Curse

arcanemacabre · Jul 23, 2007, 01:53 AM // 01:53

Quote:

Originally Posted by Loviatar

a dev stated that the scale was a sliding curve and that over the long term average someone solo would wind up with about twice the loot they would have gotten if they used a full party.

Quote or it didn't happen.

Loviatar · Jul 23, 2007, 02:24 AM // 02:24

Quote:

Originally Posted by arcanemacabre

Quote or it didn't happen.

it was in the dev forum and Gaile Gray can confirm it.

the search is not working for me at the moment.

also i have a very long history of not being stupid enough to state something provably false.

it happened

arcanemacabre · Jul 23, 2007, 02:32 AM // 02:32

Quote:

Originally Posted by Loviatar

it was in the dev forum and Gaile Gray can confirm it.

the search is not working for me at the moment.

also i have a very long history of not being stupid enough to state something provably false.

it happened

I am arguing the statement of things as facts that are not provable. As soon as something is provable, then I will adhere to it. Simple as that. Your track record means nothing to me, Loviatar. No offense. I will not take someone's word just for their word alone, which is the premise of this entire thread to begin with.

I will research as well, I'm not sitting on it passively awaiting proof, either. As soon as I see something confirming the average, statistical, or otherwise actual RNG loot drop code to provide real percentages, it's all malarkey to me.

EDIT:
I'd like to add my own theory to the equation. I think the percentage chance of drops (average) may be affected by other factors than just loot scaling. As others have mentioned, there may be in place something that alters the percentage chance of drops by how fast you kill multiple mobs simultaneously when soloing (or also scaled with party size; perhaps with a 100% chance drop rate with a party of 8, down to a 50% drop rate when solo in an 8-player zone - completely made up variables to show my theory).

There may also be increased or decreased drops depending on the average level of monsters in a zone, or how that zone is classified as "Beginner" (Istan, Ascalon, Shing Jea), "Advanced" (Kourna, Kaineng, Kryta/Desert), "Difficult" (Desolation, Luxon/Kurzick, Shiverpeaks/Ring of Fire), and "High-Level" (Torment/DoA, Urgoz/Deep, Fow/UW). Theoretically, each area could have its own drop rate percentage (even different for each type of drop) which may be based on that area's difficulty, or even independently controlled by Anet in case an area becomes too popular for bots/players. This might give the illusion many have described as an area being "overfarmed", and suddenly having less drops than it used to because others are farming it dry. We know this isn't true, but if Anet can control drop rates of each area, it would make much more sense. Plus, I don't see why they wouldn't be able to control something like that.

All this adds up to very wildly differing loot chance percentages depending on various factors. Only a huge database like you, Loviatar, claims to exist and is readily available can truly show a rough average for drop rates all over the lands of Tyria. Again, I'm not saying it doesn't exist at all. I am saying I would love to see it. I am also saying there may be a lot more that affects the percentages than just what is and isn't exempt from loot scaling. And no, it has nothing to do with luck, probabilities are not affected by luck - that is proven

It does, however, have everything to do with the developer's code.

Chthon · Jul 23, 2007, 04:18 AM // 04:18

Quote:

Originally Posted by manitoba1073

OK Ill try to explain it as simple as possible. Since some people are confused alot on how Loot Scaling actually works while soloing and not soloing.

First all drops are random.

Now in a zone that has a party size of 8 you have a solo drop rate of around 24%-28% chance to get the drop.

Now in a zone that has a party size of 6 you have a solo drop rate of around 32%-40% chance to get the drop.

Now in a zone that has a party size of 4 you have a solo drop rate of around 48%-56% chance to get the drop.

Now for preseering you have a near 98% chance since its a maximum size group of 2.

That is just the effects of Loot Scalling alone on soloing, with HM involved the only difference is you have a higher chance on the exemptions dropping instead of whites. In NM Its similar without the higher chance of exemptions dropping, while still being exempt.

Now for the rest of it.

Loot Scaling had NO EFFECT on its own except lower drop rates for less than max groups.Dont believe me look at the first 12 hours it was driving prices on everything up, then Anerf had a OH %^$& moment, and added the exemptions list is what allowed the prices to come down by increasing supply.

With HM the only difference is that the chance for drops from the exemptions list is increased.

So for all those out there that thinks it was Loot Scaling that helped you, you are wrong, it was the exemptions list.

Also mods try to leave this stickied so people dont get confused on Loot Scalling anymore.

I hate to say it, but this post is worse than useless. Construing Gaile's "about twice as much" as an excuse to just multiply by 2 and post imprecise statistics is little more than ex-rectum argumentation.

Quote:

Originally Posted by arcanemacabre

So far, Gawa's post was the best because it provides some actual statistical data.

True. Though, as much as it pains me, I have to agree with Lovitar - 1000 samples is not nearly enough.

Quote:

Originally Posted by MSecorsky

OK, I'm completely hypothesizing here, but this feels like what's going on.

A critter has a % chance in any area (which may vary) to drop an item. IF that creature is calculated to drop an item, then the item to be dropped is determined. If it's an exempt item, it is dropped. If it's an item subject to scaling then based on the number of party members there's a percent chance the item will not drop.

So... sum it up:

Creature Dies.
Determine if creature drops item (based on a percent chance to drop).
-- If yes, determine what item drops. Is the item exempt from scaling?
---- If yes, drop item.
---- Else, determine percent chance on item dropping based on party size. Roll random.
------ If the random is still in the percent chance to drop, then drop, else do not drop.

Now... this actually makes a simple flat percent to get something nearly impossible (or at best rather difficult) to calculate with any precision... but if I wrote the code, it would look something like that. Simpler from a logic perspective.

You're probably not horribly far off. (Arcane, since the odds on a dye drop are already near zero anyway, the fact that they aren't further reduced does not make them drop "like rain.")
If you're really interested in pursuing this idea, I suggest you look into how the D2 item generation/treasure class system works. I expect GW's system is a similar, improved system, since (1) a lot of the same people made both games, and (2) GW has the same telltale "certain-monsters-certain-drops" behavior. This is a good place to find the details of how the item generation system in D2 works; and this does a slightly better job of explaining the recursive nature the D2 treasure classes.

arcanemacabre · Jul 23, 2007, 04:37 AM // 04:37

Quote:

Originally Posted by Chthon

True. Though, as much as it pains me, I have to agree with Lovitar - 1000 samples is not nearly enough.

That's one thing I agree with Loviatar on, actually. I said it was actual statistical data, not that it was sufficient statistical data. In no way is that information enough to provide reasonable percentages, but it provides better information than made-up percentages based on misinterpreted quotes. That was my point.

The Diablo 2 drop rate system is all very interesting, but also extremely rational and logical that I don't need to delve into it to understand it. The only thing in dispute is the chance to drop when not in a full party (since the chance of any drop for a full party is 100%). That chance, which is represented by a percentage, is said to be proportionally scaled down the fewer and fewer people in a party. I suggest that the chance may also be affected by other factors, programmed into the code, and cannot be determined any other way than by a release of that code, or averaged by a wide statistical database of drops from a variety of situations.

Loviatar · Jul 23, 2007, 04:10 PM // 16:10

Quote:

Originally Posted by arcanemacabre

Quote or it didn't happen.

i aim to please

linkie to developers statement

http://www.guildwarsguru.com/forum/s...03#post2751103

the important part

Quote:

The loot scaling for gold and common loot is not linear with the number of players in the party, and it includes an element of randomness, so while the difference is not easy to quantify, it is by no means a factor of eight. Advanced solo farmers may now earn about twice as much gold and common loot from farming solo as they would if they farmed in a party. While gold and common loot are thus reduced for these players by loot scaling, certain other types of loot are completely unaffected. For example, Skill Tomes are completely unaffected by loot scaling, so they still drop eight times as frequently for solo farmers than they do for people who play in parties. Thus, advanced solo farmers will find that certain types of farming are still extremely productive for them, but they may have to change what and where they farm if they want to earn as much money as they did before.

arcanemacabre · Jul 23, 2007, 04:32 PM // 16:32

*headsmack*

Loviatar, have you been paying attention? I know about that quote. That's the only one that's been thrown around like it has some pertinent information. Truth is, it's vague. Read: "may now earn about twice as much..." Take that as you will - but it is very much all about interpretation. The element of randomness just adds to it, since it's kind of a -duh-, of course it's random. Unless she's referring to some extra random code, which is what I theorize about.

I swear, this quote is treated like it's the holy bible of loot scaling. Of course, just like the bible, it's wide open for misinterpretation. Well have at your beliefs. I believe the numbers or the hard facts - this single quote provides neither.

Loviatar · Jul 23, 2007, 05:13 PM // 17:13

[QUOTE]

Quote:

Originally Posted by arcanemacabre

*headsmack*

Loviatar, have you been paying attention? I know about that quote. That's the only one that's been thrown around like it has some pertinent information. Truth is, it's vague. Read: "may now earn about twice as much..." Take that as you will - but it is very much all about interpretation. The element of randomness just adds to it

to be honest i have paid attention and in the short run they have to be vague to keep idiots from saying my loot wasnt X2.

what i am pointing out are two flat out developer statements that leave no wiggle room for interpretation.

1. flat statement that the loot scaling is not linear according to party size (how much is up to them but not a flat scale)

Quote:

The loot scaling for gold and common loot is not linear with the number of players in the party

2. that is confirmed by this other flat developer statement

Quote:

it is by no means a factor of eight

so how/how much is vague but those two are straight simple statements that solo will get a random amount more

arcanemacabre · Jul 23, 2007, 05:19 PM // 17:19

[QUOTE=Loviatar]

Quote:

so how/how much is vague but those two are straight simple statements that solo will get a random amount more

Ok, so we agree that a soloer will get some random percentage more than a person in a full party will get. Right on! Got no problems with that obvious statement.

So do you also agree that the percentages quoted in the OP are entirely out of thin air based on that statement? That's all I've really disputed, ya know.

TabascoSauce · Jul 23, 2007, 06:28 PM // 18:28

Whoa. You find funny in the oddest places.

The title from the OP suckers you in, promising the "truth" about loot scaling.

In reality, it devolves over 6 pages to Loviatar saying that they have "flat" statements that are....vague? (and anything but flat heh)

Thanks!
TabascoSauce

manitoba1073 · Jul 25, 2007, 04:01 AM // 04:01

Quote:

Originally Posted by TabascoSauce

Whoa. You find funny in the oddest places.

The title from the OP suckers you in, promising the "truth" about loot scaling.

In reality, it devolves over 6 pages to Loviatar saying that they have "flat" statements that are....vague? (and anything but flat heh)

Thanks!
TabascoSauce

Unfortuantely it is the truth. Taken straight from Anet devs and pretty clear cut statements. Now if you have anything to contribute please do so. And so far in 6 pages noones yet proven it wrong.

arcanemacabre · Jul 25, 2007, 05:10 AM // 05:10

Quote:

Originally Posted by manitoba1073

so far in 6 pages noones yet proven it wrong.

... or right.

forgot that part.

manitoba1073 · Jul 25, 2007, 08:23 AM // 08:23

Quote:

Originally Posted by arcanemacabre

... or right.

forgot that part.

Funny but I have proven I was right numerous times including with what you tried to show how I was wrong.

funkytoejams · Dec 30, 2007, 01:06 AM // 01:06

random??? thats just a lie.... please explain to me how a computer can come up with random numbers (other than having it monitor random natural events, like radioactive decay or weather conditions,etc, and then assign numbers to the readings), i think imho that its pretty safe to say they arent doing this , MATH doesnt allow you to follow the same equation and get a diff random answer everytime.

DivineEnvoy · Dec 30, 2007, 01:46 AM // 01:46

Quote:

Originally Posted by funkytoejams

random??? thats just a lie.... please explain to me how a computer can come up with random numbers (other than having it monitor random natural events, like radioactive decay or weather conditions,etc, and then assign numbers to the readings), i think imho that its pretty safe to say they arent doing this , MATH doesnt allow you to follow the same equation and get a diff random answer everytime.

In Java, there's a method called Math.random(), which auto generates a number between 0.0 and 1.0. Yes, math does allow you to have the same programming code and getting a different random answer everytime.

Mork from Ork · Dec 30, 2007, 01:55 AM // 01:55

Quote:

Originally Posted by DivineEnvoy

In Java, there's a method called Math.random(), which auto generates a number between 0.0 and 1.0. Yes, math does allow you to have the same programming code and getting a different random answer everytime.

That is incorrect - Math.random(), instructs the computer to use a pseudo random number generator to create the appearance of a random number. Computers can not do true random numbers unless, like funky says, they monitor a truly random event such as radioactive decay. Pseudo random generators do a good enough job for general use - but they DO NOT produce truly random numbers.

DivineEnvoy · Dec 30, 2007, 02:08 AM // 02:08

Quote:

Originally Posted by Mork from Ork

That is incorrect - Math.random(), instructs the computer to use a pseudo random number generator to create the appearance of a random number. Computers can not do true random numbers unless, like funky says, they monitor a truly random event such as radioactive decay. Pseudo random generators do a good enough job for general use - but they DO NOT produce truly random numbers.

In all fairness, the person did not ask for how computers produce a truly random number, but this person was merely asking how computers generate random numbers.

Secondly, Math.random() is only an example to how computers may produce a random number with the same settings; there are many other methods. How are you so sure that computers cannot produce the so-called true random numbers?

Qwertyfied · Dec 30, 2007, 02:10 AM // 02:10

Quote:

Originally Posted by DivineEnvoy

In Java, there's a method called Math.random(), which auto generates a number between 0.0 and 1.0. Yes, math does allow you to have the same programming code and getting a different random answer everytime.

Basically just agreeing with Ork here, but one can't retrieve true random numbers via software generators, they are only attained via natural randomness, such as an optical quantum random number generator.

ensoriki · Dec 30, 2007, 02:17 AM // 02:17

Wow you guys are silly.
Whenever someone asks what the evidence behind your numbers is,
you say, The theory of I dunno.

Quote:

Basically just agreeing with Ork here, but one can't retrieve true random numbers via software generators, they are only attained via natural randomness, such as an optical quantum random number generator.

Is that a computer thing though? Computers don't create true randomness.

If you want random numbers you have to ask an idiot.

Qwertyfied · Dec 30, 2007, 02:22 AM // 02:22

Quote:

Originally Posted by ensoriki

Is that a computer thing though? Computers don't create true randomness.

If you want random numbers you have to ask an idiot.

You really need a suitable diagram to explain it, but no it does not use computers to generate the number, it's where photons are fired at a semi transparent mirror, whereby the end location can be in 1 of two different places, creating a binary number. The fact that the photon incident on this mirror is intrinsically random and cannot be influenced by any external parameters, explains the randomness.