> 2533274801176260;9422:
> That is contradictory. If engangements can happen from opposite sides of the map, then a killed player does not imply the presence of opponents, as he could have easily been cross-mapped.
Maybe there was an “often” missing in there. Now that we’ve fixed that contradiction, the same point still stands: where teammates are dying, there’s a high probability that there are also opponents there. This influences decisions made by the player.
> 2533274801176260;9422:
> I completely agree that there are certain “focal points” for players to accumultate at, but once dead, the game respawns you as far away from the skirmish as possible, so I feel that this compensates sufficiently enough for this issue. It might need to be taken into account when you want to use this formula to actually calculate the mean free path, but not if all you’re interested is proportionality.
You feel, but I’m not convinced. I guess it’s good for you that you feel like it has no effect on the proportionality, but ultimately that’s just it: your feeling.
> 2533274801176260;9422:
> As for the lines of sight, I feel like you’re severely overestimating this issue. There might be some spots on certain maps where you can cross-map enemies, but they are few and far between to be relevant. For every Haven, there is a Skyline. For every Solace there is a Monolith. Even on maps like Abandon and Landfall, large parts of the map are blocked off from view, and being out in the open is the exception, not the rule. Of course this changes rapidly once you switch to Big Team Battle or those kinds of maps, but that is why I focused my analysis on 4v4 Team Slayer, the “bread and butter” of Halo gametypes.
It’s not even about seeing across the whole map, but on most Halo maps there is a huge number of locations where you can see further than half the length of the map. Not to mention, it’s precisely these locations players try to get to because they offer an advantage. Either way, the distance which a player can see almost anywhere on almost any (4v4) map is greater than the average distance between players. There’s just no way you can argue that the interaction distance of players is significantly shorter than their average distance.
> 2533274801176260;9422:
> If absolutely need be, I could repeat the analysis assuming behaviour according to a Leonard Jones Potential, where particles “attract each other at larger distances and repel each other at shorter distances”. But in that case we need a clear definition as to what “temperature” means in this scenario, as the parameters don’t neatly cancel each other out anymore.
Well, it might possibly be less wrong, but that doesn’t make it a reasonable approximation of reality.
> 2533274801176260;9422:
> EDIT: By the way, I just realized, the effect you’re describing just increases the downtime. If 4-6 players are fighting in a room and the game respawns you on the other side of the map, statistically speaking, you are less likely to run into another enemy. Especially if every player would just beeline for said room, since their paths wouldn’t cross until they arrived. And the larger the map, the more prominent the effect. If anything, my estimate gives a lower limit for the mean free path and the downtime.
Now you’re just assuming that the situation is always going to be everyone packed in a single room. But there are two problems here. First, when I said the presences of teammates (often) implies the presence of opponents, I did not say that it implies the entire opposing team is where one player just died. I meant that it implies that there is (with significant probability) at least one member of the opposing team in that part of the map. The other three might still be anywhere else. Secondly, you are using one special scenario to argue something about what happens statistically. Sure, if I have gas in a box, there might exist a state where the molecules all move vertically from top to bottom, colliding almost never, but the possible existence of such a state doesn’t imply anything about the mean time between collisions.
> 2533274801176260;9422:
> I’m sorry, but we’ll just have to disagree there. I see no fault with my model, especially since I observed the exact increase in MFP/downtime that I expected from the assumptions I put into it. Could it be coincidence? Certainly. Will I have to re-validate the model in future games? Definitely. But so far, both theory and observation point towards the same consequence: That - within Halo(!) - the disproportionately scaling in levels due to the inclusion of sprint has led to an increase in downtime and thereby slowed down the overall game.
Tell me again, how have you made these observations? Have you gone and played a bunch of games, recording the time between each encounter? How many matches is this based on?
Maybe I’ll have to go dig some game histories to see how game speed has changed over time. But as I’ve said, this far I haven’t see any data that would bump me in either direction about whether gameplay has even become faster or slower over time.
