test content
Logo
What is the Arc Client?
Install Arc
60% Chance to Make Gemmed Pants/Shirts... uh, no
rsomazzi
Member, Neverwinter Beta Users, Neverwinter Guardian Users Posts: 16 Arc User
I've made a slew of pants and shirts, with epic Platesmith and epic tools. Although the UI says 60% chance to get a gemmed version, I am barely earning HALF that.
Perhaps they got the math wrong, or they've engineered crafting so that you cannot get the advertised rate. I know it's an average, and it's not guaranteed, but I am too far below (30%) the advertised rate to think this is really a 40/60 roll.
Anyway, I wanted to hear what other crafters are averaging with their shirts/pants.
thanks
Perhaps they got the math wrong, or they've engineered crafting so that you cannot get the advertised rate. I know it's an average, and it's not guaranteed, but I am too far below (30%) the advertised rate to think this is really a 40/60 roll.
Anyway, I wanted to hear what other crafters are averaging with their shirts/pants.
thanks
Post edited by rsomazzi on
0
Comments
I have failed 4 95% rolls in a row. Now that is quite a feat.
But unless we see the actual formulas or have someone willing to put in a very large set of numbers it is all but impossible to know for sure.
That pissed me off so I went to sell off my epic tools for a little under 4M AD (I had invested >4M to get them)... I am glad I did...
That said, I've always found computer-AI RNG algorithms in games to feel a bit off, especially depending on what they're using as the seed. They definitely do have tendencies towards going in "spurts." You'll succeed 5 times in a row then fail 15 times in a row etc. I've seen whole articles based around this. It's not something that's easy to solve, or at least maybe isn't worth it to get/do better algorithms for video gaming. Even wiki mentions this trait:
http://en.wikipedia.org/wiki/Random_number_generation#Computational_methods
While you are correct about sample sizes, you miss the point that there is definitely something off about the RNG/success rate.
With a 60-40 split, a 200 sample size has a margin of error of 6.8% and a 300 sample size's margin is 5.6%. So the average is 6.2%.
A 60% success rate with a 6.2% margin of error, should provide the true success rate of ~56%-64%. Why is my success rate below 40% and closer to 35%?
And no one ever sees the margin of error in their benefit, always benefits the failure side. There is something very wrong with the RNG!
Actual, this is at least in part a manifestation of a psychological effect called negativity bias, wherein people remember when things didn't go their way far more easily than they do when they did.
http://en.wikipedia.org/wiki/Negativity_bias
I don't numerically track my results. But I do get more successes than failures when I expect them and more failures than successes when I expect that.
Neverwinter Census 2017
All posts pending disapproval by Cecilia
I remember back in WoW, people constantly complaining about not getting the 15% chance improved Alchemy results from the various Masteries. But when you run the numbers, it's quite possible to get failure runs of 10-20 in a row (1:1000 odds are still pretty good, when you've got enough people doing something. If 100k people are involved, 100 will have that "bad luck". And if only a quarter go to the forum to complain, it still looks like "everyone" is getting <font color="orange">HAMSTER</font> results, OMG it must be broken!)
See, first mistake.
Seriously, I tend to get a success way more often than not these days. I exclusively run one armor task at a time, and am on Mindflayer shard. I'm also doing cloth shirt/pants.
Make 1000 then get back to me with numbers, you might have a case then
However a simple math error where the chance becomes 1 - % chance instead of just (chance) is entirely possible. (so 40% instead of 60% as intended)
"feels" like that to me too with the many shirts/pants i've crafted. But never did any solid tests.
Since the same bug existed for a long time with several things (EG: Valiant Knight set was proccing 75% of the time instead of 25% for months.)
Though if that was the case, trying with the just say 10% on the a rank 3, should give you a 90% shot at it..
-Epic Dread Vault Crushed.
Characters (Dragon): Axer (60 Guardian, Leader of Crush It!), Controller (60 Wizard), Warlocker (60 Warlock)
I have 1 blue tool and overall I have 'felt' I was doing much better than expected with mailsmithing then I hit a bad run with cleric pants, 4 unslotted in a row.
When i failed looting a skill node, and tried again quickly a couple of times. I could could get into a kind of everlasting fail where every one failed until I took a couple of seconds pause before trying again.
I guessed this had something to do with when the system updates "random" values. Like new seed after every other second from last start of attempt, which could end in this loop, if there was not a couple of seconds between starts.
But that is of course pure speculations... and has nothing with the long periods in professions
No amount of false fallacy covers an almost regular appearance of rolls of both 1 and 100 within the same dungeon run by the same player with at most 4 bosses/chances per run.
There is definitely something off about the RNG in this game.
Well if they got the lockbox issue wrong, who's to say it can't go the other way. The thing about the lockbox situation was that it would lose them money and was completely obvious. When odds are too low, then it's pretty hard to find out if there is an issue. I'm not trying to accuse them, but I can't have faith that they have it right and if it was wrong, I don't have faith that they'd notice/fix it in a hurry. Having said that, there seems to be a lot of bad streaks in this game.
Whatever it is, it's simple and easily overlooked.
Neverwinter Wonderland
Neverwinter Wet & Wild
Neverwinter 2nd Anniversary Montage
Neverwinter Anniversary Montage
Protector's Jubilee Speech
Oh the Carnage
You're probably right. It's not consistent, and of course it's a percent (I'd say 70% is more realistic considering how much AD you have to invest in it in the first place), but yeah, sometimes I'm like WTF.
Is it possible it's messed up? Sure but we'll need a MUCH larger sample size to even consider this
Incidentally these types of threads pop up in any game i've ever played which contains some sort of random mechanics
Sadly, not understanding the principles behind these things doesn't keep people from posting their irrelevant opinions
I was mostly using your quote to add to my opinion on the matter rather than arguing with you :P