Yes, cost per transistor is what matters, but still it's not clear that Intel's cost per transistor isn't more expensive than other foundries, as they make high margin CPUs and can easily absorb larger costs. Also, the fact that Intel has to heavily subsidise their Bay Trail chips to compete with ARM SoC's is a possible indication of higher manufacture costs.

Reminds me of the joke, "My dick is 12 inchs long, but I don't use it as a rule"

And if you want a standalone app there's SumatraPDF - much faster and about 100 times less bloated

One thing that that article didn't mention was that App approvial is mostly automated. That's why so many garbage apps get in (Apps that just display a single jpeg, or Apps that are just repackaged game demos).

Unfortunately, there is no SPEC benchmark available for iOS.

Well, no one is going to write graphically demanding games until the hardware capable of running them is available.

Maybe he wanted to say "the past", but had a stroke at the last moment.

Better yet, how about sending dead people?
Dead people eat even less food than women and let's face it they're going to end up dead anyway, so this just gets it over with quickly.

Obviously, they should publish the full contract, just so slashdotters can give it their seal of approvial

Of course, the data is not sold to thrid parties. It's used to monitor who is making negitive comments about Apple. Once Apple unleash their invincible robot army and takes over the world, these herectics will be sent to the labor camps.

I think a better summery is "blogger creates click-bait post by sowing acute ancetodes about mechanical turk, etc. and a few technical looking graphs together to form a article that appears interesting, but really is just a bunch of waffle."

If you actually read the article you would see that the PowerVR GPU performs better than the Intel graphics

The reason why is because Intel's modems haven't been certified yet for American operators.

No, I'm talking about fusion. The neutrons need to be slowed down for effective tritium breeding and that's done by a modifier (usually graphite)

That's interesting, but the devil is in the details.
A fusion rector needs a neutron multiplier and a neutron moderator, as well as the lithium breeding material. Even if the container material is somehow shielded, the moderator and multiplier materials will absorb neutrons becoming radioactive.