Fundamentally dive computers are running a mathematical model which is attempting to model gas saturation in the body during a dive - descent, bottom time, ascent (and surface interval).
The first thing to understand is that we DON"T really know what is happening.
A number of different models have been developed that, as long as you follow the individual model rules, it is expected to get you out of the water safely. No individual model will guarantee that you won't get a bend, it is just unlikely, i.e. a fraction of a percent.
(Bends - decompression illness, do occur even when you dive 'within the table' these are often referred to as undeserved bends, or in table bends).
If we fully understood what was happening, there would only be one model!
Also the existing models wouldn't have version number - VPM-B, VPM-C, gradient factors or other compensations / adjustments
So, if you 'break the rules' of the particular model your dive computer is using it is going to get confused, hopefully, it will attempt to give a 'best guess'. The VR3 (early mixed gas computer) was known for giving you the joyous message REVERT TO TABLE if you really did something stupid that it couldn't determine an exit strategy for. The upside was you new you where in trouble!
Small violations, repetitive dives with short surface intervals, yo yo'ing, missed stops, fast ascents etc are all violations of the majority of models. Early computers weren't that clever, and would lock you out for 24Hours. Modern computer have much more computing power, have padding and additional punishment for minor indiscretions, like adding safety stops, additional decompression stops, reducing no stop time etc.
In addition modern computers often attempt to compensate for other things it determines are risk factors, cold water diving, high gas consumption etc.
Just because you are using a dive computer does not means the 'rules' don't apply. Whilst we may not fully understand how the body is working. We do understand a number of factors that are considered aggressive (likely to increase the risk of a bend). If you keep this in mind during a dive, you are less likely to 'upset' your computer or give yourself an 'undeserved bend'.
I would far prefer to dive in an environment where compulsory decompression was considered normal, rather than a fixation with attempting to push every second out of a no-stop dive, which is a very aggressive practice. Those doing decompression diving have an expectation that they will need to have good decompression protocols and ascent practice, and will generally pad their stops on the ascent if they have been aggressive during the dive.
Regardless of your dive computer, understand what is good dive practice will improve your safety and reduce the risk of you getting an 'in table bend'.
Gareth