Rainbow Loot | Forums

[SuspiciousPixel]

Since this is a contentious topic and people from the community didn't like the sudden backtracking from this:



to



I decided to Target Farm the Gearset people experienced a high percentage of rainbow loot (Providence Defence) to try and estimate the new value.

Please note this is a limited set of data and my World Difficulty was Heroic. All activities represent what players are likely to do during end game. I will continue to add to this spreadsheet when Providence is targeted in missions and strongholds. If a stronghold is available I will do it at Legendary difficulty.

Sheet
https://docs.google.com/spreadsheets...it?usp=sharing

Conclusion
- There is a 70% chance that the first random attribute matches the core for Providence Defence.

[dagrommit]

Good to see I'm not the only one capturing actual stats, though I'm too lazy to break stuff down by brand/item type. These are my results from running 60+ levels of Summit across two days. Hana-U and Overlord were targeted loot.

Total gear items (excluding gear sets): 133
Items where 1st attribute is same color as core: 79 (60%)
Items where 1st attribute is different fromcore: 54 (40%)

Items with rainbow rolls (all three attributes different): 18 (13%)

[edit] screenshots of the first 100 items for reference:

https://imgur.com/a/TH3WCCr
https://imgur.com/a/CDztkPv

Note: I don't believe difficulty makes a difference to the types of attributes that roll, just their quality.

[xcel30]

Might as well thow my own small sample of loot from doing challenging floors on summit yesterday

youtube has been deactivated.
Do you accept the use of cookies to display and allow you to watch the video content?

It alos matches 60% but a really really small sample

[dagrommit]

Ran floors 80 -93 on Summit and picked up another 51 gear pieces. Excluding gear sets::

Items where 1st attribute is same color as core: 27 (53%)
Items where 1st attribute is different from core: 24 (47%)

Items with rainbow rolls (all three attributes different): 11 (21%)

Compared to the other results, consider this a reminder of why sample size matters.

Adding those to the previous samples:

Total gear items (excluding gear sets): 184

Items where 1st attribute is same color as core: 106 (58%)
Items where 1st attribute is different from core: 78 (42%)

Items with rainbow rolls (all three attributes different): 29 (15%)

As always, RNG can and will vary. But based on the data so far, rainbow rolls are far from being a majority.

[edit] added screenshots of the 51

[Brando_0070]

I feel the problem isnt totally down to the amount of rainbow rolls.

I did a few experiments similar to this ages ago where I wrote down everything I farmed for a few play sessions.
During that time I farmed 686 different pieces where 499 were armour, the rest were guns, mods etc.

Out of that 499 there were only 107 full rainbow or 22%
but there was another 80 (16%) that had the 2 minor cores matching but were different from the Major core, ie. a red major (weapon damage) then 2 blues or 2 yellows.
These would also be useless if someone was looking for a full red build.

So even back then when everyone 1st started complaining about the amount of rainbow I was only seeing, at most, 38% of gear that couldnt be useable with a single recalibration.
(I know everyones experience is different but that is what I recorded during my own sessions)

The bigger question in my mind is - how many of those pieces are actually an upgrade or something everyone would be happy to use?

I know quality is a different topic to the spread of the colours, but, I feel, many people just see that they are not getting any upgrades, see a few rainbows and non matching colours and put all the blame there. That is why there are many reports of 99% rainbow, because that is all they remember.

If the stats are not as high as you are looking for (and the exact combination, say just CHC and CHD) it wouldnt matter if they changed it so Providence only ever dropped with all 3 red, 99% would still be trash to many people that had already got pretty good gear or was looking for something really specific.

When its all said and done people just want upgrades, tweaking the amount of rainbow isnt really going to solve that longterm, because as more people get the colour combination that they want they will soon hit a wall on how high they can get the actual stats.

The complaints will then just change to - I cant get upgrades for my stats.

[dagrommit]

Out of that 499 there were only 107 full rainbow or 22%

That matches up with what I saw in the PTS for TU10.1, so TU11 is an improvement. Whether that's a big enough improvement is up for debate.

I know quality is a different topic to the spread of the colours, but, I feel, many people just see that they are not getting any upgrades, see a few rainbows and non matching colours and put all the blame there. That is why there are many reports of 99% rainbow, because that is all they remember.

Confirmation bias is almost certainly a big factor here - i.e. people see a few rainbow rolls and perceive that everything is a rainbow roll. That's why it's important to capture what people actually got. That allows for an informed discussion on the issue rather than one based on guesses.

As for roll quality, the reality is that it will always become more difficult to get an upgrade as you climb up the power curve. For that to change, the distribution of god rolls will need to increase, but that will result in people complaining that they've maxed their builds etc etc.

[Merlinsdragon1]

so its at least 20% guaranteed garbage without looking at the stat value at all and another 50% where its not really useful because 2 foreign stats from core, so 70% guaranteed garbage.
i guess its an improvement from the percieved 99%, so there seems light at the end of the tunnel

[Brando_0070]

That matches up with what I saw in the PTS for TU10.1, so TU11 is an improvement. Whether that's a big enough improvement is up for debate.

As I said, I think you could make it 100% chance of all colours being the same and that might make things go quiet for a while for a subset of the community, whilst simultaneously upsetting another subset.

That's why it's important to capture what people actually got. That allows for an informed discussion on the issue rather than one based on guesses.

I agree, thats why I took the time to write everything down too. But I still feel this dicussion is almost mute as its not the real problem with loot.

As for roll quality, the reality is that it will always become more difficult to get an upgrade as you climb up the power curve

Many people dont care about that, they just want upgrades until everything is as maxed as it can be.
Its what many did on Div1 and its what they want to do now.
Rainbow loot has just been the thing they perceive as standing in their way to achieving that, when in reality the rolls are just not high enough across the board to make that remotely possible for the average player.
Look at all the pics of gear you and the other guy posted, how many of those would satisfy the majority?
Not many I bet.
I know that was only a small sample, but many people complain when they have played for a few hours and not seen anything worth keeping, so those samples are of a relevant amount of stuff collected.

For that to change, the distribution of god rolls will need to increase, but that will result in people complaining that they've maxed their builds etc etc.

Many say that wont be the end for them and they would then happily move onto finding and maxing a new build, and then another.
So noone can say for sure if those complaints would materialise or not.

But im pretty sure they wouldnt be as loud and toxic as the many, many loot complaints are now

[SuspiciousPixel]

The more data I collect the lower the percentage is getting so 60% is looking likely.

[dagrommit]

Completed open world activities and a couple of heroic CP's. Added them to the totals:

Total gear items (excluding gear sets): 235

Items where 1st attribute is same color as core: 131 (56%)
Items where 1st attribute is different from core: 104 (44%)

Items with rainbow rolls (all three attributes different): 37 (15%)

The rainbow percentage has stayed relatively stable (13 - 15%), but the decrease in 1st attributes matching (60 - 56%) is a bit concerning.