say having access to the most efficient bits of the Core line of processors also helps. Unfortunately, there is much more to a mobile SoC than the CPU performance/watt (and I think Intel should have pushed "performance" more with Silvermont), and this is where Intel continues to struggle. In particular, Intel still needs to get the modem built on its leading edge process and integrated, improve the quality of its in-house GPU IP for mobile applications, take leadership with ISP/camera, and the like.
"So, at 14 nm or below, I should be able to put a Core processor in my phone and use it as a processor and hot spot for my tablet and lap top that could be just dumb monitors."
I suspect that eventually all of Intel's CPUs will be SoCs and I wouldn't be surprised if there were a version of Skylake or Cannonlake that would fit in a phone, surrounded by the attendant mobile-oriented IP blocks.
It was NEVER about ARM. ARM, as that getanid61 guy used to say, was just a "rubber stamp" - a common instruction set that enabled a fleet of competitors. That said, the companies utilizing ARM are better - at least today - at putting together the right SoC package for smartphones/tablets. Intel is getting much better, but they're not there yet. I expect Broxton will be the first platform from Intel that is unequivocally leadership - and this is why, despite my frustration, I hang on.
Yep, it's Goldmont that comes with SHA instructions, not Silvermont. My apologies.
Oops, I'm sorry - he confirmed AES, not SHA. I'm actually not sure if Silvermont has SHA or if it is Goldmont. Let me dig around a bit...I do remember a SHA related post on Intel's developer network site.
Thanks for the analysis. According to marsavian I owe the world an apology for highlighting the performance results of the only decent low-level CPU benchmark we have, but I think that if Intel processors were dramatically outperforming the competition he'd be singing a wildly different tune.
Do note that marsavian himself would often link Geekbench scores (before Silvermont was a known quantity) to help us get an impression of the speed/performance of various SoCs. It's only when Silvermont came up short that suddenly the results were worthless.
The fact is, it's all we've got, good or bad, and OEMs knowing that they want to ship devices that "win" the benchmarks will be influenced by benchmarks like Geekbench, AnTuTu, Vellamo, etc. even if they're "bad" benchmarks.
Let me know when I can buy a Merrifield based device to perform an independent analysis ;-)
I do plan to purchase the first one that is available.
" he appears willing to double down and to say whatever he has to to damage Intel even if it requires significant assumptions on which he has no information. "
Coming from the author of the infamous "Humpty Report" this is downright laughable.
Max turbo for C2750 is 2.6GHz so that accounts for about 8% of the difference. The remainder is likely due to 64-bit v.s. 32-bit.
Merrifield would have been much more competitive if they had clocked it near the high end of the Silvermont range.
Bay Trail GB3 64 bit offered negligible improvement. GB3 compiled with ICC gave Silvermont a nice boost from ~974 to ~1100 in ST.
TSMC's roadmaps are usually pure fiction, which I've written about extensively. I doubt very much that 16FF will be available in products on shelves until 2016.
The better question, though is whether that really matters. At 28nm, the ARM SoC vendors have been able to fend off Intel's 22nm FinFET efforts quite nicely. It's not all about manufacturing technology and if you actually knew what you were talking about, you'd understand this.
p.s. you still can't buy a 22nm smartphone with an Intel chip, so right now TSMC has a perf/transistor and density advantage in shipping products (28nm v.s. 32nm). Just FYI.
When Apple hit $500, Tim Cook announced that he bought back $14 billion worth of stock over 2 weeks. This caused the shares to pop nicely, delivering 10%, erasing the post-earnings slump.
This is what Apple has done to build shareholder value. Not to say Apple is perfect (margin compression is a real problem for them, as is Samsung), but they are as high quality a company as they come - one of the few that I would say is even higher quality than Intel.
But I wouldn't buy AAPL's shares because there's much less upside to be had than in Intel's if Intel gets its mobile act together.
Look at AAPL's 5 year performance and gauge it against INTC's. There is no contest: Apple delivered a 5 bagger. INTC delivered a double. 10 year chart is even better in favor of AAPL.
Because Apple is the world's richest and most powerful company and it has been growing sales at much faster clip than Intel has.
Its older A6 is about as fast as Intel's new Merrifield. The A7 is far ahead. Both have been shipping for a long, long time. That's the difference.