I believe this was due to the GPU being long in the tooth. This machine had two panic events logged in the past month during mild use but nothing more sysinternal. I realized the GPU had been in service since before my time and re-seated the heat sink with fresh thermal paste. Good as new.
BTW while I am on the topic of thermal issues. I had a problem where after setting BIOS to defaults a mobo setting for CPU thermal throttling was mistakenly turned on, for whatever reason this kicked in immediately even though the CPU was at a healthy temperature. It took me an hour to find the offending BIOS setting.
Caveat emptor.
J
Lesson learned.