One sort of issue I used to have back when I used to GM was how to determine whether or not a character has been spotted. In principle, yes, you use a perception test, but the way that perception tests work was so open ended that I always felt that it would be hard to be fair and consistient. Let me give you a few examples of situations where I'd feel a bit uncomfortable with the perception test mechanic.
Let's say that a shadowrunner, pictured here, is lying prone 500 meters away from a platoon of 30 UCAS marines and is observing them through binoculars. How do we handle the perception tests the marines get to notice him? Does each marine get to roll Perception against some high TN to have a chance to notice the shadowrunner? If so, even if we assume that all the marines have average intelligence, that would add up to 30 rolls of 3 dice, or 90 rolls alltogether. Even if the TN to spot the shadowrunner is around 15-20 it seems likely that the shadowrunner would be spotted, since there are so many rolls taking place. But I'm not sure if that's satisfactory or realistic because it means that the best defense against stealth is just to have many many people present since this adds up to more dice rolls.
Secondly, would the marines immediately know the shadowrunner is there and be able to open fire on him if they got a single success? If not, what would one success mean? They sense a disturbance in the force? How many successes would they need to be able to open fire? I'm not sure how satisfactory it would be in the circumstances described to have the shadowrunner observe the marines from 500 meters way for 3 seconds and then within those three seconds a few of the marines aimbot-spin to squeeze an aimed burst at him.
Thirdly, given that a combat turn is only 3 seconds, how many combat turns should pass before the marines who didn't notice the shadowrunner are allowed to open fire on him? Again, my concern would be the surreal situation where at 500 meters within the first 3 seconds a few marines are firing on the shadowrunner, and then 3-6 seconds later following a few "speak a word" actions all 30 marines are firing aimed bursts at the shadowrunner so again by probability he likely gets ripped to shreds. I'm not a military expert but or anything but it seems a bit strange that looking at a platoon through binoculars for 6 seconds would result in instant death.
Since DSF knows all the rules, what is the "correct" way to handle perception tests in the situation described above?