I am having a similar problem with my 7-Segment 5 digit displays. I am using a program called SimvimX a free software based on flight simulation. I have four of these displays mounted on MAX7219 boards and are connected to a 16 channel mux and Arduino Mega2560. At one point in time (about 4 months ago) all was working fine. Then all of a sudden my displays started to display random stuff and one display went completely blank. After a little troubleshooting and rearranging wires, three of the 4 displays began working and the fourth still blank. Swapping from one display to another did not work. Swapping wiring made the blank display work. So one would suspect wiring. Deeper we dig and no problems found with wiring. Power: power is provided to everything via an old 1000w PC power supply with Arduino powered through USB. All devices are grounded to a one-point common ground. This is how it looks from the SimvimX site. Not exact as I have 5digit displays, but the hook-up is the same.
Displays are from?
Can't tell if they are good or not. You got the module version with both MAX and display?
About that PSU, you said it's old, is the 5V+ output stable? If unsure, a small electrolyte in the size of 10's of uF could be wired in, it won't hurt.
Also, you wrote about wiring, and even if you concluded the wiring was ok, but before you wrote that swapping wires got one display working again. That doesn't sound right. Everything's connected with jumper wires?
Let me be more clear. The power supply by old it really isn't old, just old to me it is from a 2014 build and is an ULTRA 1000w. It was a high-end power supply. Very little ripple and rock solid 5 and 12 volt output.
Swapping wires meant that I removed the wiring harness from a working display, plugged it into the one that wasn't and it worked proving Max7219 and the 7-segment 5 digit display is working properly.
All bus wires are distributed via block terminals. All wires soldered and inserted into the terminals so if one works, they all work. Power is distributed as well.
Soounds good, rigorous.
I try to understand from the schematics, are the displays connected in parallell, with a 74hc4067 providing clk? What's the benefit? I've only seen daisy chained.
The MAX7219s are in parallel on the bus not on the mux, the mux provides the clock signal, the Arduino provides data (Din and CS).
So far I'm with you. I don't get why that mux is controlling the CLK signal, and not the Arduino. Normally you'd have CLK from Aurdino and different pins to each units CS. Here it seems CS have a passive role, and you chose display by enabling CLK for that particular display when it's time to transmit data.
Anyway, the PSU, wiring and the display should be good. What's left is that mux, the Arduino, and the software. If it was the software, a reset would fix the problem, so unlikely (I mean, you have restarted, haven't you?). I'm really out of ideas at this point, hope someone else have new fresh ideas 
@BBQSteve, please do not hijack. Thread split.
I am having a problem with my 7-Segment 5 digit displays. I am using a program called SimvimX a free software based on flight simulation. I have four of these displays mounted on MAX7219 boards and are connected to a 16 channel mux and Arduino Mega2560. At one point in time (about 4 months ago) all was working fine. Then all of a sudden my displays started to display random stuff and one display went completely blank. After a little troubleshooting and rearranging wires, three of the 4 displays began working and the fourth still blank. Swapping from one display to another did not work. Swapping wiring made the blank display work. So one would suspect wiring. Deeper we dig and no problems found with wiring. Power: power is provided to everything via a 1000w PC power supply with Arduino powered through USB. All devices are grounded to a one-point common ground. This is how it looks from the SimvimX site. Not exact as I have 5digit displays, but the hook-up is the same.
Check the 4th displays separately with a power supply. See if the digits glow.
Displays tested and are fully functional. Even tested a spare in this position and it does not work. Tested non-working display in another position and it works.
please show YOUR wiring.
Real pictures of your modules/setup.
Each and every wire must be clearly visible on the pictures so we can follow the route.
Why?
Usually these displays have a set of connections on both ends so that multiple displays can be daisy-chained. No multiplexer needed.
(Notice that one end has pin marked "DIN" and the other end has pin marked "DOUT")
Even if your displays don't have that, using 4 Mega pins for the 4 LOAD pins instead of 4 pins to control the multiplexer, would remove the need for the multiplexer.
Yeah it's some weird setup there, we've been discussing it already in this thread. Dunno why Steve started a new thread, but here we are.
My wiring is EXACTLY like the schematic. The only difference is that I use 5 digits vice 8.
Because I was told not to hijack someone elses thread even though I thought they were similar problems.
That works well if all your displays are located close to each other, but if they are five to six feet apart in different consoles, not so much.
No worries mate, the moderator split the thread.
Have you reached out to that company Realsimcontrol, if they got any insight?
I have reached out to Vlad and we have been discussing basically the same obvious solutions. That is why I reached out to you guys, for some out of the box possibilities.
Look, some background. I was an electronics technician for 38 years so I am very familiar with troubleshooting and do not ask for help unless I have exhausted all of my know how.
I have tested the wiring both with a meter and an oscilloscope. I have substituted all devices except the Mega2560 and I cannot find any logical reason for three of four displays all hooked up the same way and same devices to work and the fourth one does not. I have had this setup working since January and all of a sudden a month ago the problem crops up.
I appreciate your help, but I think it is more a control issue vice assembly.