whether USB 2.0 or USB 3.0 can basically be answered as follows:
The speed of USB 2.0 is quite sufficient for MobiFlight and the Homecockpit applications.
Theoretically, USB 2.0 can guarantee data transfer rates of up to 480Mbps. That's about 48 MB / sec. Even the best captain in his cockpit can not move the switches so fast. USB 3.0 transmits up to 5 GBit / s, which is 500 MB / sec, which is 10 times more than USB 2.0
These high speeds are not needed in our hobby for external wiring.
The power supply looks similar. According to the specification for USB 2.0, a current of up to 0.5A or 2.5W can be taken. For USB 3.0 is 0.9A or 4.5W.
Now comes the mandatory ABER: The power supplies, which are delivered for the USB hubs, have a max. Current of 3.5A. Some HUB power supplies deliver only 2.0A of power.
These power supplies are designed so that never all ports on the hub the max. Can supply power, because 10 x 0.5A = 5A and 10 x 0.9A = 9A.
The power of the power supply is designed so that individual ports only temporarily the max. Electricity can or should deliver. A permanent current consumption at the limit can not work.
For the MEGA, it does not matter if it is connected to USB2.0 or 3.0. This does not increase the transmission rate (the MEGA can deliver a maximum of 16 MHz) or the current to be extracted, because the max. Power loss of the MEGA remains independent of USB.
Pizman is so right with his consideration of power consumption!
To the MAX7219 modules. Again, I agree with pizman in principle.
The 7219 module PCBs have a design flaw that affects each time the modules are connected in a chain.
The reason for this is the diode D1 present on the PCB. It serves solely to protect against reverse polarity, which would otherwise inevitably destroy the MAX chip.
The problem with the daisy-chain is that 5V are connected to the 1st module. Through the diode, the 1st module is only operated with 4.3V, since the missing 0.7V drop across the diode. These 4.3V are now forwarded to the next module and thus the second module (identical to module 1) is operated with only 3.6V, since again drop 0.7V across the diode of the second module. At the 3rd module, these 3.6V, diode, only 2.7V !!! The MAX7219 requires a run datasheet but at least 3.4V for trouble-free operation. Thus, the voltage for the 3rd module is no longer enough!
The different voltages can already lead to perceiving different brightnesses on the displays. As an external power supply synonymous benefits nothing.
1. Remove the diode D1 from all MAX boards and replace it with a wire jumper. Then the 5V run over board 1 to board 2 to board 3, etc. All boards have the same supply voltage and there are no brightness differences (except those caused by component production)
2. Each board is powered separately with 5V directly and the diode remains on the board. Then all boards run only with 4.3V, but there should be no differences in brightness.
Basically, the brightness of the tubes can be manipulated. On the board there is a resistor in SMD construction with the inscription 104 for 10kOhm. This resistance must not be chosen smaller, it would destroy the MAX chip. I myself exchanged it with my modules and made good experiences with a value of 47kOhm. If you like, you can put a potentiometer (say 250kOhms) in series with the 10kOhm resistor and experiment with it a bit. Then measure the potentiometer and add the 10kOhm. Then use this value.
With these and pizman's hints, you should be able to make a decision whether to use USB 2.0 or 3.0. Likewise, you are able to assess the brightness differences of your displays and determine the cause. The differences in brightness should therefore also be a thing of the past.
Please let us know which type of USB you have chosen and you could solve the problems with your display.
Stephan (Time: UTC+2)