With the latest firmware 1.1.x the position of the camsync has natural limitatios for N+1 triggers because you have to time it exactly before or after events so that there is no way to confuse them. This is easy with simple trigger on the crank, like 2 teeth for a 4 cyl as you just place the cam tooth in between the primary trigger events.
For Porsche/Audi with their 135+1 you really need the masking signal from the crank home tooth to get the timing right, people seem to erroneously assume that this works with other systems without the home signal to mask the cam signal, they have apparently not strobed the timing at all loads and rpms - there is a big risk that you get a 360/135 = 3 degree error in the Audi case when it skips a tooth.
With N-x (60-2, 36-1, etc) you must make sure it does not race with the trigger teeth of choice (your trigger tooth ref table). Someone mentioned it earlier, the cam pulse is usually at TDC for cyl #1 on many engines and means it is normally just a matter of setting it up and go.
I've set up cam sync on maybe 10 engines so far, the majority using 1.0.x firmware, there were no tricks with the settings really. I haven't really considered that there is any limp-home mode for it, I would like to see it as an option in case the sensor is not reliable and you're a long way from home.
If you think the engine runs better without the cam sync you're just fooling yourself. It's a gamble which injection sequence will be used on each startup. At least with cam sync it will always run as bad all the time, as it will use the same injection sequence every time. I've played with this plenty of times and on engines with hot cams where it should matter the most I can hardly discern a difference when rotating the injector group table, however slight - you see the problem is that with all VEMS firmwares so far you will never inject fuel at the "right time". You can always rotate the table but the injection events are started every "trigger tooth" (afaik) which is rarely at the correct time if you look at the cam/intake events.
Do not mistake "injector trim" for phasing, trim is for compensating injection pulsewidth for varying VE between cylinders (strange intake runners) and in rare cases non-flow matched injectors, while phase compensates for injection timing and naturally needs cam sync while trim doesn't.
I'd say "trim" is really hard to get right, while it's nice to apply overall individual injector scaling for WOT and high duty cycle use, in a world of perfectionism the injector characteristics and errors are non-linear which requires the input of injector flow curves for each injector.
Phasing has been written by Andrey, no idea on how much testing there has been but apparently not enough to be publically available, apparently it was using just one fixed angle (enough really for a good idle). I will push for getting it done sooner rather than later. A 4x4 table with TPSxRPM would suit most peoples desires?