Atlas Passive Tree - Save Layouts

I swap between two strategies. One for juicing Deli maps and the other for speed farming beasts, essences, map bosses for alva missions and guardian and synthesis maps. It's a hassle to respec my tree manually every time I want to make the swap. I feel it intuitively makes sense to have the ability to save a tree layout and swap to it as easily as choosing it from a drop down menu and paying the respec point cost.

I like this idea in principle, although I'm not sure how it could be implemented with only Orbs of Unmaking.

Trying to determine exactly how many nodes are different between two arbitrary sets of passive trees might not be a simple task. The absolute difference in nodes could be calculated, but the path to get to them could throw everything off dramatically. For instance, let's say passive tree A and passive tree B only share 20 nodes in common. The calculation of the number of Orbs of Unmaking required to build tree B out of tree A could be different from the simple difference of (total nodes - 20).

For example, as with the passive skill tree, it's possible that you can't just undo 3 nodes and assign those points somewhere else. I've often had to undo 3 nodes that I actually want to keep, fill in a duplicate path with those 3 points, undo the 3 nodes I wanted to undo originally but couldn't—because of a dependency between one area and another, and then redo the 3 points I'd originally removed in order to accomplish all of that. In short, to change this particular tree in the way I want, I'd have to spend 6 Orbs of Unmaking, even though it's only 3 nodes that are different between tree A and tree B.

I predict that if something like this were to be done, as a kind of snapshot, there would have to be a new kind of currency. Just as we can save voidstone modifiers to a Surveyor's Compass, you'd save your passive tree to a blank Atlas Orb (or something like that). Using a saved Atlas Orb on the atlas passive tree would then restore the saved tree, overwriting the current one.

But the cost of doing this could be something quite different from what you'd expect. Perhaps one Orb of Unmaking for every two allocated nodes in the entire passive tree, in addition to the cost of the blank Atlas Orb itself. And I suspect that while you might even be able to trade these (imagine giving your atlas tree to somebody else), they could only be usable if the atlas tree they were applied to had enough nodes available to accommodate the saved nodes. (If it had more, the saved tree would be applied, and you would get extra points.) Also, given how GGG has implemented other things like this, I suspect the Atlas Orbs would be consumed when applied.

The cost of doing this would be the same, regardless of the number of differences between passive tree A and passive tree B. So, it would be exorbitant to switch between two trees with a relatively small number of differences. For that, you'd want to do it manually. However, it could be worth it if the trees were dramatically different.

The problem you mention is only a problem for trees that have very little difference. In your example where only 20 points are in common, that means there are up to 112 points not in common. It is physically impossible for you to be forced to unpsec 20 of the nodes you are keeping temporarily to resolve pathing issues. If you are forced to do so, that is because you inefficiently allocated your free points. Which means this is only an issue as the trees become more similar. Therefore this is easily resolved by having a base cost of unmaking + differences. Ie. it cost 10 orbs to swap trees + the cost of differing nodes. In large changes you are paying a 10 extra to forgo needing allocate manually. For small differences you may pay extra or the same. There is a very small chance that you would actually pay less than physically doing it. If it cost 20 points base, it would be virtually 0 instances where manual respec would cost more than saving a tree.

But if you wanted an exact algorithm to calculate exactly the points you need every time, it is actually not that difficult to make one. After all, it is only comparing 2 trees and the points needed to go from tree A to tree B. It isn't calculating the most efficient path for you or anything. It would look something like:
Calculate all the points you can respec without breaking pathing.
Spec points for new tree prioritizing pathing points (this is the hardest part to actually code. Determining what is a pathing point)
You would just loop through those 2 things. After each iteration of the loop you would update the tree diff between what you have and what you're going to.
If there was an instance where you get no points returned from the calculate loop then the system would unspec a single point that is not a pathing point and continue.

Since this is a process that wouldn't perform often it isn't even important to improve the speed of the algorithm. No matter what, it will be faster than you manually doing it. I think anything below 10 seconds is acceptable for this and with what I just mentioned, I don't see it taking more than a few seconds. There aren't that many points we are dealing with after all. The time complexity is about O(log(n)n). Not exact. It's just better than n^2 but worse than n. n being the number of points required to respec.

Tbh, time complexity doesn't even matter. You could simply program it where every time you change your current tree, all your saved trees will run the algorithm in the background against your current tree and therefore each tree can display the points required to swap over in a live update manner. And when it comes to actually swapping trees, it would be instant with no calculations done. Just points removed and a tree change.

Not being able to easily change your tree limits what you can farm and stops people from dominating all content at one time which creates opportunity for other setups, This is needed to keep the game healthy and GGG has finally done it with this atlas passive tree.

You're clearly asking for this system which limits you from dominating more content than your atlas passive tree can handle to be made easier which i understand but that is kind of the point.

the option discussed here simply should not be allowed to come to pass and one thing they could do which would help a great deal is to make the atlas passive tree unique per character allowing you to have a second character for your farming strategy at the minor cost of leveling it up.

"

SilentSymphony wrote:

Not being able to easily change your tree limits what you can farm and stops people from dominating all content at one time which creates opportunity for other setups, This is needed to keep the game healthy and GGG has finally done it with this atlas passive tree.

You're clearly asking for this system which limits you from dominating more content than your atlas passive tree can handle to be made easier which i understand but that is kind of the point.

the option discussed here simply should not be allowed to come to pass and one thing they could do which would help a great deal is to make the atlas passive tree unique per character allowing you to have a second character for your farming strategy at the minor cost of leveling it up.

If it did happen i would laugh as the cost of these orbs goes up significantly until they are unobtainable by your average player in any significant number.

You are making assumptions that aren't true and also speaking on behalf of GGG determining their intentions which they haven't said nor do I agree with. I don't know why you state it as a fact.

1. For the opinion that the atlas passive tree is there to limit the player in anyway, I don't know where you got this idea from. That's like saying scarabs are there to limit the player because you can only use 4 of them. You are trying to draw a correlation to a system having built-in limitations to the system being made to limit the player. Sextants, scarabs, atlas passives. They are all made to enable the player. Not limit. If a system was made to limit then it would reduce other mechanics power while increasing others. Ie. Abyss scarabs would add an Abyss but now allow a breach to spawn. But that isn't the case. PoE has no such mechanics.

2. The system I propose does not change anything except the quality of life. It doesn't allow players to dominate all content at once as you talk about. You are simply swapping passives. Things I already made clear that I do and the majority that I have observed do as well. I'm not even asking for a system to dominate all content. I want essences when I'm speed clearing toxic sewers and I want flash breach when I'm juicing deli promenade. It would be pointless to have these nodes in the opposite strategy. I'm going to make these changes regardless. No one is going to gather 50 guardian maps through normal play and then run them on their atlas tree that isn't spec'd to the guardian nodes. They are going to swap their passive nodes to get the most out of it. If you think swapping passives to different strats counts as dominating all content, then everyone is already doing that except you I guess. That is the reason I said that it "intuitively makes sense to allow people to save atlas tree layouts". We are already constantly swapping.

3. Your speculation on unmaking orbs is laughable. People are already constantly swapping their atlas around. Saving layouts would have virtually no impact as it is just a quality of life improvement on a thing people are already doing. You can just buy unmaking orbs with regrets and regrets can be bought with scouring orbs. It's almost impossible for them to become unobtainable.

"

KaseyM21 wrote:

2. The system I propose does not change anything except the quality of life.

I don't think its the right direction to go, You don't have to agree.

Everything is limited to some amount giving player agency in how you fill those things in, We cant have everything and that is a huge part of this game

There's got to be better solutions than a button for those who can afford to vomit currency at their tree, if you disagree with that then i guess we'll just have to disagree

Your frustrations are legitimate, I would like to see them alleviated but not if it benefits the top % of players while doing nothing for your average player.

"

SilentSymphony wrote:

"

KaseyM21 wrote:

2. The system I propose does not change anything except the quality of life.

I don't think its the right direction to go, You don't have to agree.

Everything is limited to some amount giving player agency in how you fill those things in, We cant have everything and that is a huge part of this game

There's got to be better solutions than a button for those who can afford to vomit currency at their tree, if you disagree with that then i guess we'll just have to disagree

Your frustrations are legitimate, I would like to see them alleviated but not if it benefits the top % of players while doing nothing for your average player.

It's not a matter of whether we agree or not. Having limits within a system is not the same thing as a system limiting the player. That is the core issue I have with your statement. How to tell if a system is a limiting one or an enabling one: take away the system, is the game more limiting or enabling now? If you think the atlas tree is a limit on the player then the answer would be that the player is more enabled without it. That is obviously not the case.

Your opinion is objectively wrong because it's based on false conclusions.

You seem to be under the false impression that this only benefits a specific strata of players. You are wrong. This is making NO changes to the current system. This is not a topic where I am advocating for them to make any changes to the current system. It is only a quality of life change that mimics what is already happening in the game. Everyone benefits from it. I don't know why you keep ignoring the fact that this is only a quality of life change I am proposing. Not a redesign of the system.

If you want to advocate to see changes in the system itself, that is a completely separate topic. You are simply disagreeing with me because I'm not solving the issues you have with the tree when I was never trying to. You want separate trees for different players? Fine, I don't care, that and what I propose aren't mutually exclusive. After all, my change is only a quality of life change. Yours is changing the system itself.

"

KaseyM21 wrote:

The problem you mention is only a problem for trees that have very little difference.

Any solution has to include all of the people who are playing the game.

"

KaseyM21 wrote:

In your example where only 20 points are in common, that means there are up to 112 points not in common.

Only if you've completed all content. The majority of players haven't. Again, you have to come up with a strategy that applies to everyone.

"

KaseyM21 wrote:

If there was an instance where you get no points returned from the calculate loop then the system would unspec a single point that is not a pathing point and continue.

Yes, an algorithm is possible. However, my opinion is that GGG will never invest in the time needed to implement this kind of solution. What I proposed is far more in line with how they've historically implemented such things, and would require far fewer resources.

"

SilentSymphony wrote:

one thing they could do which would help a great deal is to make the atlas passive tree unique per character allowing you to have a second character for your farming strategy at the minor cost of leveling it up.

I both do and don't agree with this. I'm not sure which strategy would be most wanted by the community: each character sharing an atlas passive tree, or having a different one for each character. I can see pros and cons to each.

From a thematic perspective, if the layout of the tree were not shared, it would make equal sense that the number of passive points unlocked should also not be shared.

Since I mainly only play a single character, this isn't much of a concern. But I do have other characters, and they aren't as powerful as my main character. Were I to switch over to one of them at this moment, I'd have to undo several of the things in my current tree before playing them.

If I think about it, I kind of like the additional challenge of having to juggle the tree around based on my current needs.

However, having said that, if I completely redesign my tree, and the monetary cost of doing so is the same (either with the current manual method or some kind of easier "select a tree" method), I don't see a downside to having an easier method. But I do think that the monetary cost should be high enough that it simply would not be worth using it to switch between minor variations of the tree. It should be an expense that is only justified for significant changes.

"

Only if you've completed all content. The majority of players haven't. Again, you have to come up with a strategy that applies to everyone.

What kind of argument is that? Guess we can't have an atlas because not all players can get there? it's a quality of life addition. Not a change. All players can use it. It doesn't matter where your progression is at. If you have access the atlas skill tree, you can use it. It is not adding a new mechanic. It is improving a mechanic to make it easier to use.

"

Yes, an algorithm is possible. However, my opinion is that GGG will never invest in the time needed to implement this kind of solution. What I proposed is far more in line with how they've historically implemented such things, and would require far fewer resources.

It's really not. It doesn't really make sense to create a new item for this when like I said, just adding base cost to the difference is a much more simpler and elegant solution. What you are specifically proposing is more complicated than the algorithm. In simplicity order it would be:
base cost -> algorithm -> creating a new item.

This started off as an interesting discussion, but has quickly devolved into an attack by the OP on anybody who disagrees with them and who only wants to see things a certain way, so I am dropping out of the conversation at this point.