Re: "Safe" cell phone WiFi capabilities?

On 5/22/2024 7:17 AM, Martin Brown wrote:

My instinct is that there are still a *lot* of phones out there (at least in the UK) where wireless n is the fastest they support.
Give it another 3 years or so and that will change. Do you care if some

>
Do you expect "whatever is the current standard" to be the new "design in"?

 Not necessarily. I suspect the market will diversify between those who value respectable battery service life and those that insist on streaming full 4k video onto an 4" OLED screen. They must have incredible eyesight is all I can say!

Do you think the radioS are a single component/chipset?  Or, components that
are individually selected and thrown together?  Given the high volumes
involved (and the small spaces available).  I.e., will the phone designer
have this choice or the *radio* designer?

Or, will there still be some manufacturing premium for that (even in cell
phone quantities!) so that something "a bit older" (at THAT time) ends up
being the new "commodity level"?

 We have sort of reached that stage with the consumer grade PCs. Improvements for single threaded usage are now very slow indeed.

Yes.  But, I wonder how much of that is related to the fact that
PCs (esp "consumer grade") are not really the subject of intense
optimization/improvement, any longer?

Luddites can't use whatever it is you are making?

>
No; that was the point of my "1950's Philco" anecdote.
>
BUT (!), I don't want to restrict the market to only folks who like
gold-plating!
>
Think of the disappointment you feel when something you want isn't
compatible with "what you have".  E.g., why can't I run Windows
on my ______.  Or, why can't I keep using my 5 year old printer
(has marking technology advance to the point that 5 years makes
something obsolete??)   Or...

 Pretty much. You can extend the life of some kit with third party drivers or other tricks but the OS has a nasty habit of prohibiting useful things to make life easier for the lowest common denominator of user. Windows Safe mode or whatever it is called Lock In for instance.
 Many things today are clipped and glued together on a one time basis with no reasonable prospect of safely prizing it apart again even though the glue used is nominally thermoplastic.

Yup.  I have adopted the "no user serviceable parts inside" philosophy.
It's almost essential if you want small, inexpensive and reliable.
And, as you drive your product costs down, the notion of repair is
almost impractical.

    "For 'nominal' cell phones (i.e., taking into consideration
     that not ever subscriber buys The Latest and Greatest), ..."
>
It's a balancing act; if you extend your support "backwards"
(in time/capabilities), then you potentially address more users.
But, you offer a product with diminished capabilities.

 Famously IBM got it completely wrong insisting that OS/2 must run on the then shipping 286 models even though it was dire on that hardware. Conflating the OS/2 launch with lockin PS/2 MCA hardware didn't help.
 It opened a Window for MS to take market share with Win3 that targetted the 386 only. We all know the outcome of that monumental IBM cock up. It united all of the PC clone makers to produce the EISA standard bus.

Sadly, people keep making the "non-portable" design decisions and, as
a result, keep having to reinvent the wheel.  OS designers are smart
enough to realize the value of an HAL; but, application designers
haven't made that realization, yet!