chrstgtr

I don't really care. But others seem to. Regardless, I think buildings were well balanced before the mod change. Yes, non-random arrows. It seems, at best, to be preferred as much as random arrows but more likely disfavored compared to random arrows. Keeping random arrows doesn't require additional balance changes This has been the problem the whole time with this change. It was widely held that rushes were balanced before the change. And, if you have a problem with the hero aura then the hero aura should change. Buildings working against armies without hero indicates that it was properly balanced before. Your proposal just feels ptolemaic. At the end of the day, I want to be able to rush the units around a CC in p1. I don't see how that can ever be possible without losing a bunch of units. I also don't want to lose a considerable portion of my army guarding rams from a couple garrisoned swords in p3 after I have clearly won a battle of units.

I always thought this should be true. But that's a different issue from buildingAI behavior.

There was a pretty explicit change to unify building arrows recently: https://code.wildfiregames.com/rP25309 Personally, I'm ready to walk away from any change to buildingAI. Random vs. nearest unit buildingAI just depends on personal preference and, after some testing, the player base doesn't seem to want a change in behavior. As it has been said in other places, no one seems to have a problem with rush or late-game building-unit balance, so this change also doesn't "fix" anything (aside from behavior preferences). But as we have seen, it can introduce a problems with building-unit balance. At this point, this seems that this doesn't fix anything, it creates balance problems, and it changes buildingAI to a disfavored form. Seems like an experiment that was worth trying but has run its course at this point.

I don't think anything has changed. It has been a known problem for a long time. It is just being reported here now. To the extent the mod did cause more occurrences, I think it is because the new mod makes some games last longer/more units to be produced/etc. (i.e., the mod is causing the game conditions under which the error occurs instead of the mod itself causing errors)

Got it. Sometimes game-ending OOS errors show that the host and only the host is OOS. But other times everyone, including the host, is OOS.

That makes sense. But why would it happen to all players at the exact same time? If it was one player causing it, wouldn't it just list that single player as OOS?

Yeah, it was unartful language by me. My fault. It is the OOS error @real_tabasco_sauce @Barcodes, and I mentioned earlier in the thread where all players go out of OOS around min 40. For the 10 minutes before that happened, various players went OOS or actually crashed. Sorry.

Yes, I reach the normal "You have reached the end of the replay" message, which makes sense because the game never truly crashes--it goes OOS and makes it impossible to continue

16GB

Mac host. Game crashed at min 42:11. For about the last time minutes of play, other players were dropping like every minute 2024-03-05_0011.zip

Please not range. Camping from a safe distance discourages fights. Also, long pack times are just annoying. No one likes playing with units that don’t feel responsive—it almost feels like lag is built into microing cata

Yes. It also only tends to happen in 4v4 games (i.e., 1v1s that last 40 minutes do not go OOS)

Mac hosts using 64 bit versions have also crashed around the 40 minute mark. The real problem @Barcodes is referring to is when all players go OOS around that time mark. Typically, a few players will go out of sync before that but they can leave and rejoin to continue. For some reason, around the 40 minute mark ALL players will go OOS.

I think the wait and see approach is the way to go--opinions are still forming. With the melee patch, it is clear the community wants something different than community.mod.v4. It's not clear the community wants something like community.mod.v6, though. With the buildingAI patch, it isn't clear if the community wants any change from community.mod.v4.

That is a separate issue. I don't think anyone has an issue with being able to override buidlingAI. But there is a lot of debate around whether buildingAI should shoot randomly or at the closest unit. It isn't clear what the preference of the community is, and if the "closet unit" system can even be balanced in a desirable way. But buildingAI was certainly changed to address a units "don't die" problem. I personally don't think that was ever really a problem. But now we might as well see if this other system is better/workable. In short, everyone should vote in the poll in the other thread once they develop a firm opinion. There is no clear right answer--people have different preferences for legitimate reasons. The community at large should decide. EDIT: Sorry, I didn't all the way down to the bottom of your comment where you said random buildingAI with player override was a good compromise. It could be. I think player override should exist. The buildingAI should be community determined.

Disagree. But assuming you're right then armor should increase. The underlying point is that changing hp will certainly have downriver effects that are unintended and will cause a cascade of imbalances. The underlying cause of quicker battles (whether that is armor or melee dmg) should be addressed instead of changing yet another variable. I don't think anyone disagrees on this and @real_tabasco_sauce hit it on the head when he said that melee rank up was an old hot fix for melee balance that didn't actually fix the underlying melee balance problem. I've always said that melee was strong against range and that the problem was melee's inability to reach range units. All of these other melee advantages v. range get enlarged when melee units actually get engage range units.

The problem with changing hp is that it has a lot of downstream effects. It changes building/unit balance, inf/cav balance, melee/range balance, etc. It just seems a lot cleaner to to nerf the melee attack dmg. Otherwise, we just create more problems for ourselves. Also, the changes of the melee patch suggest that changes to melee attack dmg are the problem. The melee patch gave a major buff to melee BUT gave a nerf to range attack dmg. That means units are dying faster because of melee's buff despite range's nerf. The problem has to be that melee are killing range too quickly. The old meta had melee killing each other in the middle before breaking through and slowly killing the range. The new meta still has melee killing each other in the middle before breaking through and quickly killing the range. Sniping is present in both metas but that doesn't change anything. Battles are only ending quicker once there is a breakthrough and melee are able to rip through the range.

Iphri already got a nerf. It just isn’t in the mod yet. I also think we should consider just a general nerf to melee attack dmg instead of increasing health. It will have the same impact of making battles last longer without further nerfing range units. Melee is stronger now because once they get to range units they really rip them apart. Giving melee more health will let melee do that more effectively.

The more I play, the more I think this is the answer. Not sure if it is necessary for towers but I am leaning towards no. I don't think the meta has really been figured out yet. I think we still need to wait to see how things shake out. With that said...the below are all things that I was independently thinking. I also think there is a general problem with how champions aren't being made as much anymore. I like how melee is being made more now. But that diversity has come at the cost of less cav and champs. Personally, both neither system is better than the other for me. But I do think we could make the current system better to address the cav and champ issues.

Again, this is your preference. Some would say the old version makes a more dynamic game, which lends to more comebacks, strategic dives under CCs/forts, etc. etc. Let's let the community decide what they prefer. My comment above was responding to @Philip the Swaggerless where he said he thought CC dmg output would be lower but believed that the dmg output is actually higher in the current mod. That is incorrect. The dmg. output is lower in the current mod compared to previous versions. But, in the current mod, the kill count is higher, quicker. So, in the current mod, buildingAI is both stronger (kills units earlier) and weaker (takes longer to kill all units) than before. That stronger and weaker function will always exist with current buildingAI behavior (unless dmg output was increased so that it was stronger/stronger but that obviously won't happen). If we want the current buildingAI behavior then we have to get comfortable with that paradoxical strength/weakness.

It was. Buildings are at once stronger (kills units earlier) and weaker (takes longer to kill all units). That functionality is something that will always be a problem.

I’m still learning/adjusting my thinking, but right now, I think this might make sense. I’d like to see how a rush on the wood line works when someone 2x towers it, esp when it is an inf rush. For various reasons, I’m not liking any if the other ideas you mention. I don’t think I’ve heard any complaints about focus fire, which is a feature long overdue imo. I have really big concerns about this right now. I additionally have really big concerns about it making gameplay too turtle-ly late game because a player cannot stand under CCs/forts for any real amount of time, so melee taking out a CC is much harder and defenders can take out rams too easily.

I would recommend reserving judgment for a little bit and to cast your vote in a week. Right now, I would be one yes and one no for the two big changes (melee/buildingAI) but I could see both, either, or neither of my opinions changing so I haven’t voted yet. No matter what, I think this round of testing is a massive success for the community mod concept. We implemented two radical changes and we are getting real play testing to confirm, modify, or reject the proposed changes. It’s the first time it’ll be an actual experimental mod and not just an avenue for quicker updates.

Honestly, I think the best way to control sniping is to just make it an inherent part of the game that unitAI does automatically. There was a forum topic awhile where no one was able to agree on it--I think we should revisit it.