Curse

My Green Storage · Apr 28, 2007, 02:08 AM // 02:08

E/A (and the occasional A/E) HARD MODE -
Solo farm of:
Chor
Mindclouder
TomTom Spiriteater (Factions Rit Boss)

Chor - Over the span of 2 days, close to 80 runs.
I got a gold, black dye and I think another gold within the first 5 runs on day 2 (Gold bow had 20/20 string)
Then it just teetered out to nothing. I could go 15 runs and nothing would drop. Not even white items.
For a few runs it seemed like a scroll of adventurer's insight would drop often.

Mindclouder - Close to 70 runs
I honestly don't remember him dropping anything except gold or a scroll every 10-15 runs.
I *did* get a number of gold heket armor drops off of the hekets. I'd say about 3-5 gold runes in the span of about 20 runs .. which of course is fairly good.
TomTom Spiriteater - Close to 100 runs
Absolutely nothing.

The sad things was, I didn't give a honk about their greens .. I wanted their damn Elite Tomes.
---Merched 5 SpiritEater offhands, 3 Chor's Axe and 1 Mindclouder---

And Tamirxd. They're saying if you're taking on groups of 8+ it'll affect the drops. Not sure how the solo necro build works (Spoil Victor? SS?), but you're not taking on groups of 15 smites at a time like an Sandstorm Ele would (and could).

Funny you bring this up, I always thought the LONGER it took you to kill a boss, the better a green drop would be.
Seemed like that has happened a few times I noticed BEFORE hard mode, not sure if the same holds true.

But the time it takes might have an affect too (I'm sure it was mentioned).
Kill a boss in 15 seconds. Low Drop Rate
Kill a boss when it takes you longer then 60 seconds. Better Chance.

Especially noticeable when dealing with heroes, flagging a few away when it's time for the boss.

As for the UW, I'll test it also as an ele. Usually once you reach the area after the 2 smite groups you first run into, I try to get all 4-5 groups in 1 pop ... only to NOW see 3 whole drops on the group after all 15-20 are dead ;/
So it looks like I'll spend some quality time with each group seperately

**Did 1 run and will say doing each group seperate (3-6 smites per group) did get me more money then I have noticed then when I did 8-18 smites.
I take out 2k for each run, spent 1k, so for me to get back 955 gold is very good now a days.
I know times doing 2+ groups at a time, I'd come out of there with 300 gold.
So maybe it does hold true taking on less = more
http://hidebehind.com/46B436

kill myself to live · Apr 28, 2007, 02:55 AM // 02:55

i have noticed that the smaller groups yield more drops and if you kill them slower too. since hardmode came out i have been getting 1-5 ectos per solo run in uw on normal mode too. and just last night i got a req 9 gold max fellblade and req 12 gold max spatha in one group of 4 smites in normal. hard mode has given me lots of golds in other places too with small groups killing one at a time even after repeating the map several times.

Lurid · Apr 28, 2007, 02:59 AM // 02:59

Yep, I can also verify this. I always get better loot from the stragglers that are left over when the main mob is killed. Takes longer to kill them, but I generally get more golds from 'em.

brian78wa · Apr 28, 2007, 03:21 AM // 03:21

Quote:

Originally Posted by tamirxd

this is my ecto rate[solo necro],this is the normal mode stats,killing aataxe and smites.
First run 4 ectos.
Second run 3 ectos.
Third run 3 ectos.
In Hard Mode:
First run 3 ectos[After i killed something like 4-6 aataxe].
If the drop sucks for you guys make a mission or 2 it will improve your drop rate[I did 2 hour run with no ecto, then did 2 mission and as you can see above its MUCH better].

5 runs for me 0 ectos. And I either break even or lose money for going to UW. This is after doing missions.

SotiCoto · Apr 28, 2007, 11:36 AM // 11:36

Lady Luck is a harsh mistress.... which is ok if you're a masochist and into all that kinky stuff....

But I'll bet, odd as it may seem.... that all of this particular speculation into matters influencing drop-rate is at best inaccurate.

You see.... there has to be a probability regarding drop-rate that something will drop... and a separate probability for WHAT will drop if something does...
Now... I suppose it is possible that in any given group of enemies killed simultaneously.... that the system works out an average essentially for the entire group as one as opposed to individual enemies.... thus taking which enemies have been killed in the last second or so into account to adhere to the mean.... AND... on that same basis, to enemies killed separately, several seconds apart and whatnot..... it might reboot those settings and start its average all over again.

But even so.... irrespective of kill order, kill time or the size of the group.... it most likely evens out to average over time... In the short-term you might suspect you're seeing some sort of difference, but that can most likely be attributed to random fluctuation... i.e. noise. Anomalous results. Give it long enough and I strongly suspect that it will average out...

Why?
Because I suspect the A-Net folks are probably far too lazy to implement something so complex as you suspect...

Let us not forget.... The human mind is specially coordinated to find pattern in things. It what gives us our particular brand of sapience. It is however far from perfect, and tends to find patterns where in fact there are none, simply due to the ability of randomness to occasionally generate consistancy.

[Excuse me if any of this isn't totally clear.... I'm kinda drunk.]

FoxBat · Apr 28, 2007, 01:50 PM // 13:50

Quote:

Originally Posted by SotiCoto

Let us not forget.... The human mind is specially coordinated to find pattern in things. It what gives us our particular brand of sapience. It is however far from perfect, and tends to find patterns where in fact there are none, simply due to the ability of randomness to occasionally generate consistancy.

This is what testing AKA "science" is for. Hearing about "My experience" one way or the other isn't terribly convincing. But make 5 or 10 runs doing it one way, and another 5 or 10 runs the other, with the same character, at the same spot. Screenshot all your drops, count them afterward, and create an arithmetic mean. Tell me if you don't notice a difference.

Its true the human mind is good at making up patterns that aren't there, but that doesn't mean patterns don't exist, especially in something as deterministic and mathematical as computer software. We have the scientific method as a way of distinguishing between the two, and can apply it here on a small scale by creating a control group, running multiple times, and having other people do the same thing and report their results. It's alot better at producing a good answer than speculating back and forth.

Gaile *did* say in that infamous dev update thread that loot scaling is done by a "complex algorithm" and 1/4th drops in an 8-man was just an estimate. This is completely in line with that.