I've read many threads on this topic of accuracy requirements and I hesitate to get into a debate. But this thread is about my new product so I suppose I might as well chime in once. I know math, I know e-targets, and I know running matches, and I know competing in F-class. Here is the intersection of all four.
The answer to how much accuracy is required comes down to statistics. Suppose you have 4mm of random error. You are shooting a group at 600 yards that is 3 inches (1/2 moa).
Yes it is true that on each shot, the shot might be reported up to 4mm in either direction, and 4mm is enough to put you outside the line, or inside and change the score you would have got on manual targets for that shot.
But you shouldn't think about it in terms of whether individual shots are reported correctly. You don't see the paper, you only see what happens on the electronics, you are scored on the electronics, and everyone you are shooting against is shooting on the same target. You shoot multiple shots, and so does everyone else. The error is random and unpredictable.
The question is does that amount of error added compared to scoring manually negatively influence the competition? If the random error was large then it would be like everyone is shooting a slightly less accurate rifle and then the results would tend more towards random luck, and the value of the competition would be reduced.
So we can work it out. 3 inches is 76.2 mm. If your "true" group on paper was 76.2mm, then the target with random error of 4mm would increase your group size to... sqrt(76.2*76.2 + 4*4) = 76.3 mm.
This level of accuracy is so ridiculously insignificant in terms of competition. It's not even worth thinking about.
4mm at 600 yards is 0.026 minutes. When you are making 0.125 moa adjustments in F-class, another 0.02 moa of random error is small enough that it will not influence any decisions you make while shooting. It's just a tiny bit of noise that no one will ever notice. At 300 yards, it can be argued that it is noticeable, but just barely.
There are more important things to think about. The real problem is whether the target will throw a shot way out. Practical issues such as shockwave interference from other shooters can not only cause missed shots, but shots to be reported way off. Having 8 mics helps with this. If the target is shaking in the wind, or worse, rocking back and forth, this can cause 1/2 to 1 moa of vertical error (unless you have 8 mics). This is unacceptable. Using e-targets in matches requires properly built, sturdy, stable target frames. It does not require 4mm raw accuracy capability of the system.
Whether it's 4mm or 8mm, both the ShotMarker and the SMT are equally capable of the raw accuracy required for the top level of competition. The question is whether the practical implementation issues are addressed, such as network reliability, protocols for troubleshooting and handling target errors, protocols for crossfires, etc. E-targets bring new challenges in terms of running matches in a fair way, and over the last 3 years I've seen it all. E-targets are the future and to get there sooner we need to accept that raw precision of the system has progressed far enough across the board (with ALL vendors of e-targets, at this point) that it no longer matters, and start to understand what does matter.
I intend the ShotMarker to be used in competition, and I will put my focus into addressing these practical concerns through better software, warning messages and providing relevant data to the shooter/scorer when odd things happen, reliable network, an easy to use interface, educating ranges on what is important when building target frames, etc. I've shot e-targets in a concrete tunnel and the accuracy can be incredible - 1mm or thereabouts. The problem is not the technology, it's how we choose to use it.