fig0451

Also desync.

Is the Stary/Novy Sobor spawn still a thing? I haven't died in a while so I'm not sure.

With the exception of being on the north edge of the map, I don't think there is anywhere like that on the maps. And even there, you can still move to a new loot area quicker than say an hour respawn timer. @alpha: Hopefully the 'quadrants' aren't literally huge quadrants of the map, and are more localized.

Going by my interpertation of the quadrant system, a looted item is anything that's been moved from its initial spawn position. So yes, if you picked up something and then dropped it, it would count as moved. Players could trigger a premature respawn ... if they knew all the items in a group. So if one house was one quadrant, thats maybe 10 item spawns, and if you did that, then sure, you can get some more things in an hour. However, there is a relatively simple counter to this: Don't make a house one quadrant. Make one item in House A, 3 items in House B down the street, and 4 items in House C 200 meters away in a quadrant. You'd be nuts to map out every quadrant if they were all randomized like that. Also ... is this really a reasonable thing for a player to do? If they had the inventory space, wouldn't they just take items? And if they don't have the space ... why are they trying to get more things? They want different things? Why not just go to the next house and check it instead of waiting for an hour for a crapshoot?

Kawai'i AK74-SU

Ok, even assuming you are doing this the most intelligent way by making ONE table of viable spawn locations, and assuming that there are only 1,000 item spawn locations on the map, you still have to do a three dimensional spherical trace from EVERY PLAYER to EVERY EMPTY LOOT SPAWN. Assuming a full server at 40 (and remember the devs are aiming for 100 person servers!) that's up to 40,000 spherical traces + the 1,000 "is there an item here already" checks. Exactly it would be #LootSpawns + (#Players * #EmptyLootSpawns)^, where ^ denotes computations that are much more intensive. And no, your sanity check doesn't count because you haven't described how you get your acceptable locations. What current list? Where? Determined by what algorithm? Compare that to how the quadrant system is likely to work with your numbers, which would simply be 1,000 yes/no checks on a list. Your method is basically ludicrous in comparison. In reality, there are probably 1,000 spawn locations in each large city, I'd wager probably 10,000 spawn locations, at least, throughout the whole map, especially given that something like half the buildings in the game aren't properly spawning things. That's 10,000 * 100 for the goal of players in the server, which is literally a million spherical traces. Oh I forgot, that's in addition to the 10,000 "is there an item here checks". Again, compared to how the quadrant system is likely to work, thats 1,010,000 operations compared to 10,0000. And you are using sphere traces, which means calculating a line between the player pos and item_spawn pos, which is more mathematically intensive then just checking for a 1 or a 0. 1) and 2) are literally exactly the same, stated in reverse. And it doesn't matter, as your method of achieving it is ludicrously computation intensive and almost certainly not workable. You say you're a developer but it seems like you have literally no experience with the Arma2/3 engine. As for 3) and 4), have you ever considered that there being high and low traffic areas is a good thing? It's nice to have a general sense of areas that are dangerous and areas that are safe (even though nowhere is really ever totally safe). It sounds like you're trying to force the game toward exploration and spreading out across the map. IMO it would be kind of dumb if this plan succeeded and made the entire map have roughly the same player density. Differences are interesting! EDIT: Also upon further consideration your system is inherently self-defeating. The goal of your system is to spread players out everywhere, so we have a more even distribution that's less weighted toward what are currently hot-spots due to loot and/or proximity to spawns. Well ... if players are spread out everywhere, your item spawning system will be even less able to actually spawn items. If there are people evenly spread through out the map, far less areas are going to be possible candidates for a new item spawn (as at least one player will be nearby).

I don't really see how you can say that, given that you know literally none of the details, as you point out in another of your own posts. No one knows. It could literally mean empty, or it could mean 75% of the items have been taken leaving 25% of the items still there, or any other number. Quadrant could restock known empty positions and leave the things that are still there ... still there. Better question: How big is a quadrant? If a quadrant is large, it would make sense to have a threshold level as I've described above. If a quadrant is small, that threshold could be smaller, or absolute. For the first two, the answer is probably yes, and the answer to the third is probably no. Why? Fairly simple: The quadrant system probably only cares about item spawn locations. So, things in the place they are spawned count, and other things don't. This is optimal because then to do a respawn check, the system only has to check it's known locations, as opposed to doing an intensive search to find all items in the map and then define whether or not they are in some geographical area. It's much more likely that a 'quadrant' is just a defined group of item spawns, as this would be less intensive on the script. No one knows, but a reasonable solution seems to be something like ... oh I dunno, say an hour delay. Every 15 minutes have each quadrant check to see if it is empty, or meets the emptiness threshold as I described above, then that starts an 45 minute timer for it to restock. If Timer is in effect, quit checking, until the spawn happens where one last check is done to see if any additional items have been taken, then you've got you're proper 'put new stuff at these locations' instruction. Probably not, that would be really stupid. 1) As I said above, quadrants are not likely to be actually defined by some kind of geometrical container. It's waaaaaay easier to just say item_spawn_1 item_spawn_2 and item_spawn_3 are part of quadrant 1, and item_spawn_4 and item_spawn_5 are part of quadrant 2. So, it's entirely possible that a city could be one large quadrant, or a city could be made up of many smaller quadrants. 2) That depends on if the respawning system is part of or tied into an item persistence system. We don't really know about that right now, but that is certainly within the scope of this sort of thing. This question is misleading though. Why do servers restart? Primarily, to restock items. If the restocking system works ... we don't need frequent restarts any more. So this question is moot. You describe a situation literally identical to how the current system works (OMG it's hard to get items in a highly populated area! People might fight and stuff!) ... ... and conclude that this will result in more KoS and less items. Pretty much nonsense. Let's say the quadrants restock in an hour as I hypothesized above. Currently, most servers seem to reset every 2 hours. That'd mean we'd have twice as many items compared to a server that restarts every 2 hours. Or hey, even if it's a 2 hour respawn, if it's tied into the item-persistence system, we still don't have to restart our servers any more unless we REALLY want more items. Come to Joe's Stop and Shop 30 minute respawn server! Again, nonsensical, servers don't have to restart all the time if the quadrant system is implemented. TL:DR; your solution is hamfisted and inelegant. Continued Below!

Can't wait for the next one :P

Certainly the most mild-mannered hacker I've ever heard of, hahahah!

It's an early alpha and everything that strains network performance is subject to massive amounts of optimization and complete recoding.

Fires 9x19, so it's a submachinegun.

Yeah that seemed a bit strange to me ... An M4 should be a bit slower, but not much. Maybe it's to compensate for all the pistols being hilariously inaccurate at the moment?

Someday my .357 won't be woefully under-powered and laughably inaccurate.... someday.

It's probably going to be a hunting crossbow, which realistically, would be effective out to about 65-70m (and that's pushing it a bit) assuming you are an expert and have a really nice xbow. In other words, similar to the effective range that a pistol currently (though unrealistically, especially for the 357) has right now. Medieval Crossbow != Modern Hunting Crossbow. Gameplay-wise it will likely make an excellent close range assassination weapon, and useful hunting weapon once animals are added.

It's possible I was one of the three guys. From our perspective, we were looting there and then an unknown showed up and started shooting at us, so we all returned fire. My hardcore experience is otherwise incredibly similar :D