Curse

Bobulation · Jun 20, 2008, 06:34 PM // 18:34

Quote:

Originally Posted by Nittle Grasper

http://www.guildwarsguru.com/forum/s...php?t=10225077 <explains that only what time you enter an area matters for what loot you get.

Well, I have been testing this with both of my accounts. My main account assassin and my other account assassin enter an area at the same time and farm the same stuff with different builds. So far I have tested probably about 50 times, and its very shocking. Almost every single time one account gets a gold drop, the other account will get the same gold with the same mods on it. So personally I believe that how you kill the monster doesn't matter, it's when you enter an area.

Damian979 · Jun 20, 2008, 06:57 PM // 18:57

http://wiki.guildwars.com/wiki/Loot_scaling

This explains loot scaling by anet's definition. Unfortunately what it doesn't explain is that the total anti farming effects also include the amount of time played consecutively on one char, the number of consecutive times one char enters and re-enters an area and the speed at which consecutive enemies are killed via non AoE damage per attack (i.e. one duration of sliver armor is considered one single attack).

I find it funny that wiki had a full page on this topic when anti farm codes were introduced, now this little quip of info is all the "official" info I can find.

AshenX · Jun 20, 2008, 07:05 PM // 19:05

Months ago, after reading someone's post about drop rate variation based on mob kill vs single kill rates, I decided to test the theory for myself.

I decided to use the Gates of Kryta farm as my test location and used my elementalist for the test character. My builds centered around sliver armor for individual kills, and fairly generic nuking for the mob kills.

My first series of tests relied on a few hundred runs for each build. The drop rate for the sliver armor build was several times higher than that for the nuking build.

I have since done a larger test with 1500 runs on each build. The sliver armor build consistently received a better drop rate. In the end I received just over 4 times as many drops on average using the sliver armor build than the nuking build.

I have found the same pattern repeated when using a perma SF sin. When I use a dagger build, generating individual kills, my drop rate is much higher than when using Mark + Lava. Of course the run takes forever using a dagger build so the fire build still works out as more efficient over time but the actual number of drops per run is better on the dagger build.

Sorry no screen shots.

Ashen

Maximumraver · Jun 20, 2008, 07:15 PM // 19:15

Quote:

Originally Posted by mazza558

-Does Hard Mode really make a difference?

Yes, a lot, i never play easy mode anymore, the drops... baed.

PANTHOR · Jun 20, 2008, 07:30 PM // 19:30

Quote:

-Is loot scaling affected by the speed of using skills and the speed of kills?
Yes
-Is loot scaling affected by the type of skills? Do you get more drops if you use attack skills? AoE? Skills that kill one at a time
Yes

Be more wrong kthx

Fay Vert · Jun 20, 2008, 07:42 PM // 19:42

Quote:

Originally Posted by Darksun

Actually, you're quite simply WRONG.
That's not loot scaling. It's the basic drop code that's always been there. What Anet said was true, loot scaling (which was implemented after NF came out) is effected by what they said it was effected by.

One of the most misinformed, incorrect and stupid posts ever made. Check your facts, check the research and try again. Its very easy to prove, try reading my post again and then read the original dev update which explained what loot scaling was. As so many have posted so many times before over the last year, it is very easy to prove that it is inacurate. ANet have refused to acknowledge there is a problem despite undeniable proof, though they never state that findings are wrong (how could they) but merely things are as they intended. The dev update is a lie, and anyone who denies the rate of kill factor is ingorant/lazy/fanboy/gullible/newbie.

You can't just post that provable and repeatable facts are wrong purely because you want to believe something. By all means feel free to agree with the decision ANet made with loot scaling, but don't deny facts and don't think for one second that ANet told the truth, because they very provably didn't.

KZaske · Jun 20, 2008, 07:43 PM // 19:43

My answers to your questions are based on my observations and are MY opinion and only my opinion. I am sure there are some that will disagree with me.

Is loot scaling affected by what time of day people farm? Yes. The fewer people on, the better the drops.
Is loot scaling affected by the amount of other players farming the area? Yes
Will secret farming locations give people better drops? Yes
Does loot scaling gradually decrease as people kill more enemies in the explorable/mission? Not sure what you mean. If you mean "gradually decrease" in drops: No.
Is loot scaling affected by the speed of using skills and the speed of kills? Yes and No. Depends on your party make up.
Is loot scaling affected by the type of skills? Yes.
Do you get more drops if you use attack skills? Yes.
Is loot scaling reduced if you barely survive with your build, or when you're at full health all the time? Loot Stealing is in active or inactive, there does not appear to be a scale that loot stealing applies to any zone I have found.
Do higher level enemies really drop better stuff? Better stuff, Yes; as often, No.
Do level 28s beat level 24s in terms of the chance of gold drops? Not from what I have seen.
Does Hard Mode really make a difference? Yes, as for the varity and value of drops.
Do locations which require lots of traveling beforehand give you better drops? Depends on the area. The old anti-farming code is still active so going to a heavly farmed area does reduce the drops.
Why are some locations with enemies of the same level still have different quantities of drops?
Is loot scaling affected by how you use your character? Yes. Part of the old anti-farm code.
Do characters dedicated to farming get worse drops than general PvE characters? Yes, see above.
Will playing PvP or a long PvE mission improve drops if you then return to farming? Yes, see above.
Does farming on different server matters to loot drops? No.
Does favor effect drops in an area that needs favor to go in? No.
Why do drop levels gradually decay after the 3rd of 4th time of entering an area then? That is the old anti-farm code that we were told would be removed. Apparently it wasn’t.

romeus petrus · Jun 20, 2008, 07:53 PM // 19:53

Quote:

Originally Posted by Darksun

Actually, you're quite simply WRONG.
That's not loot scaling. It's the basic drop code that's always been there. What Anet said was true, loot scaling (which was implemented after NF came out) is effected by what they said it was effected by.

I just loled.

Didn't think anyone who frequented these forums would ever deny that large groups of monsters dying at the same time = less drops.

kvndoom · Jun 20, 2008, 09:08 PM // 21:08

What I believe but can't prove is that anti-farming code was weakened, but not totally removed. Are we supposed to believe everything Anet says? Did Anet not say it was impossible to stop GoLE cast-faking? Did Anet not say it was impossible to separate PVE and PVP skill function? But realistically, I don't see much effect of farming any place repeatedly.

What I know and can prove (and so can anyone):

Hard mode drops are basically the same as "normal mode" before there was such a thing as hard mode.

The time elapsed between enemy deaths affects drops TREMENDOUSLY. On several UW Chaos Plains runs, I used different methods to kill Specters.

1) Fastest kill- 3 Lava Fonts in succession. All adjacent enemies die simultaneously. Drops are almost non-existent. Non-lootscaled items (ecto) drop normally. Can kill 18 enemies at the same time and MAYBE have 2 drops.

2) Slower kill- 2 Lava Fonts, then wand each enemy individually so they die one at a time. Only a couple seconds between deaths though, since you just target the next foe and hit them with staff to finish them off. Total drops increase dramatically. Get more gold coins and more blues, whites, golds and purples. I'd say off the top of my head a 5x higher chance of getting something to drop. The drops may be crap, but even crap adds up at the merchant.

3) Slowest kill- use conjure flame + MoR, hit with staff or fiery scythe. Takes a lot longer, but drops are a slight bit better than method 2. The time spent versus the items dropped isn't really worth it.

There really is no mystery. These results can be very easily reproduced by anyone. If all the enemies die at once, you will get JACK S**T for drops, especially in Normal Mode. I seldom AOE farm in hard mode, so I can't give conclusive data there.

As for Loot Scaling, it's pretty simple. You more or less get the same drops solo as you would in a full party. Nothing too hard to prove, just take a full party or H/H on one run and do it again solo and compare your share of drops.

Chthon · Jun 20, 2008, 09:25 PM // 21:25

Quote:

Originally Posted by BlueNovember

Specualation and rumour!
http://www.guildwarsguru.com/forum/s...php?t=10225077
^^

It may seem that time taken to kill mobs has an effect, but the above thread conclusively shows that it does not.

Please re-read that thread more closely. It conclusively shows that the effect is very small, but it definitely exists.

Quote:

Originally Posted by Bobulation

Well, I have been testing this with both of my accounts. My main account assassin and my other account assassin enter an area at the same time and farm the same stuff with different builds. So far I have tested probably about 50 times, and its very shocking. Almost every single time one account gets a gold drop, the other account will get the same gold with the same mods on it. So personally I believe that how you kill the monster doesn't matter, it's when you enter an area.

If you know even a modest bit about programming, it's not shocking at all. Let me explain what's going on: Computers cannot generate truly random numbers. The best they can do is to produce a (long, but repeating) sequence of numbers with an even distribution across a range where the terms have a very inelegant relation to each other, and then pick a different starting point in the sequence each time it is used so that users don't get deja vu. You pick the starting point by giving your pseudo-random number generator an input. Standard industry practice is to use the very small digits from the system time as the input, since it creates the illusion of randomness at a very fine granularity, and it's virtually impossible to the user to control their input that precisely. GW is taking the system time when you enter a zone as the "seed" input to initialize its pseudo-random number generator.

(For the more metaphor inclined: GW is not rolling the dice every time you kill a monster. Rather it's rolling a hundred dice once when you enter the zone, then looking at the first die when you kill the first monster, and looking at the second die when you kill the second monster, etc. It's also using the the same set of 100 dice for everyone who enters the zone at the same time. So, if you and I synch join, it will roll just one set of 100 dice, then look at the first die when you kill your first monster and when I kill my fist monster, and look at the second die when you kill your second monster, and when I kill my second monster, etc.)

So where does this small difference that the synch joining experiments found come from? Tmakinen proposed the best explanation I've seen, and backed it up with some experiments.

First, let's imagine a non-loot-scaling world where every critter has a 50% chance to drop an item. How would you program that? Well, a simple way would be to take the next pseudo-random number off your series, range 1 to 100, and drop an item if it's over 50. Or, to put it another way, "if (roll > threshold), then drop item."

Loot scale appears to make two modifications to this simple system:

First, the threshold that your pseudo-random number has to exceed to get an item to drop varies with party size. Going by some old comments by Gaile, it scales from zero for a full party (every monster drops something) up to the top of the range on the pseudo-random numbers minus two divided by the max party size (you get twice as many drops solo as you'd get in a full party, after loot division among party members).

Second, before your pseudo-random number gets checked to see if it clears the threshold, a small value might get added to or subtracted from it. This value seems to be defined by reference to the rate at which you've been making gold (as measured by the merc value of the things that drop for you). If you're making money too fast, something is subtracted off your pseudo-random numbers to lower your odds of getting more drops in the immediate future. If you're making money slower than the limit a-net set, pseudo-random numbers might get a small boost so that you're more likely to get drops in the immediate future.

That leaves us with, "If ((roll +/- adjustment) > threshold), then drop item," where adjustment is related to how far above or below a-net's "speed limit" your drops have been recently, and threshold depends on how many people are in the party.

Fay Vert · Jun 20, 2008, 09:45 PM // 21:45

Quote:

Originally Posted by Chthon

Please re-read that thread more closely. It conclusively shows that the effect is very small, but it definitely exists.

Try full party trapping or EoE bombing and you will see its not "very small"
Try solo farming one kill at a time and see a drop from 95% of the mobs, then say its very small.

The OP first screen shot in that thread proves it: Solo run, 70 mobs yet 50-60 drops? Loot scaling is not as ANet says

Chuk Charcoales · Jun 20, 2008, 09:51 PM // 21:51

Quote:

Originally Posted by Fay Vert

Try full party trapping or EoE bombing and you will see its not "very small"
Try solo farming one kill at a time and see a drop from 95% of the mobs, then say its very small.

QFT
I remember back when you could farm with splinter/barrager; always got way lower drops than I did with my sv necro even though I killed faster with my barrager.

Crom The Pale · Jun 20, 2008, 09:52 PM // 21:52

It is hard to get any exactly acurate data on Loot Scalling simply because you have to factor in the Random chance of a drop.

Enter a zone solo and kill just one monster and leave, repeat. You might get a drop every time or just 1/100. Try that same test with a full party and see if the results change.

Loot scaling alters the probability of getting a drop as well as the quality of the drop, but its altering a random variable so the change may not be measurable in some instances and very noticable in others.

Ate of DK · Jun 20, 2008, 10:26 PM // 22:26

What do we know about Loot Scaling?
Loot Scaling does not affect full parties. It reduces the number of drops when less people are in the party. As soon as you play with less than a full party you won't see all items drop.

There is a difference between anti-farming effects and Loot Scaling!
Anti-farming effects are the so called AoE-bug or RoK.
The AoE-bug: makes sure that you get less drops when you kill multiple enemies at the same time.
RoK (Rate of Kill): Some people think that it's not only the AoE damage but only the number of enemies that you kill in a short period.

To know if Loot Scaling is working correctly, you have to know if every opponent in the game drops an item when you're in a full party.
We tested this on a different forum some time ago and a few people added information. We killed every single enemy in an area that didn't take much time to complete. We saw that every enemy dropped an item when we had a full party. Then, these few people, completed that same area ~50 times with hero and henchman. Average numbers after all these runs showed that we got 1/8 of the drops in that area. (It was an 8-man area)

Does that mean Loot Scaling works then?

A: We discovered that Anet has a few Loot Tables for several area's.

Example:
People were complaining that they didn't get 1/8 of the drops when they farmed the Snowman Dungeon. They expected to get 1/8 of the drops but a smart player will immediatly notice that the Snowmans don't drop gold coins! Anet changed the Loot Tables of these enemies. Probably because the area was used for quick Derlver points and the low level of the huge mobs.

There are more area's that have different effects. People who farm the first Stygian Veil quest can also tell you that the mobs they farm won't drop gold coins.

Thus it's very hard to tell what you can expect in some area's. But it's safe to say that most explorable area's that you can vanquish will earn you 1/8 of the drops. (And 1/4 if it's an 4-man area)

B: Bosses
Bosses can drop up to 4 items. Which messes up statistics. It's possible to get more drops in a full party than there are enemies on the map.

C: Gold coins
How to count the gold coins? Statistics gave us the impression that a single gold coins drop can be considered to be 1 drop. Meaning that in a full party: 8x gold coins drop for the party = 1 drop for a single player. (You get 1/8 of each gold coin stack in a full party. --> 8/8 = 1)

Conclusion:
Nobody can tell you if Loot Scaling is working correctly. The only ones that know if Loot Scaling has the effect it's used for is Anet. Anet won't reveal any information on this, they haven't done that since they implemented it and gave the additional dev. update.

I'm still convinced that Loot Scaling is working like intended. My gaming experience became more fun. Since they implemented Loot Scaling I didn't feel the need to farm anymore to buy all equipment. Low prices is more fun for everybody.

manitoba1073 · Jun 20, 2008, 10:44 PM // 22:44

If you want to know how LS works here is the thread to see.
http://www.guildwarsguru.com/forum/s...lo ot+scaling

That is the plain and simple way loot scaling works for soloing.

Nothing else matters. Most dont want to aggre but its there.

Cab Tastic · Jun 20, 2008, 10:47 PM // 22:47

Quote:

Originally Posted by Darksun

Actually, you're quite simply WRONG.
That's not loot scaling. It's the basic drop code that's always been there. What Anet said was true, loot scaling (which was implemented after NF came out) is effected by what they said it was effected by.

No you are quite simply WRONG.
Before LS there was never an issue with rate of kill (turai farm anyone?).

I was discussing this with a guildie a fwe days ago. If you observe your rate of kills you will see that if you are too efficient your drops will decrease.
This is basically a catch all by anet. no matter how many kills you make/hour you loot not change dramatically.

Anet have lied about how LS works. I think it was a mistake that plagues this game to this day.

Fay Vert · Jun 20, 2008, 11:12 PM // 23:12

Quote:

Originally Posted by Ate of DK

What do we know about Loot Scaling?
Loot Scaling does not affect full parties. It reduces the number of drops when less people are in the party. As soon as you play with less than a full party you won't see all items drop.

Instant fail

So how do explain that I can farm a certain way solo and get a drop off of nearly every mob, and yet I can enter Urgoz with a full party of 12 and EoE bomb mobs and get very few drops for the entire party?

eeks · Jun 20, 2008, 11:20 PM // 23:20

Hm. Considering the success of the recent SF ecto runs or even smite runs, could it be the run to Chaos Planes or to the smites (in which nothing is killed), that prevents proper loot scaling from occurring? That is to say, the type of loot scaling that's mentioned here (less gold drops over a period of time), unless that's anti-farming code.

I dunno. I pulled that from my behind.

Weirdly enough, I get more drops if I complain in-game about getting less drops.

romeus petrus · Jun 20, 2008, 11:30 PM // 23:30

Quote:

Originally Posted by Fay Vert

Instant fail

So how do explain that I can farm a certain way solo and get a drop off of nearly every mob, and yet I can enter Urgoz with a full party of 12 and EoE bomb mobs and get very few drops for the entire party?

Instant win for Fay just coz Urgoz was mentioned.

My guild has had the entire room killed instantly in an EoE bomb with a full 12 man team, and got a grand total of 0 drops. It is clearly not the size of the party, it is the rate of killing. Denying that is like saying oil prices didn't go up over the past year.

manitoba1073 · Jun 21, 2008, 12:16 AM // 00:16

Quote:

Originally Posted by romeus petrus

Instant win for Fay just coz Urgoz was mentioned.

My guild has had the entire room killed instantly in an EoE bomb with a full 12 man team, and got a grand total of 0 drops. It is clearly not the size of the party, it is the rate of killing. Denying that is like saying oil prices didn't go up over the past year.

Afraid not. It actually has to do with the Timing of kills. Not the rate or how fast you kill them.