You come back to me with a GPU clock speed quote that comes from actual Samsung literature and not a benchmark app or a source-less web page, and then we'll talk some more.

Firstly, the comment in the article clearly got its clock speed claim from the article. Secondly, I think every one of those articles got the clock rate from a benchmarking application because Samsung didn't release a GPU clock speed and it's standard practice in the Android community.

Does it?

Provide an example of a thirdparty app running at 533MHz on the S4.

"It is supposedly clocked at 533mHz, which is more than double the iPad 4."?

"What the device is capable of" is a function of the device's current state - clock speed, cooling, voltages, power supply, etc.. You want to test the device in the same state that it will actually be used. A 533MHz benchmark is a good indication of what this particular chipset would be capable of when it is running at 533MHz. It is not a good indicator of what this chipset would be capable of if you clocked it to 400MHz, or at 1600MHz, or 3GHz.

That would depend on whether thirdparty apps get the higher performance too. I don't think the fact that a handful of built-in Samsung functions use the higher clock is a good guideline to how the device will perform in typical use.

I'm curious as to how Samsung is maintaining this list of videogames that require an underclock, and how often it's updated. Wouldn't your phone's performance on the new Deus Ex game drop with the next software update?

First line of the article:

Somebody with a Galaxy S IV has just performed an Antutu benchmark and revealed all the specifications of the device!

Antutu being one of the benchmarks that supposedly activates the overclock. You must be a real threat down at the debating club.

Pony and trap, crap. AC is a cockney.

The CPU isn't "rated" for anything, Samsung didn't release any GPU or CPU clock figures until the denial quoted above.

And in fact, Anandtech specifically points out the opposite:

It's interesting that this is sort of the reverse of what we saw GPU vendors do in FurMark. [...] In order to avoid creating a situation where thermals were higher than they'd be while playing a normal game (and to avoid damaging graphics cards without thermal protection), we saw GPU vendors limit the clock frequency of their GPUs when they detected these power-virus style of apps.

The section of code that activates the changes is actually called "BenchmarkBooster". Someone will be fired for that I'm sure.

That doesn't tally with the information extracted from the S4 code: it lists several benchmark apps, which when detected activate a "boost" feature that changes the CPU clock.

More formally, your knowledge of certain aspects of the language is strictly bounded by the loss of this information. Historical linguistics as a field knows this and does research into it. I don't see that this undermines the expertise of a historical linguist any more than the uncertainty principle undermines the expertise of the theoretical physicist.

What experience? They don't steer the ship and they don't create the charts.