Low-end SoC market is already full of competitors. A new ugly chip (external design with Intel label, and no previous base) is unlikely to change anything. The bad Intel records into the SoC market don't help either.

Possible that this do not make a difference, but I still don't see why this is an advantage to buy an external design over cut down an internal one.

As mentioned, first arm64 will not touch the top server market. It will more likely take place on the low end.

Well, on the server market, the arm64 introduction will be interesting. The first chips will probably not be top performers, but there could potentially change the market price of there performance segment.

CFLAGS+=-Wall -Wextra -Wshadow -Werror -pedantic-errors

And then use tool like http://valgrind.org/

Intel is overcharging there products, just look at the Intel annual benefice. If it's true that the Z3740 is charged below cost, it's not only a bad sign of weakness but probably an illegal way to gain market.

I wonder why Intel have to buy an external design to bring the X3 chips on the market. What prevent them to cut down the X5 into a X3 ?

Even more strange is the fact that there don't integrate one of there own GPU core.

Your Apple speculative theory is interesting but completely unproved at this stage. Or did you have some information to share ?

Like for Itanium and the Xscale for example...

I wonder how long this strategy can be profitable as peoples that really need top end PC represent a relatively small market.

If true, this is unlikely that a single customer like BMW will really impact the market share of the x86 SoC compared to the ARM Soc. More likely BMW could be a an opportunity for Intel to test the current state of there x86 SoC strategy on a real project with a real customer in the hope to get more customers.

Until the price and the modem support and performances are verified, it's too early to pretend that integrated modem is an advantage. On the SoC market Intel have for years making big press release of chips that vanished into insignificant niche market compared to the SoC leading chips that massively use ARM cores.

I share your point of view.

On embedded systems I have worked on, sysvinit was useless and in each project someone have wrote an application that manage services in a why that fit the dynamic nature of the system's hardware. Dependencies between services is a must.

It's not a tool trick, but I found valuable in some project to rename functions and variables to make them telling really what there do. It's not rare that the name was a poor choice or that his semantic changed in the evolution of the project. From my point of view, it's a kind of documentation.