Game after game cant get the units to move logically, just two examples out of many:
I switched to halberdiers in a game of BF and instead of taking the logical way around the woodline they choose to run straight into the choke i had walled of because i didnt want to get flanked from that side. Lost me the game because i couldnt get my units in the right place and had to manually set waypoints for them to understand. Trash.
I tried to set up trade in a amazon tunnel diplo game and i cut to the guy next to me for trade income. The trade carts didnt like my plan though and choose to run through the middle where all the fighting was going on. It didnt matter if i walled it off completely or closed the gates, the only acceptable way for the trade to run was through the middle. Long story short i almost pulled my hair out after rebuilding the trade 500 times and then lost because i couldnt get gold income when i needed it. Trashtrashtrash fu game.
Yeah the pathing is completely broken right now and so far it seems like the Devs are refusing to address it. Nothing we can do really. Just have to hope they actually read the forums, Reddit, pro player videos, content creators etc. who are all complaining about it.
I’m not convinced a lot can be done with pathing.
I’m no expert in pathing, and I don’t play any other pathfinding heavy games, but basically afaik there are different algorithms out there for pathfinding. In general there are fast(not computationally intensive) algorithms that produce ok results, and slow(computationally intensive) algorithms that produce optimal results.
I don’t have PROOF but I assume aoe2, to provide overall optimal performance, uses one of the faster algorithms.
for an aggressive 1v1 where things typically end faster (and thus at a point in the game where you have fewer units), maybe the performance decrease wouldn’t be too bad if they adopted an “optimal” algorithm, but conversely imagine a post-imp 4v4, 800ish unit mosh pit.
Admittedly I don’t have hard numbers to refer to, but I wouldn’t be surprised if really REALLY fixing pathfinding has some computational limitations. They can work around the edges, make incremental improvements here and there, but FIXING it might require more compute to run than the average gamer can be expected to have.
Regardless of the truth of any of that, I would say it’d be nice if they devs came out and said “Hey it’ll slow the game too much” or “Sorry but it’ll cost 100s of k in payroll to get it fixed and we can’t spare that” or “It’s literally the best the game engine can do” or whatever the reason is. The devs have mentioned in patch after patch that the pathfinding is better in specific situations, and when you compare DE to HD or AOC, yeah it’s hands down better but it seems to be a series of iterative improvements, not “being fixed”. We keep hearing it’s better but not really feeling it on the day to day.
This is a lie. I recently went back to Voobly because DE pathing made me vomit. And guess what? HD/AoC clearly does not have all these bullshit pathing issues we have now, like units randomly freeze while attacking, villagers zigzag rather take straight path etc.
Let’s not find excuses for the devs, and use phrases like “spaghetti code” as a shield. Reality is code from 1999 does more logical things than now.
Yeah this is even worse. Being dishonest upon being bad at your job is just infuriating.
If you disagree that is your prerogative, but I think it’s unwise to assume intent to deceive. If you observe better path finding with HD then I can’t argue what you experience. That is not congruent with my experience but who knows.
I get some people think anything that isn’t a solution to pathfinding is some excuse, but let’s employ occam’s razor. Everybody wants better pathfinding. most of the people working on the game are fans of the game and presumably want to see the game be as good as it can be. I don’t think they’re just not delivering the goods cause they don’t want to. I assume something is preventing this from being easily done. If you disagree then I guess we can agree to disagree.
If that is true, all we want is for them to come clean and be upfront about the problems; to explain what limitations and hurdles are getting in the way of fixing it all and to detail exactly what possible solutions there might be. Because at the moment it feels like they’re just adding shiny new stuff to the game while sticking their fingers in their ears and singing “LA LA LA!” every time someone brings up the pathing problems.
I hear the same excuse of “20 year old code” being thrown around a lot. Can somebody with programming knowledge explain to me why they can’t just write new code to fix this? Either by adding new code to the old code or by rewriting the whole game code from scratch using the old code as a template.
I remember they said they have been doing this (refactoring) work this year. We can’t judge the complexity without further context.
But what we know is that devs can really mess up basic things. Like when range indicator first launched, it is not correct at all. Or the in-game map pool voting displaying percentage before you vote. Lack of common sense for a software engineer.
So I’m actually a web developer. Definitely not apples to apples. But the possibilities that come to my mind are what I mentioned above.
Compute - It might just be that a pathfinding algorithm that produces an optimal result is to compute intensive. The game will run slowly.
Introducing new bugs - I’d say half of my bugs aren’t “well I was being dumb, typo’d, reversed function arguments, missing character, etc” but especially on larger more complex projects you’ll run into “ok so this updated function spits out the basically same result but in a few edge cases it’s slightly different and that makes sense, and that function over there that’s obviously related is either fine with it or I updated that too, but this seemingly un-related thing I’d honestly hadn’t considered ATM, is also referencing that value for non-obvious but important reasons, but I never considered how that’d interact there” etc, etc, etc. So they might be apprehensive to overhaul something so integral to the game.
Presumably they updated numerous things simultaneously and therefore rolling back pathfinding would mean having to roll back something else because it doesn’t work with old pathfinding, but they don’t want to rollback everything so they don’t.
Yeah this sounds wrong to me. I work on an app that has a fairly large user base. And our rule is that we cannot ship a version if it contains critical bugs. The pathing right now is game breaking and should not be considered launchable by most if not all business standards.
I don’t see how regressed pathing has to be bound to some other critical fix that they don’t want to roll back. It feels like sloppiness.
IDK. I played yesterday and the day before and It wasn’t causing me any major problems. From my experience I would not characterize the pathfinding as game breaking. it will occasionally do something stupid but by and large I don’t have too many problems.
Personally I have 10x more problems recently getting double click to work. I typically have to try 2, 3, 4 times before I successfully select the objects in question. Maybe it’s because I play on a 43" 4k screen, maybe I’m just a pixel off or something but the first click seems to work and I’ve been playing on such a screen for over five years and only recall issue in the last six months or so.
When you need to sacrifice 4 scouts to kill one villager, because 3 of them freeze in the air, this is pretty game-swinging and should be considered game breaking imo. You can easily get away with taking near 0 damage when pathing works.
Oooof. Yeah I’ve been lucky enough not to experience anything like that.
Those shotels uphill against organ guns were especially heartbreaking.
If I do something dumb just for the lols, like 4v4 with me and 7 ai 500 pop on ludicrous I see stuff like that. It’s been years since I’ve played that setup. IDK if it’d happen nowadays.
I wonder if it isn’t “pathfinding” so much as it’s lag issue.
That is the thing, lots of voices complaining about the pathing or path finding, but nothing has been said officially by the devs, not even the giant selection boxes for market, town center and siege workshop, once you have archers on ur enemy’s base you will hit the market instead of running or hitting the enemy, the game is in a terrible spot right now, freezing units if they are tasked to attack.
The pathing has never been that good in this game version, it was decent once they fixed the attack move and the knight pathing, cause for like 2 years xbows were melting knights cause they would refuse to hit the units but instead re grouping behind the first knight.
The path that broke the selection boxes and pathing was the xbox cross, which by the way you guys would be amazed to know the number of players that we got from that successful port, numbers aren’t bigger than 3k players and it is decreasing steadily, cause console players play on ##### #isplays with the controller, playing aoe2 isn’t fun to play on 50" inches or more being sat on ur coach, so 70k constant multiplayer users have to suffer as usual for the minorities.
And for those who keep claiming that the game is in the best moment ever, i double bet they do not play competitively or use commands, so for them these problems are not real or from their interests, they just want to keep adding moar tribes as civs wiith copy and paste bonuses, moar charge attacks and more crazy armor and attack for cheap gold cost and if possible flying units.
yeah, I see a lot of “MOAR” too. I can get behind some of it but most of it I’d pass on personally. My least favorite is the big push for unique castle skins, monk/monastery skins, etc. doesn’t add much imo.
