PSA: Instanced Loot Drops > Non-Instanced loot drops

Grim Dawn Gameplay Discussion
1

HO7Zfhd.jpg1920×1080 717 KB

^What this thread is about, whether you host a game with that option on or off.
Having it selected (red, and it’s selected by default just fyi) = instanced loot. Turning that option off = non-instanced loot.

Ran SR 75-76 with the same 3 people.

2 runs on instanced, 2 runs on non-instanced:

first non instanced run 3 players: 32 legendaries total
second non instanced run 3 players: 28 legendaries total

first run instanced 3 players: 43 legendaries total
second run instanced 3 players: 47 legendaries total

The disparity would be even bigger with 4 people.

Alright I found what seems to be actual proof since I think people won’t believe me unless I go to this length:

yljjFFe.png886×286 11.9 KB

So as you can see in single player a nemesis chest will spawn 14-15 items in ultimate, which I believe just cycles through loot 1-6.

But for 2 player it is just 18-19 items. So you’re only getting about 4 extra drops cycling through loots1-6 in 2 player. This includes things like components/potions. What I mean is legendaries can’t drop from every single loot1,2,3,4,5 and 6. Only a couple of those usually have a chance to drop a legendary.

I checked the file for what seemed to be the loot chests for shattered realms and they seemed to be similar. Except what I think was a high tier chest in SR had 7+(numberOfPlayer*3).

It’s not that simple either though:

ngKGEIA.png1576×739 39.5 KB

And for the sake of more information: As you can see loot1 has a 100% chance to drop a legendary when its selected by the chest drop system.
Loot 6 has a 2/3 chance to drop a legendary, but a 1/3 chance to drop an epic.

The way it works is that each loot1-6 has a different weight depending on which item is being selected to drop. You’ll just have to take my word for it since I’m not going to take a more screenshots, but you see the weight of 1000.0 in indexes 1 and 2? Well I looked at loot2-6 and they’re all 0. That’s why this tier of SR chest will always drop AT LEAST 2 legendaries. Then for the third item from the chest there is a weight of 200.0 which is weighed again loot2’s options which have a weight of 700.0. So your third item out of that tier of chest only has a 2/9 chance of being a legendary, and loot2’s options are rares and MI’s.

And so forth and so forth.

So now I can say definitively that this is not what happens in instanced loot. In instanced loot the chest drops as if each player is in singleplayer, and that’s why more legendaries drop overall.

1 Like
2

Seriously? Your n == 2. I am not saying you are wrong, but your sample size is too small to say anything.

3

I also looked at the loot tables a long while back and it seems to function as I suspected. (unless I hallucinated)
The disparity is too great for you to make this statement dude.

If the difference between the 2 was within ~5 you’d have a point. But with such a huge gap I am certain my premise is correct. 28 legendaries /3 = 9.3 per player.
So to make that 28 legendary run non-instanced not prove my point, you’d have to say that not only did each player get 3.7 legendaries less than I’ve EVER seen drop for any player in 75-76 instanced… but that it happened to all 3 of us at the same time.

I have done EXTENSIVE testing on how many legendaries drop per player solo in 75-76. The average is about 15.3 per player

But the low amount that dropped in non-instanced loot with the same amount of players, the same exact players was significantly lower. If I did 4 players I’d reckon you’d see ~60 legendaries per run instanced and mid to high 30’s non-instanced.

Again, from looking at drop tables:
It seems that non-instanced additional players is like just an additional couple of chances for certain drops on the tables.

Whereas instanced loot is a literal loot multiplier.

Furthermore I had noted this occurring MANY times in the past, just didn’t record it. If anyone from the forum would like to help me test with 4 players to get numbers (the disparity should be even bigger between the two modes with 4 players) hit me up. But the disparity alone should be convincing enough.

The question now I guess is how do people feel about this? I think it’s fine since it prevents players from overly exploiting using instanced-loot-off + splitting up.

4

Please understand RNG and sample size. Even your “extensive” testing is still far too small a size to indicate anything.

If you’re so certain, surely the game code should tell anyone what they need to know, especially somone that is able to make mods, like yourself, Why can you not simply check how the game works in the code and get a direct answer. You say you haven’t even checked the drop tables since “A long time ago”

Either check the code and report facts or don’t expect any belief from a single set of results

5

Regardless: again, I edited my post for clarity above.

I have farmed legendaries for groups of 3-4 in instanced runs over 100 times. I have never seen drop rates anywhere close to 28 for 3 people, let alone 32.

What you’re not hearing is that the drop rates in SR are super tight. 75-76 instanced = almost always 14-16 legendaries. On super rare occassions 13 or 17. I think I’ve seen 12 once.

The chance that they dropped 3.7 legendaries less than the aboslute minimum I’ve ever seen drop from SR76… and that this happened for every single player in a 75-76 run (yes we got the speed bonus chests every time) is just not possible. I don’t need to check the game code when the difference is so large. It is WELL beyond the minimum amount I’ve seen in instanced loot.

I will do more tests in the future, hopefully with 4 players to show the disparity gets bigger.

Here is an image of 10 runs I recorded solo. I can only tell you from doing well over a 100 runs in multiplayer than when I have instanced loot on, these numbers are reflected for everyone. EACH player is achieving the average you see in the image. Yet whenever I turn instanced loot off, I see a huge dip in the amount of legendaries that drop.

I have since edited more information screenshotted from the game’s files into my opening post to prove my premise.

6

No… I have a problem with anyone that repeatedly wants builds nerfed based off crucible performance, or claims xyz with just a single sample and insists a single sample is adequate.

Just because I disagree with a lot of your recent comments is nothing against you personally.

Off topic slightly… I don’t understand why your recent threads asking for changes are all in the public and not testing forums, I don’t see other tersters doing this, or are you trying to get more support for your suggestions from the public forums than the testing ones???

8

Call BS as much as you like… You totally miss the Crucible concept and comments. It’s not about not caring abouit the Crucible it’s about the fact that you seem to think HOW a build performs in Crucible (ignoring campaign) should be enough of a reason to call for nerfs to a build which will effect the Campaign build. I’m not the only tester that has said this to you repeatedly about not basing your call for nerfs purely on Crucible performance.

Still that’s going off topic enough.

9

You take an extremely small portion of my testing work, that I do in-line with what the dev has told us, and think this is a good reason to hash this old nothing up again, publicly to throw shade at me? Be better than this.

It is off-topic. You skirt around how I pointed out your not so subtle character attack. I did address the Crucible comment. The only issue with my previous ponting-outs of some Crucible builds is sometimes I didn’t take into consideration how MI-reliant they were and how super RNG MI drops shouldn’t be balanced around. I’ve since adjusted and don’t bring up super crazy MI builds at all.

I’ve talked to the other testers about my approach on crucible build power criticism-making. I’ve taken their feedback into my approach. You’re the only one left who simply clings to the dislike that I factor them in AT ALL. And I addressed that too:
“…Zantai cares about crucible build performance to some degree, which is the REASON I do.”

Your point makes no sense as well. You could nerf a Crucible build from 6 minutes to 9 minutes (NOT what I am suggesting btw) and it’d still utterly crush campaign. If you’re saying it might hurt how it performs in SR? When it comes to buff/nerf suggestions I actually try to keep in consideration how a build does in Crucible AND SR, with minimal build tweaking to adjust for each game-mode.

“I am not the only tester…” Guess what, you’re also not the dev, and some testers agree with me. But who cares about whether a tester agrees or disagree? You’re entitled to your view of game design, I’m entitled to mine, and both are somewhat irrelevant to me since I actually just try to figure out what the dev is going for and support that.

Spellscourge had issues in Crucible and SR but it easily dismantles all campaign content. Yet I still think it should get buffed for how much worse it performs compared to other sets in both modes of challenge content. This runs counter to this idea that I’m nerf-crazy. The majority of my suggestions are buffs to make campaign building around items more fun. The best thing about this little tiff if you constantly make brief statements to put me on the defensive, because it’s so much easier to poke holes instead of point out solutions.

Point out a crucible build I suggested should be nerfed that is now bad in campaign.

I’ve said it before: If I don’t mention how a suggested nerf will affect campaign performance it’s not because I haven’t thought about it, it’s because I’m of the opinion that the suggested nerf won’t make any meaningful impact on the build’s ability to steamroll through campaign if fully built.

10

Chill guys.

11

So… what does instanced and non instanced mean here exactly?

12

Multiplayer loot. Instanced = Each person gets, and sees, their own loot and does not see what drops for others until they toss it out.

Non-instanced = Free-for-all, quickest or nearest to grab it gets it. Everyone gets, and sees, the same drops. Also known as ninja-looting. Never been a fan of it for obvious reasons.

13

Ah, so if OP’s findings are correct, this should be a bug right?

14

I don’t believe he’s stating this as a bug report, but more as a suggestion to always use instanced loot as an option so that you get more loot per run on a total basis (and as an individual player where you get your fair share).

This really seems like common sense to me though and reflects what I saw happen when using these options playing through the game with my son. Instanced loot is really the only way to play as everything in the game is effectively doubled regarding chance to get loot (and it should be!). Although, if I remember correctly, there was a bug that was fixed in the Iron drops that greatly inflated them with more players than what was originally intended.

15

Well to be clear, there is still merit in using non-instanced loot while splitting up if you’re playing online with strangers and don’t want to share loot. Then you can all each do a challenge dungeon each and enjoy higher than solo loot drops.

But yes, basically if you’re playing with anyone you trust and want to exchange loot with freely to each character’s needs, then instanced loot is always the way to go.

I doubt this is a bug. I could imagine that it’s an artifact of instanced loot being added later in development. Alternatively it could be made that way so that people can’t overly exploit the non-instanced loot setting by splitting up.
Maybe it just worked out that way.

This is simply a Public Service Announcement in the hopes of making sure our members aren’t suckered into using non-instanced loot when playing with friends. Because psychologically they might “see” more drops and think it’s better when it’s actually worse.