Map/match analyzer project

[Cat1981England]

AntiTCC2009r6
TitanTeamFix
ONSPlus_101beta31
UTPlusStandaloneHitSounds
ServerLogo4a
TA100
MutWheeledVehicleStunts

That's everything.

[Wormbo]

Okay, I think it's an incompatibility with ONSPlus - but only because I abused FindPlayerStart as trigger for detecting player spawning instead of e.g. using ModifyPlayer or other means. ONSPlus not always passes the FindPlayerStart call down the GameRules chain, probably for good reasons.

I'll straighten this out when converting the mutator to work in ServerActor-only mode. The additional overhead is likely insignificant and the inaccuracies of that approach are well below what will be displayed on the generated images.

[Cat1981England]

Ok, thanks for the update. I can't see another ServerActor being a problem, the server has quite considerable headroom in regards to both the CPU and Ram.

As soon as you're ready for another test run just give us a shout

[Pegasus]

Been almost entirely preoccupied with AFK affairs for the past few days, but there's one thought/suggestion about this tool that's been on my mind even before then that I forgot to bring up: the feasibility of supporting map-specific data stacking from multiple matches in future versions. Far more informative than just being able to review behavioural patterns in any specific match, with all the circumstantial caveats that can entail (time of day or day of the week and all the variables those influence, player count, other exceptional occasions etc.), having such an option to control for all those factors through averages derived from much greater amounts of data would no doubt provide an invaluable degree of insight and offer a glimpse at the bigger picture regarding the underlying trends as they're shaped by the maps' own features themselves. Here's hoping expansion towards such functionality is in the cards .

[Wormbo]

I don't see any problem with that kind of data aggregation, except that individual players will not be identifiable across several games, other than by their name. In fact, currently they aren't even identifiable across a reconnect.