Walter writes:
" High altitudes increase the voltage required to jump the gap on the plug."

??!?!?!?!?!?

I know you know better, so you must be considering a secondary or tertiary aspect.

The dielectric strength of air decreases with density so the voltage required to jump a given gap decreases with density. Pressurized mags are required to prevent ionization under the cap and the resulting crossfire to a plug with low density gas in its gap.

I use dual Bendix 1200 mags (on my TIO540 powered IV-P) that are not pressurized. The 1200s have a larger rotor diameter than the Slicks so they do not need the extra dielectric strength and can do without the condensation and corrosion. I regularly operate them at FL260 at 75% power and have tested them (pre RVSM) to FL300.  Mags work fine as long as the ignition system is meticulously maintained (Mags, plugs, wires and P-Leads).

IMHO (based on dyno and operational experience) hotter, longer and more energy spark claims have no direct impact on performance assuming you are comparing them to an ignition system that is operating within specifications on an engine which is also operating within specifications. What hotter, longer and more does give you is more MARGIN for situations where you are too lean, too rich, close to fouling, cold starting, etc.  Think of it this way, if the spark plug "lights the fuse" on the combustion process will the explosion be any bigger if it is started with a match or a blow torch?

Once you put the requisite 140 mJ (more or less) into the mixture the combustion process has started and, like being a "little pregnant",  adding additional energy won't get it "more started".

Aircraft engines have dual ignition systems for increased detonation margins, NOT for reliability from redundancy. Aircraft combustion chambers are so freaking big that by the time a flame front has traversed the chamber the remaining gas has been compressed and heated to the point of detonation.  Starting the fire in two places shortens the combustion time significantly and reduces the opportunity for detonation. This is also the mechanism for the RPM drop during a mag check.  When you turn off one mag the combustion event takes longer which has a similar effect as starting the process later, AKA "retarding the timing".  So when you turn off one of the mags during a ramp check you may think the engine is slowing down because you are adding less spark energy but what is really happening is that you are changing the timing and anyone who has adjusted the timing on a car knows that timing changes result in RPM changes at low power settings.

Regarding variable timing on turbocharged engines, IMHO it is a wasted effort. Most ignition systems have some facility to retard the timing during starting. Once a turbocharged engine is started it will spend most of its power producing life in a very narrow range of operating parameters (2,400 - 2,700 RPM and 65-100% power) compared to an automotive engine (1,500 - 6,000 RPM and 10-100% power) . As Taylor observed, there is a minimum effect of timing changes at high power settings. The complexity of adding variable timing is not justified by the minimal performance gains.  For naturally aspirated engines there is a fuel efficiency gain to be seen by advancing the timing at  low manifold settings. In automobiles (back in the cretaceous period) this had been implemented as "vacuum advance". 

My concern with most aftermarket ignition systems is an apparent lack of robustness. The use of remote "brain boxes" with D-Sub connectors; the lack of demonstrated lightning, EMI, HIRF protection; exposed crank sensors that are vulnerable to bird strike, belt failures, accidental damage all make me apprehensive of the failure of this critical system. Builders who address the robustness issue with mixed technology (mag and EI) may improve reliability while introducing a critical question. Given that there is an "ignition delay" between the static timing and the actual plug firing, that this delay is different for mags and EI and that simultaneous firing is important to detonation margin, how do you know the EI and mag are firing at the same time?

What really scares me is the availability of  builder programmable (adjustable) timing advance characteristics.  You might as well bring him to the rim of the Grand Canyon, blindfold the poor bastard and tell him to walk around until he finds a high spot.  The gains are small and the hazards are deep since a "holed" piston WILL ruin your day. Timing curves can only be developed by knowledgeable individuals using a properly equipped dynamometer. The GAMI or Barrett facilities are good examples.

I love technology and have been involved in the development of some of the most advanced avionics systems available today, but I also love simplicity and elegance. A magneto is a primitive device, granted, but it does its one thing very well. You have to respect it for that.

Regards
Brent Regan