Well friends, I'm finally at a point with two cars where I don't have much to work with here regarding a timing map to start with.
1st car is my 3.0 v6 turbo swapped focus - this engine was never turbocharged from the factory, and has hypereutectic pistons, fracture split rods, ~10.3:1 compression, and an 89mm bore with pentroof heads. This engine should in theory have decent knock resistance with a large quench area, which might be countered by it's relatively high CR. My goal with this engine is around 500hp (which should be relatively easy with around an atmosphere of boost, stock the engine makes 250). These engines sell already pulled for around $300 and my engine needs to come out anyway, so I am not overly worried about killing it.
2nd car is a single turbo S4 with aftermarket rods, stock pistons, 2.8 heads&cams. Unfortunately this engine retains the stock compression ratio of 9.something:1. Guys goal is basically "well 700hp would be cool but lets see what this thing will take before it pops (but lets also try to avoid killing the motor doing something stupid and preventable)"

Both of these cars will be running 93 AKI pump gas with water meth in the works. The intercooler on the S4 is fucking huge and IATs are not a problem...yet. The S4 has code added in which performs fuel enrichment regardless of lambts based on load and RPM, so we do not need to worry about fueling ever going leaner than what I want, no lamfaw hack required.

At least when I tuned my K04 car, I simply set my timing around the point where we start to get knock retard, and then back off a degree, so that we shouldn't ever see knock retard under normal conditions and let them do their job as a safety net. Obviously though, relying on knock retard numbers at 30 psi of boost at 7000 rpm is very difficult with a single knock event being very much in the realm of beating the shit out of the rod bearings on the S4, while 14 psi on the v6 is very much in the realm of breaking a piston.

So my question is: How do we determine a ball park estimate of what is safe and what isn't? Would it be worth putting an amplifier between the knock sensors feeding a set of headphones to listen to any activity the ECU might ignore? I am also working against the fact that I'm a broke bitch who doesn't have a chassis dyno which means the only consistent means of logging is 3rd gear pulls down country backroads (which are thankfully pretty straight, flat, and quiet besides cracked out druggies).

And finally - am I being too paranoid here? Obviously we don't want to run 30 degrees at that magical 30psi/7000rpm but will I be taking an unnecessary risk slowly adding 0.5 degrees at a time and testing until we see knock?