As everyone else I’ve been picking up a ton of unique items. The “system” I’ve adopted is - keep one of each unique and if I pick up a better one replace it. Most of the unique items I have don’t have any legendary potential but I keep them just so that I have something if I suddenly stumble upon a build that uses it.
One thing that I noticed lately is that items that I’m picking up that have legendary potential are far worse than the ones I already had without the potential. E.g. I have an item with 0 LP but decent rolls. I pick up another item of the same type that has 2 LP and start comparing it to the one I have it and it turns out that pretty much all affixes (and/or implicits) are worse than what I already have.
In my view that should never be the case and if something has legendary potential then that should mean it’s truly good. One fix would be to bind legendary potential to the quality of the item that drops. E.g. anything below a hidden 50% quality rating would always drop with 0LP and very average affixes/implicits. Drops between 50% and 70% would drop with 1LP and stats that are better but not as good as 2LP, 3LP and 4LP drops. And so on. The final LP4 tier could be the top 5% and have THE best affix rolls possible.
As it stants it’s RNG on top of RNG on top of RNG and you never get anything that’s actually good. You either keep it because it’s OK, or just sell it because it’s useless.
Can confirm the same experience re LP dropped Uniques.
Higher LP “tend” to have poorer rolls than lower LP items… Its not always true obviously (RNG and all) but after collecting way more than one or two of each Unique item (yes, I have a hoarding problem) I definitely notice this trend overall.
I regularly keep multiple items (3 or 4, sometimes more) and do the exact same comparison and rotate out the poorer rolls - I can have 20 of the same unique ring and typically the higher LP ones will have poorer rolls - way more than average so that is hinting to me (with my poor maths brain) that its intentional.
I dont recall Mike or anyone (Tunk, Dammit) confirming that there is some code in place to add weighting like this to drops - maybe someone else can confirm if its intentional?
For me its a definite “feels bad” kind of thing to get a LP3 drop only to realise that its rolled with worse stats than an LP0/1 copy. The chances of a higher LP dropping are already very rare in some cases that one could assume (right or wrongly) that the overall item should be generally better.
Beyond the feels bad, I am not sure if this is a MUST have better stats or just on AVERAGE SHOULD have better stats kind of thing.
It almost feels like there’s a number of points that can be allocated to affixes AND legendary potential and if you end up with higher LP you have less left to allocate to affixes (or implicits). The system just pure sucks because once you have one 0LP item with good rolls you’re very unlikely to ever get a better one with 1LP or more.
Perhaps… No idea on the math behind the rolls but I agree it seems like there is something beyond pure RNG happening here. It isnt great on average but I have experienced high LP with great rolls so its not impossible. For me the issue is also the rarity - its not like LP3/4s drop like candy in the first place. They are REALLY rare especially for Uniques that are actually useful in late game or build defining.
Dont like to ping Mike directly but hes likely the best person to respond to this to confirm the math/intention.
1 or 2 data points? Yeah, that’s statistically significant. This is the “fracturing chance is way higher than the % shown” tin foil hattery back in the day.
I have hundreds of uniques so its most definitely not cycling one or two that I am basing my perception on - its something I have noticed very often while collecting and throwing away the worst of multiple of the same items and its enough for me to wonder if there is something beyond RNG involved.
I did a stash clean out before 0.9 dropped and used this exact criteria to confirm what I should throw out - and this must have been for at least 500 uniques, sometimes 20 or 30 of the same, especially small items like rings and I noticed this low rolls on higher LPs generally - it made it hard to quickly chose what to throw away so I remember it.
But sure, its a pretty small sample - thats why Mike’s likely the best person to refute or clarify this once and for all.
As you and the OP have both said, the LP item isn’t being compared to random uniques of the same item. They are being compared to the best version of that unique you have found to date.
I haven’t noticed this as a phenomenon but I also have hundreds of uniques. I’ll take a look and compare also. Like others have said, it may mean nothing. But either way, I tend to agree - I think the rolls for items with LP should generally start at a higher range altogether so that the minimum rolls are better along with the maximums. Ideally they’d be even higher than that, but people would then have no reason to use the ones without LP I suppose. And the ones with LP are pretty rare.
Ideally, being able to add LP to the one you like would be the way to go, and there would just be a key dungeon for that. That might also be a good solution.
Yes, I had the same experience, seems we have some sort of bias, but I really try to be objective with this.
On average, I get better rolls on 0LP items than in 1+LP items. Sometimes I had real luck and got a high rolled 2-3LP one, but is the exception, not the norm.
Is like if an item with LP gets into a second roll of the RNG stats and the lower values of the two rolls are applied. Much like the feeling I have with Idols, is quite hard to get good rolls in them. I wouldn’t be surprised at all, after all the design team has some sort of terror about OP items. Making items with 100 or 110 level for LP maybe is not enough.
The best rolled affixes on a 0LP Arboreal Circuit you found will be similar to the best rolled affixes on a 2LP Arboreal Circuit… assuming you found the same number of them (which essentially never happens happens) and that number is large enough to smooth out impact of outliers.
Doesnt change the feeling of getting a low-rolled 2LP circuit though.
As for the suggestion, gating LP behind quality levels would make them even rarer than they are now and people would riot. Me no likey
I’m way more of a hoarder. I keep the best 2 (period) and then, after that, the best 2 with LP (most LP). Sometimes that results in all 4 having LP, sometimes not.
As I said multiple times Le is a modern day slot machine and every item you pick up is a gamble. Rng is rng is rng deal with it or get to the root of the problem and talk about the bad itemisation in LE in the full spectrum.
BTW after checking I have LP items with higer roll then non LP items.
In the clean out I mentioned above, I did a comparison to tens or more of almost every unique item that I have hoarded over a long time (cannot recall when LP was added exactly) - i.e. it was not individuals compared against the latest drop. It was just a bunch of items I kept because they all had LP compared against each other at a moment in time. I.e. I opened a randomn bag of items and compared them to see what to keep.
Yeah - I am the first to say similar. I’d like, however, for peace of mind, to confirm that this is the case and there isnt some link between LP level and overall roll quality on items. I know it IS possible to get good rolls on high LP. I have a few LP3/4 items that have almost perfect rolls, but I am still not sure if there is any weighting.
TBH, if Mike replied and said its all 100% RNG or there is a specific weighting involved with LP and LP rarity items, then I would be happy either way.
Again, perhaps its just my perception on this particular issue. I will be the first to accept it if the devs can confirm. AS to the general itemisation problems - thats a bigger topic for a different thread
Yeaah there needs to be a different thread for a look at the big picture ^^. You might get a faster answer if you ask in discord then waiting for it here. But I’m pretty sure they will run a test and say everything is fine .
Secretly hoping Mike will do this like he did with the old fracturing issue on crafting. Honestly, if I had to put my life on the line I would say its just RNG, but seeing as its not life threatening, there is enough doubt for me to question the math.
In theory I’d think something is fishy because almost all my LP uniques are better then the non LP uniques I have and me beeing lucky simply don’t happens… at all! Then again I’m pretty sure EHG is using the same rng method everywhere so it’ll be a save bet all is fine even when I’m confused by beeing lucky ^^.
