As Zaodon quite rightly pointed out, what you are looking for is achievable using the lootfilter.
I too only started playing in May, and although I do spend a considerable amount of hours p.wk playing, it was like any game a learning curve. Once you’ve tried a few builds you soon get used to the types of upgrades you look for in gear progression. It’s then a matter of writing/editing a loot filter to give you that.
The way I approach my loot filters is a layered approach. At the very bottom I basically block everything. Above that I have a section that will colour change anything at all with shards I am low on for crafting of any level, above that when starting a new build I have a section recolouring any normal/magic bases I could use for crafting. These are my “roots”.
My next highest level is each item base category of each type I use that have 1 out of the 4 affixes I need at T4+.
Above that I have all bases that have 2 out of the 4 affixes I need at T3+.
Then above I have those bases that have 3 out of the 4 affixes but with no Tier filter. “Green recolours”
Second to last section is “endgame bases” which have 4 out of the 4 (or 5,6 depending on item) affixes I need, again with no tier requirement. These tend to be quite rare drops, as they are items that require no purging and are good to try just upgrading immediately. These are normally my “Red” colour items.
My top section is “endgame” which have each item group with 4 required affixes that are all minimum T3. To be honest I have yet to see one drop, but when they do I am sure I will be quite excited. These are the items that I know I stand a fair chance of making T20.
Of course, I also have sections for T6/7 and certain uniques I want extra highlighted.
The use of recolour in conjunction with the lowest level of blocking almost everything enables you to simply play around and know immediately what is dropping and how excited to get. The lootfilter is exactly what you make of it. It can be a simple tool, or it can be highly complex and enable you to identify items simply by their colour. Personally I have come to love it, and my filters evolve considerably as I play each character.
Here’s an example of one that I am using currently on my sorceror based on Bear’s ED character.
Yeah, that’s a good call. The problem is you don’t have a good idea of what valuable affixes are that early on, or that they even exist, and when you first look at that list of 652 affixes you may just go “oh no”. Being overwhelmed is generally worse than missing a few good affixes, because once you learn the system and end up playing a lot more you’ll quickly make up for lost time.
But game wise, I think it’s a similar problem to the idols. For + skills in particular, tier is just not as relevant, and I feel there should be something in the loot filter to reflect that, if some affixes are on a level higher than others (and there do seem to be some shards that are much more rare).
I just make a rule and check any affix i need in the Class Specific section (all rare ones are usually there). Then add some affixes I need too like Damage Penetration or Dodge on potion on belts etc.
This is exactly why I am opposed to lootfilters (and the need for them due to abundance of trash drops): you are creating a game for people who like to code. All the rest of us (majority of the population, I assume) are left in the dark and at the mercy of “loot filter makers”. The loot filter is as complicated as comparing items in the game, when looking for an upgrade.
I think a solution would require a huge rework of the system, where you’d have “Major stats” that can be summarized in “Power” and “protection”, and “minor stats” that represent “utility”. The minor stats could be plenty, and be there to create variety in builds and playstyles, while the major ones would be the ones you compare initially to see if your new loot is an upgrade. This would be sufficient for the more casual players, and the more experienced ones could dig deeper into comparisons and filters to tweak/optimize their builds.
Idea: if you want a loot filter for all players, make an interactive one, where you can right click an item dropped, and “mark it as trash”, or “mark as valuable”. Use Machine learning if you need to, it will help.
I’m hoping LE will not turn out to be a game where you need to optimize to get there (“end game”, max level, final boss, etc.), but where optimizing will get you there faste, easier and with more power. Instead of games like POE, where not optimizing means you’re locked out of any content even remotely approaching that “end game” state,
I mean, I’ve never played an ARPG that didn’t have a mess of loot dropping that you had to sift through. It’s effectively core to this genre. You can remove the loot filters and you’ll still be left with that problem.
Personally, the loot filter is brilliant, but I’ll admit I’m a software developer so I indeed like to code. But I think rather than removing the loot filter entirely, the game could offer some good nice defaults. Most players are already used to, I think, concepts like “I only want to see magic items” from modern ARPGs, which is nothing more than a very simple loot filter.
So I rather think some of the affix areas can be tweaked a bit to more clearly represent a general “power” coefficient so that simpler loot filter settings could then be applied to them and given as defaults. And maybe some understanding that defaults are not necessarily ideal and if you truly want to min max you make your own filter. Us casual players are usually quite fine with “not ideal”.
I think mostly your average player will go through these phases (and just assume everyone wants unique/set/exalted at all times, forever, lol):
I don’t need normal items anymore
I don’t want to see any items irrelevant to my class unless they are super shatter worthy
I don’t want to see items with barely any tiers because no way they won’t shatter before I get anywhere → I think this is something the magic-rare system should be able to represent, tbh. But still kind of leaves the problem of “I still want to see +skill items”, which is why I think that should be systemized in some way
The tough one that cannot be easily resolved is this one:
I don’t want to see items not relevant to my build
Because the game has no way of telling what your build is. But a lot of really basic stuff would go here, like “I only need to see 2H weapons”.
Your “major/minor” stats idea seems to be an intriguing one and could well have a lot of merit. In fact I think I vaguely remember playing a game many moons ago that used to have different colour stats for this purpose, but I forget which one.
As far as coders etc go, yes I will admit to being somewhat of a geek. In fact I actually do enjoy having a lootfilter system that can be as complex as the one we have. However, it should be pointed out that you can also make very simple lootfilters that also do the job. I think that’s the beauty of the current system they use, you can go as shallow or as deep into it as you want.
There’s always going to be a certain amount of complexity in items and gear. This comes with the territory of creating and giving diversity in any game to enable players to make their own way and style of playing. It’s a part of min/maxing. However, it IS an elective process. If you want to simply play a character, you can play it with any old gear and find things that look as though they would work for you. In order to play at higher levels though with consistently less deaths/more kills you will always have to refine things, finding the right gear types, the right affixes, etc.
Sadly the alternative to this is to have less build options, gear without crafting, possibly gear without affixes altogether. I’ve played arpgs like that in the old days, before things got more advanced, and they can be fun if you play in a guild say with large groups but that was the main focus of those games. Playing solo without build diversity and gear diversity soon becomes very boring.
Sadly, I think it’s a matter of “damned if you do, damned if you don’t”. The compromise is where you add in all the diversity, but then give the players the tools to either make it complicated or not (and that is what the loot filter system does). I think LE have quite a nice balance of this dilemma thus far.
Yes, such statements are exactly what a player is thinking. The challenge is to translate them into code, which is far from trivial. Once you start trying to “code the human utility function”, you find out it’s usually more complicated than initially thought, contains contradicting rules and even worse: that the human in question is unable to actually tell you the exact rules or utility they want. As a programmer you probably experienced the “client vs developer” dilemma product owners live through
Yep! I think this is a very valuable topic that I imagine developers themselves have a lot of thoughts on. And ultimately it comes down somewhat to collecting user feedback, and I hope they have in mind players of different kinds and backgrounds to see what kind of thoughts and complaints people have. That’s often why you want to hand the game over to a random buddy who doesn’t play too many ARPGs and see what their pure reaction is. It’s not a trivial problem by any means, but an important one depending on which audience you wish to attract.
The Loot Filter interface for LE is not “code”. It is a User Interface which makes setting up filter rules simple and fairly intuitive. You are playing LE on a computer. You can click the Windows “Start” button, double click the LE Icon to launch the app, put in your ID and Password. None of that is “coding” either. Its computer use. The loot filter is computer use (app use, technically. Using the user interface.) It isn’t “coding”.
Seems a bit pedantic. They said it’s a system that is designed for people who like to code, which I"ll agree with. But it’s a system that requires an application of logic and conditionals, and that’s a big chunk of what programming consists of.
People can flag emails as “Spam” using a UI.
People can have a browser “Remember my password” by using a UI.
People can use Last Epoch Build Planner to plan out their builds by using a UI.
As a developer, using “logic” (in my opinion) isn’t indicative of “coding”.
Being able to write:
function filterLoot () {
var items = new List();
var item = new Item();
for (let i = 0; i < items.length; i++) {
item = items[i];
if (item.properties.tier > 10) {
etc…
}
}
}
That’s coding.
Saying:
Action: Hide
Type: Helmet
Affix: Level of Zombie
Advanced: [checked]
Total Tier: less than 4
That isn’t coding.
Edit: I do want to say that improving the Filter UI is certainly possible to make it easier to use, so would support that fully. But as I find it easy to use already, I can’t give specific feedback on that.
Fair enough. I maybe should have written “pseudo- coding” or “programming” instead. You are translating vague thoughts, like “show me only good loot and mark the items that are better than I have in a special way” into (a series) of rules. I don’t want to define rules, other than the sentence I wrote. I want the computer (the game) or the game designers to do the translation into rules and then into code for me.
Thanks for the tips explanations and suggestions. It looks like that I’m currently not using the lootfilter to it’s max potential so I’ll see if I can narrow it down more. I do struggle a bit with what to put in the lootfilter, not the basics as I know more or less what stats I’m looking for but more the tiers and how high to set them. Zaodon’s suggestion may help there by taking my current gear as base but I may still run into some issues because of tems I upgraded via crafting and drops that are potentially lower tier but may be better if also upgraded via crafting.
From what I can see you can select idols and their suptype in the lootfilter but not filter on their stats. If I saw this correctly then adding the stats would be a great addition to the lootfilter.
With regards to the lootfilter being code etc, I think Revenchule phrased it nicely that it requires an application of logic and conditionals. If you have an affinity to IT or for example math/databases/analytics etc this is often 2nd nature to you so you’ll have no issue using the lootfilter. But for people who don’t have that the lootfilter may be difficult to use.
I mean, Prolog and other logic programming languages do exist. But I do understand what you’re trying to say, I just think that as “programmers” we often forget that logic is the very basis of the languages we use to code.
I’m also one of those wierdo’s who love even more complication to the loot filter as well. As Zaodon has said, I would not only like a general “all more or equal to 3” but the ability to add up to all 4 of my affixes individually.
As an example:
I look for Engraved Gauntlets with T5 Intelligence, Increased Crit Chance, Hybrid Health & Dodge Rating.
Now, chances of engraveds with all 4 of those affixes dropping, let alone t3+ are slim. I nonetheless have that in my filter as both the “all 4” and the “all 4 min T3” say. No problems doing that at present.
However, there’s far more of a chance that they will drop with 2 of those 4 affixes. At present I have that set as “2 of the 4” with min T3. Still not too many to experiment with.
IF I could put in more than one affix though I could relax that rule slightly to see more drops by saying say “2 of the 4”, Int 1+, Crit 2+, Hyb Heal 3+, Dodge 1+.
It’s a small QOL change, and one that might possibly infuriate people who want it less complicated even more, but for me that would be massive. I could then fine tune my geeky filters to have 1 God Tier, 1 all 4 affixes, 1 crafting potential 3 aff’s etc etc.
One other addition on my personal wish list is a “not” type function. Supposing my pool of wishful thinking affixes is say 6/7 on an item. I would also like to be able to say Item X with 2 or 3 out of those 6 but with no affixes “not” listed. That would not show any items that needed an affix scouring at all. Presently, my "2 out of " and “3 out of” categories do show items that I then need to try and remove an affix from. It’s no hardship doing that and occasionally it works, but this is merely another permutation on my wish list
