May be Out of Topic, but still, a Q about OS Design

[Martin Vahi]

I was wondering about 2 things:

observation_1)

    It takes a "big" market to drive down
    the unit cost of anything that requires
    development, specially hardware. The
    more people buy a microchip, the more
    the cost of its development can be spread
    across the end users of the microchip.

observation_2)

    The more universal a piece of equipment
    is, the greater its potential market.

observation_3)

   The WiFi chipsets and GPU-s
   are ridiculously cheap compared to
   the effort that went in to designing
   them. Semiconductor Foundries that
   service multiple vendors, produce
   totally different chips, are a stellar
   example, how gigantic investment can be
   diluted to the point that people
   consider their products cheap commodities.


The question, "The Q", is:
if drivers for WiFi cards, GPU-s, USB pheripherals,
etc. are such an issue, then why isn't there
an economic incentive to just produce
loads-and-loads of FPGA-fabric and make
every OS just read in, how the fabric
has been connected in the current hardware,
re-compile VHDL and load its own code
to the FPGA-fabric, ELIMINATING THE NEEDS
FOR HARDWARE VENDORS TO INVEST TO DRIVER
DEVELOPMENT and INCREASING THE POTENTIAL MARKET
FOR THE HARDWARE, FOR THE PRODUCTS OF
THE HARDWARE PRODUCESRS?

What am I missing?

The economics is there,
technology is there, a RELATIVELY RECENT precedent
of introducing new types of computers
to the market is there, in the form of
Android phones and touchschreen based readers.
There's even a competitor to the ARM:
http://riscv.org/

Basically, the first, if not the very first, then
at least the second, FPGA-vendor, who creates a
proper set of open tools wins the market and
buries the Intel and AMD and ARM alive. The
various FPGA-patents should also be outdated by now.
A prototype might be made by loading a
closed source FPGA with a VHDL-design that
implements the open source FPGA. Even that
has been done: "Flavia: the Free Logic Array"

https://www.element14.com/community/groups/fpga-group/blog/2014/07/21/flavia-the-free-logic-array


I admit, I do not imagine that any classical
Venture Capitalist would ever invest to that
project, but, hell, most of them did not invest
even in Google and Facebook.


Thank You for reading my letter and I hope
to receive critisism that explains, why
the idea described in this letter is hopeless
or otherwise dumb or stupid.

Thank Your. :-)

Somewhat esoteric,
Martin.Vahi at ...427...