So this is from 2010?
There are later charts, but a major point of FRAND is that rates do not change much over time
If you're not too lazy, you can always research their rates on your own. Let us know if you find anything different from what I've posted. Thanks.
When your so-called "NORMAL" method was established, phones were just phones, and the cellular chip / technology is THE (only) core of those products. Nowadays, smartphones are more of a computer than a phone, so it's obviously improper for the cellular technology to get a percentage style of cut of the whole product, because there are several MORE IMPORTANT components in the smartphones.
It's not just so-called normal, it IS one quite normal, valid way of licensing patents. This is indisputable. As for a smartphone, without cellular it becomes just a plain WiFi device like an iPod touch. But see below:
And that's going to be the next fight for IoT devices - should a company pay the same royalties for a wireless chip that goes in a phone, a refrigerator, or a car? Or should they pay in each case based on the value of the item its embedded in?
Certainly a popular feeling these days is that royalties should be based only on a patent's contribution towards the entire device. Determining that contribution can be difficult though.
For example, when sued for infringement, Apple's lawyers always claim that Apple should pay just pennies for anyone else's patents -- even if they're core technology. Yet when Apple sues someone else like Samsung, suddenly minor UI patents with workarounds are worth hundreds of millions of dollars in damages
As for paying by product price, here's a great example of doing that: Apple taking 30% of app sales!
Storing a $1,000 app costs the same as doing so for a $1 app. So why does an app developer have to pay Apple $300 in the first case, and only 30 cents in the second case. Because it's exactly like the way Qualcomm and other cellular patent holders charge: by using a fixed percentage of the price, so those who make more, pay more.