test content
Logo
What is the Arc Client?
Install Arc
Yet anothe Masterwork bug MW tasks failing constantly
dopefish18
Member Posts: 4 Arc User
AFter noticing the past few days that my masterwork tasks were consistently failing even with a 75% chance I asked some other players and they are experiencing the same issue. It appears that it is not calculating the added quality bonus of items put in the optional assets. Luckily I found this before wasting millions of AD unlike my guild mates.
0
Comments
3 out of 3 were successful in crafting the oil so it was definetely getting at least 75% to get it .
You must have hit a bad run of RNG more than likely , which isn't unheard of in this game .
When your attempts fail what level of reward are you getting and how many others are having similar issues and what tasks are they also attempting ?
If there is any issue it would be beneficial to at least list what profession / task that is being attempted .
Any of my comments not posted in orange are based on my own personal opinion and not official.
Any messages written in orange are official moderation messages. Signature images are now fixed!
Tailoring:
Commissioned Awning failed at 75% x 1
Cashmere failing at 75% x 3
Cashmere Yarn failing at 75% x ~15 with 1 success.
Alchemy:
Oil of Vitriol seems to be failing about 50% of the time at 75% for x ~10
So not sure if it is tailoring specific until I get more details from others. Hard to do a lot more testing myself since the tasks are long and expensive. Will update when I get more information.
Guildmates:
Alchemy:
Oil of Vitriol failing at 75% x9 with 1 success
Another guildmate:
He states that only getting about 20% success rate while doing Aqua Fortis and Gold Nuggets at 60% chance.
None of us are sure on exact numbers as we had not been tracking it until we realized there was a problem. We are all going to start tracking rates as of today.
Copy your toon to preview and do at least a couple of hundred.
Materials Before.
AD Before.
AD After. Final task failed.
(94033488-89434577)/8000 = 574.8 ~~575.
To check at a 2.5% significance level with a leeway for results of 0.05, I need a sample size of the following:
(2.24*sqrt(0.75*(1-0.75))/0.05)^2 = 376.32, my sample is 575, so it is significant.
Null Hypothesis: X=0.75
Alternative Hypothesis: X=/= 0.75
Lower Rejection Region: -2.24
Upper Rejection Regoin: 2.24
Number of Successes: 424
Number of Trials: 575
Test Statistic: (424/575-0.75/sqrt(0.75*(1-0.75)/575)=-0.698237948
P Value: 0.2776
Result: There is not sufficient evidence at a 2.5% significance level to reject the null hypothesis.
At the very least, it seems likely the chance to craft Oil of Vitriol to be correct. I find it highly unlikely that the chances are incorrect in general and see it more as anecdotal evidence. I may check the chances on say Bronzewood Raid rings later, when I don't mind hurting my wrists again. Friendly word of advice, don't craft this stuff, especially not the awning.
I agree with your conclusions on your data, but mine across various crafts all at 75% show a different story.
I'm 148 successes out of 225 attempts and have basically given up on mastercrafting for the moment because of this as it's bankrupting me (I'm also way down on the expected chances upgrading enchants recently with multiple 5%s and 3%s taking significantly over 100 attempts)
mean = 225x.75 = 168.75
variance = 225x.75x.25 = 42.1875
st dev = sqrt(variance) = 6.495
95% of results lie within 2 SDs of the mean, and 99% within 3, I'm outside 3.
Also - was your test on preview ? anecdotally I tend to find success rate is worse when the server is busy/laggy, I'm not sure how that compares between the servers.
Yes, my test was on preview, in port nyanzaru, next to the mailbox (I would never craft oil of vitriol on live). No, I do not believe the instance you are in, nor the number of people in it, nor any other number of external factors modify the outcome of your rng test and I am not about to ask 39 other people to help instance block an instance on preview for a control group. I craft a lot on the live server and calculate profit margins based on the given chance to upgrade. I have consistently made an expected amount of profit over a long period of time using the given chance to calculate and if it fell outside of that amount, I would myself have reason to be suspicious. As of yet, I have not (and I have kept a meticulous record of how much I expected to earn).
A little story. When they fixed the allosaur fang drop rate, I failed 11 ornament crafts, with only 1 success, in 1 crafting session. It was brutal. I told myself however, that in that particular instance, I had just got unlucky and that over time it would even out. Guess what, over an extended period of time, my expected profit did come to match my actual profit, despite that particular streak of failures. Sometimes, you will get streaks of failures (and successes!) when crafting, its the nature of RNG. You can either give up and complain about it, or tell yourself that it was bound to happen in a large enough data set and that in the long run you will make it up somewhere else. I choose to do the latter.
Not to call you out or anything (as you did your work properly) but are you sure you didn't make any mistakes recording data while checking this? I very specifically went about this with the intention of checking the chance and made absolutely certain to keep track of things properly. I find it likely (especially if you recorded that over a long period of time) that you would have made mistakes recording stuff (leaving out some successes etc in between) which would skew the results, especially since, in a larger data set, my results were not able to refute the given chance.
The sort of thing I'm seeing is it's as likely to see 3 fails in a row at 75% as 14-15 successes in a row. I'm something like 6:0 for that. I'm fairly strongly convinced that for whatever reason p(success when last trial was a fail) is NOT equal to p(success when last trial succeeded).
https://www.arcgames.com/en/forums/neverwinter/#/discussion/comment/13029451 for the experiences, first and last post that did this to me.
And really, why are you making the lichstone belt, that thing is terrible and I would never stick it on a character. Every time I see 1 of these MW rng complaint threads and I see people list the stuff they crafting I find myself asking "why are you crafting such terrible items in the first place."
I did craft a tremendous amount in the previous mastercrafting age, and the RNG was RIDICULOUSLY streaky, but the overall number of successes was a fraction low but within bounds, so I stopped recording. Then I felt it started not to be, so I recorded every attempt thereafter. I don't make stuff to sell on the AH, I craft to order for friends and guildies, with profit not the main motive, but when normally you can make stuff 1/3 cheaper than the AH, and then you get a period where 18 items in 20 are 90% AH cost or more, it's time to give up.
I will also say that your test was on an ingredient from the previous mastercrafting. I got flush on 75% for those on a 4 figure number (I split my recording of those and the final crafts from that set of MC, and the final crafts were c70% iirc). My issues have been with the most recent mastercrafting, and I'm like 2/8 for final crafts included in the stats I gave. The few old mastercrafting tasks I've had to do recently rather than just spending the GMs have been around par on a small sample size.
Up till recently, time of day had a major effect, the ONLY time I got the advertised rate or better was like 5am Eastern, 7pm Eastern was worst (I'm in UK so didn't try a little later which might be even worse). Now nothing seems to work.
That's 10.7% chance.
Would you have said the same if you had 80% chance of success and succeeded 10 times in a row? Doesn't sound so improbable doesn't it.
I am talking about maybe 200 tasks done with 75% over the last few weeks.
I had a very bad streak once, too. Failed with the stupid watch task at 35% 17 times in a row. But that was in February and it was the only "bad" incident. And with the cost of this streak being 17k GM, i can live.
Related to quality of task is some conspiracy theory stuff - craft at least a 1000 and report back.
Related to quality of task matches my experience, but not excessively so. The real issue with the new mastercrafting is how penal fails are, if they gave you a decent amount of the materials back, would be a lot more bearable.
@nitocris83 maybe something for the team already.
I would start to worry if I got 2 one in a million events in a short space of time unless I was doing long term monitoring.
To get 1 in a million you need 10 fails in a row. Log how many attempts and how many successes you get for your 50K and that starts to be worthwhile data.