On Mon, 16 Mar 2020 at 00:41, home user <mattisonw@comcast.net> wrote:
(On 3/15/20 2:59 PM, Samuel wrote)
 > Right after you select the USB flash drive from the bios menu,
 > you should get a grub menu that's titled something like
 > "Fedora Workstation Live".

Thank-you.  Found it.  Ran the memory test.  It took a long time, but I
like that it kept me informed.  The work station's memory passed.

Is there a reasonably easy, short, and reliable way to test
* the CPU(s) (is that the correct term?) (Intel Core -7-3770K CPU @ 3.5...)?

End-of-Life dateLast order date is December 26, 2014
Last shipment date is June 5, 2015

If you have a name-brand PC it is worth goggling your model.  Modern manufacturing
is very reproducible so you often see the same failures in a batch of similar systems.
   
5 years is when things with bearings start to fail (mechanical disks, fans).  Open up
the system and turn it on while watching the fans.   You may hear grinding or rattling 
from a fan that is failing.   

Many systems of this age have accumulated dust reducing effectiveness of heatsinks.   
After 5 years, low quality connectors start to suffer from loss of spring tension -- I've seen
systems where just moving the system to a new location caused connectors to fall off.   
In humid climates you can also start to get corrosion in cheap card edge connectors
as well as fungus growing on the layers of dust.

If the system is dusty inside, get some "canned air" and a dust mask, take it outside, 
and carefully blow away the dust.  In severe cases it helps to have a vacuum cleaner 
running to suck up dust so it doesn't just settle back in the system (or choose a windy day).
They do sell anti-static brushes for cases where the dust resists canned air.

Get one of those anti-static wrist straps and a can of "contact enhancer" and go thru the 
system checking for connectors with poor spring tension or green corrosion and apply contact
enhancer before reconnecting.   If you find a bad connector on the motherboard, and given the
age of the system, repair is uneconomical, but a suspect cable should be replaced.   If you have
to remove more than one cable at a time, take pictures to remind you of what went where.

* the power supply?

They do fail in ways that cause intermittent problems.  When I was working it was easy to 
swap power supplies between systems to see if the problems followed the PS.    
 
* the data bus (is that the correct term?)?

Connectors and cables are far more prone to failure than chips.
 
* the graphics card (NVidia GeForce GTX 660/PCIe/SSE2)?

 An elderly component with high power and cooling needs.

Most systems with these specs have onboard (integrated) graphics with a graphics
connector on the back.   Can you boot a live USB linux distro with the graphics card 
removed?

You can burn up a lot of time chasing software on a system with failing hardware.
Your first goal should be to ensure the hardware is stable.   Start by stripping down 
to a minimal working configuration, leave it running a benchmark and a video loop 
overnight.  If that works, and add things back one at a time.
  
[...]

--
George N. White III