Whomever coded the RNG system for the mystic either sucks, or they intentionally weighted certain rolls far above others. I know a lot of people who really don't understand computers and random number generation will just derp out "random is random", but for everyone else I think it's obvious something is seriously off with the way she rolls changes. I've had far too many times, just like everyone else where she rolls the same affix twice, sometimes even with the exact same value. Also, way too often do certain affixes show up, how often do you see increased monster experience when you are trying to actually roll something useful? I was enchanting something last night and xp showed up at least once EVERY roll, and there were at least 5-6. Like I said, I don't know if they intentionally weighted certain stats or if it just has a bug, but whatever it is, it needs to go away.
Rolls fine for me, everything usually in one roll, two tops. Luck of the gods, though I've spent 20+ trying to reroll my stupid boots and they never get the right one
Try rerolling Vitality on a rare amulet to Crit Damage and see how many Vitality rolls come up. I eventually got a 58% roll after many attempts, but should I for some reason want to try for something more, I'll have to fork over 320k gold.
You do know if you click the question mark next to the affix it will tell you what possible rolls there are? Usually there aren't more than 10 in each category.
I'm not saying it's the case, but I've seen in other games where bugs like this only affect certain players, it depends on where the seed number for the RNG comes from. There is an MMO call Asheron's Call where one player complained monsters always attacked him first in a group, his name was "Wi" he even joked Turbine (the devs) had flagged his account to get attacked first. A few other people said they also had the "Wi flag" on their account. Since it was so rare the devs and other players just dismissed it as "observer error" or only noticing when monsters attack you first. This went on for a year or more. Once day a new dev was hired, Sandra, who looked at the monster attack priority code with new eyes. She found a bug where the RNG seed was based on character name and in very rare instances this seed value is just right so that that character name always comes to the top in the monsters attack priority list. So after basically being told he was wrong for such a long period of time, Wi was vindicated and the bug was fixed. Ever since then I've had a healthy distrust of random number generators in video games. Especially since it is very hard for a computer to come up with a true random number, they just simulate it.
And of those 10 there are certain ones that ALWAYS show up and as pointed out above with the amulet, some that virtually never show up. Personally I don't think it's a bug, I think they gave a slightly increasedweight to rolls on the less useful stats as a way to slow down power creep and drain more gold from the "economy".
I know a lot of people who really don't understand computers and random number generation will just derp out "random is random", but for everyone else I think it's obvious something is seriously off with the way she rolls changes.
Why do you even bother if you're just going to start out by slamming people?
I have a very good grasp on pseudorandom numbers. I have not experienced any issues with enchanting (I seem to get a very reasonable distribution of the potential properties and it usually takes me no more than 5 or 6 tries to get what I am after). You may be right that there could be a rare bug related to seeding, but, just as with the other thread, you're never going to "figure it out" or even get any positive attention if you act like a raging teenager who hasn't had his ADHD medicine.
The only time I've had real trouble with the enchanter is trying to get Arcane Orb damage on my source. There are just so many possibilities that getting what you want, and a high roll is really, really unlikely. It gets to the point where 500k+ enchanting costs are just too prohibitive (not to mention the Forgotten Soul cost).
Why am I not surprised you locked in on the line? As for why it's there, if you read any forums related to the game where RNG is discussed, how many times do you see people line up to do exactly what I said? I have literally seen "RNG is RNG" or "Random is random" hundreds, if not thousands of times. By putting that in OP it shorts cicuits most people who think they are clever by posting that.
I've noticed similar results as well. Blizzard probably seeds the RNG based off of the current minute rather than the milli/micro second most of the time just to troll us.
It's kind of a shame you didn't record your data so we could build up a usefully large sample.
In my experience, the main reason "RNG is RNG" crops up is because people almost invariably post their bad luck (or good luck) in the form of an anecdote, and which makes it impossible to tell what role confirmation bias, clustering illusions and other common fallacies had in the narrative.
Also, remember that if there are, e.g. 10 affixes, then the odds of both re-roll options being the same unwanted affix is about one in 11, which is fairly frequent.
Someone already said about the number of random properties that could be rerolled, just press the exlamation mark and check. For example, did you know, that you couldn't roll AllRes if you already had any resistance as one of the secondary stats?
Or, there are certain categories of stats (besides primary and secondary). For example, if you want to reroll certain skilll damage % you are likely to get skill damage % but for another ability or if you reroll vitality, the probability of getting main stat or armor is higher that getting %skill damage or offensive stat like crit chance/damage.
It's kind of a shame you didn't record your data so we could build up a usefully large sample.
In my experience, the main reason "RNG is RNG" crops up is because people almost invariably post their bad luck (or good luck) in the form of an anecdote, and which makes it impossible to tell what role confirmation bias, clustering illusions and other common fallacies had in the narrative.
So much this. So much. Over and over and over and over again.
"I get bad RNG on enchanting" doesn't provide a jump-off point for identifying a problem.
There are a bajillion potential bugs that could exist relating to player, region, item type, item quality, specific set/legendary items, existing statistic, destination statistic, time of day, which Act you're standing in, number of players in the game, etc.
And then there is the fact that someone COULD just be the unlucky guy. Just because a handful of people report "bad RNG" with enchanting in a pool of millions of playes that doesn't really mean much because it's well within statistical expectations. As you pointed out, in a pool of 10 properties, there is a 1-in-10 chance that you could see a duplicate. And, duplicates aren't always bad, since the first roll may have be lower than the second roll. Imagine you're enchanting your chest and sockets pop up twice. The first one is 2, the second one is 3. In that case (and any similar cases) you've actually benefited from the duplicates.
I too have noticed a tendency for her to hone in on specific stats. This is basic probability. Look at the possible outcomes -- you can find some where the outcome could be any of nearly 15+ affixes. The probability of both returning the same affix multiple times in a row if the results were truly random is too low to be a coincidence (aside from the fact that it would be a coincidence if it were really random). But I'm certain it is not. Do you really think they want people to be able to throw a piece of gear in there and have the same chance of getting level requirement reduction as they do for monster experience or gold find?
If Blizzard wants me to test and gather data samples they can give me a testing account with unlimited mats and gold. Due to the mats and gold required no one is going to be a able to make a good enough to sample size to satisfy the "RNG is RNG" crowd that is basically taking over the thread. You're using more words, but that is essentially what you are saying. Also as in my above anecdote, bugs don't always affect everyone.
I've had crappy weapons for alternates roll sockets time and time again after only a couple rolls. I've spent about 25 Forgotten Souls trying to get a socket on a 2-hander that I found for my Crusader, and another 10 today trying to get a socket on a Serpent Sparker for my Wiz. Didn't get it on either.
So newsflash: RNG is RNG. More at 11.
Or if you wanna actually discuss whether there's some influence or bad coding there, we're gonna need some real and extensive data (either from datamining or long term data gathering), not random blabbing.
I think Blizzard could fix people complaining out RNG (or lack of it) by putting a "Buy Now" option on the Mystic. Roll for a chance at your stat a 1x normal price or buy it now (at 85% max roll) for 10x cost.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
I have a very good grasp on pseudorandom numbers. I have not experienced any issues with enchanting (I seem to get a very reasonable distribution of the potential properties and it usually takes me no more than 5 or 6 tries to get what I am after). You may be right that there could be a rare bug related to seeding, but, just as with the other thread, you're never going to "figure it out" or even get any positive attention if you act like a raging teenager who hasn't had his ADHD medicine.
And may the odds be ever in your favour.
Emmo#2406
and ill take my cake and eat it too
In my experience, the main reason "RNG is RNG" crops up is because people almost invariably post their bad luck (or good luck) in the form of an anecdote, and which makes it impossible to tell what role confirmation bias, clustering illusions and other common fallacies had in the narrative.
Also, remember that if there are, e.g. 10 affixes, then the odds of both re-roll options being the same unwanted affix is about one in 11, which is fairly frequent.
Or, there are certain categories of stats (besides primary and secondary). For example, if you want to reroll certain skilll damage % you are likely to get skill damage % but for another ability or if you reroll vitality, the probability of getting main stat or armor is higher that getting %skill damage or offensive stat like crit chance/damage.
"I get bad RNG on enchanting" doesn't provide a jump-off point for identifying a problem.
There are a bajillion potential bugs that could exist relating to player, region, item type, item quality, specific set/legendary items, existing statistic, destination statistic, time of day, which Act you're standing in, number of players in the game, etc.
And then there is the fact that someone COULD just be the unlucky guy. Just because a handful of people report "bad RNG" with enchanting in a pool of millions of playes that doesn't really mean much because it's well within statistical expectations. As you pointed out, in a pool of 10 properties, there is a 1-in-10 chance that you could see a duplicate. And, duplicates aren't always bad, since the first roll may have be lower than the second roll. Imagine you're enchanting your chest and sockets pop up twice. The first one is 2, the second one is 3. In that case (and any similar cases) you've actually benefited from the duplicates.
So it's not all bad.
So newsflash: RNG is RNG. More at 11.
Or if you wanna actually discuss whether there's some influence or bad coding there, we're gonna need some real and extensive data (either from datamining or long term data gathering), not random blabbing.