Awesome! I think nixie tube drivers forced engineers to accept the "3 significant digits" rule. By the 4th digit it flickered too much to get a valid reading.
The unit underneath is a 6-digit DVM. Here's a picture, but it's hard to get one because the tubes are scanned, unlike the HP unit which drives each tube individually. It does surprisingly well because the engineers though about what you just mentioned and made the display circuits stop long enough that the digits aren't a mess unless you've got a serious amount of slew.
I don't believe the supplies and frequency counters with nixie tube displays that I once used were scanned. The last couple of digits on the displays generally looked like 8s because they appear to flicker about every number instantaneously to the human eye.
Very cool. We have come a long way since those days. Those damn banana jack jumpers were often an issue. Other engineers would break the internal wire through normal wear and tear, throw them all back in the supply drawer anyway. No one seemed to order replacements until we were so short it was becoming a crisis. Then there was that slightly resistive layer of crud that could form on the banana contacts from years of factory 24/7 use (long before clean room methodology hit the test areas). I hated the old bench set ups. You had to check, debug, recheck, repeat until it was rock solid.
If you were using HP devices, they drove each tube individually with a 4-bit flip-flop memory card and a 1 of 10 decoder made from neon lamps on a photosensitive plate. The decoder cards were the brains, memory, and decoder for the unit.
AFAIK, most other manufacturers used custom solutions or scanned with 74141s (I think that's the driver IC) just like LEDs and LCDs are done today. You could do it super fast because neon is instant.
(post is archived)