Agostine!!!!

[gbrish]

This quest has driven me mad. I have gone through about 200 white furs and I still can not get him to make fur boots for me. Last trip I did I brought him 60 furs and he tore all of them (just for the boots).

So what was your experience with him? has anyone else had this problem?

[Chicka-Maria]

varies on character, i had a character use 200 furs and another use 2 furs lol

[KeeKee]

It's the script writing that's at fault. Random. It resets every time a fur is rejected.

This comment won't help you here but on Land Of Fire, we fixed it. Thank you Dingo Duck and Pihro!!

[Frost]

I've had similar experiences to what Chicka described: anything from almost immediate success to having to murder another family of muggles for the fur.

This comment won't help you here but on Land Of Fire, we fixed it.

How did you fix it? Did you set a maximum number of furs? Did you make the chance cumulative (e.g. 3% the 1st one, 6% the 2nd...90% the 30th)?

[SudoPlatypus]

.

[o11c]

I don't mind so much after my first character, since I have hundreds of white furs due to coal grinding.

[KeeKee]

I've had similar experiences to what Chicka described: anything from almost immediate success to having to murder another family of muggles for the fur.

This comment won't help you here but on Land Of Fire, we fixed it.

How did you fix it? Did you set a maximum number of furs? Did you make the chance cumulative (e.g. 3% the 1st one, 6% the 2nd...90% the 30th)?

Frost, hi. Yes max number of furs. The player is told by Agostine approximately how many it will take to have a good one.

[Reid]

I've had similar experiences to what Chicka described: anything from almost immediate success to having to murder another family of muggles for the fur.

This comment won't help you here but on Land Of Fire, we fixed it.

How did you fix it? Did you set a maximum number of furs? Did you make the chance cumulative (e.g. 3% the 1st one, 6% the 2nd...90% the 30th)?

Frost, hi. Yes max number of furs. The player is told by Agostine approximately how many it will take to have a good one.

It doesn't seem so random after that.

[Pihro]

The quest still is random, one of my chars got it after three tries, another took 39 tries. It can be fun up to so far imo, as a player i hated forking out 200+ furs, it got old quick. One of my main points in creating this server is to restore some balance to TMW gameplay and gameplay style, and to listen to the players. A game where players resent what the devs make them do imho is not a good way to make a good game.

[Dingoduck]

Unfortunately, many game developers believe that the RAND(30) will produce an average of 39 or so to be successful. However that would probably be the "BEST" probability. Even though, through statistical, cumulative calculations might show the average to be 39, the real time application and tracking igt show the average to be well over 150. With a resetting random number generated, the probability of hitting the ONE number to win can approach "Game Player Infinity". Where so many resources and time is spent on only one objective, the player feels that the quest is now a burden and gives up on the quest. Without quests, the game looses it's attraction...the player leaves. They might come back, but not to attempt the quest more than once or twice more.
There are methods called shuffling which makes the experience "more balanced" or where the number selected previously is eliminated from the next selection...giving a maximum amount.
To avoid adding more complex code to the game, 2 lines were added to count the number of items used and if a limit number had been reached the player would obtain the prize. The number chosen was the probable average. After polling many players, it was felt the limit was very welcome and that they would do the quest again. A quest they did not like before.

[o11c]

Unfortunately, many game developers believe that the RAND(30) will produce an average of 39 or so to be successful.

From a mathematical perspective, rand(30) will require an average of 30 calls to be successful. TMW uses a mersenne twister, which is widely acknowledged to be a very good source of random numbers. I have personally observed, after doing this quest with several alts, requirements along the lines of 2 or 20 furs (or raw logs for the forest bow).

There is a theoretical failing that will result in slightly skewed results on the order of 1 in a billion times. But that's in your favor anyway.

[gbrish]

011c your luck must be very good.

Sad to say I have massacred quiet a lot of the poor fluffies but I have attained my fur boots... I think the chances are better when you have a lot of furs with you before going to Agostine. (whether or not that is supported by any form of mathmatics)

The first time I did the quest it was only 20 or so furs i needed. The second part for the boots was where I was making trips with 30 and 60 furs and getting no results. I guess it is bad luck.. but not enough to make me leave the game. lol

[Nard]

The big scatter of this kind of achievement is caused by the probability model which is employed. There is no simple way to have a better one except by programming it with eA script, and eA script is not designed to perform complex math operations

if p is the probability of a fur to be a good one, 1-p is the probability to be a bad one.
The probability that the number of furs you have to give is larger than a given number n is the same as the probability to have no good one in n trials, thus (1-p)^n
If p is a small number, you will find that a quite high number of furs is still likely. A very high number of furs is not impossible.
The same applies to logheads and forest bow, The jack'O souls in love triangle, and in a less important way, to Book Pages in Blue sage quest.

I think the chances are better when you have a lot of furs with you before going to Agostine. (whether or not that is supported by any form of mathematics)

The chances are exactly the same if you submit the furs one by one or if you come with a lot, you just spend less time in the second case.

Nard

[o11c]

011c your luck must be very good.

No, I've just done it more times that you have so statistics has evened it out.

[Doofus]

From a mathematical perspective, rand(30) will require an average of 30 calls to be successful

This really is misleading as for the agostine/loghead quests it is a geometric distribution as Nard indicates rather than a normal one so the usual notions of average and standard deviation are quite different. Although the expectation value is 30 the distribution is highly skewed. On the up side 50% of the time one finds that <20 tests are needed (i.e. -1-log2 30/(30-1)). The trouble is when one isn't in this happy regime - the skew of the median being that much lower than the mean is balanced at the cost of the way the distribution tails off to infinity much slower than a normal distribution and there are a lot of people that end up stuck in this unhappy "why am I still playing this" regime.

I'm definitely for these potentially infinite values to be capped or better yet for the probability to slowly increase to help it tail off faster (after all the trees should be getting stronger via the process of natural selection if I find I'm only weeding out the weak ones).

I liked the blue slime quest much better as the probability being around the 0.5 mark made for a more symmetric distribution (also the fact you get experience for each dud was a nice concession).

The chances are exactly the same if you submit the furs one by one or if you come with a lot

In theory one would expect this but then how is the random number generator reseeded? Hopefully purveyors of Douglas Adams events are aware of the need for external sources of entropy - A bistromatic drive is really not that hard to code.