2 seasons in a row, with characters & items starting around sometimes as early as level 40 I will start tempering as the cost is low enough. By time I reach 60 and T1 I will be finding lots and lots of items I will temper. What I am trying to say is I have enough firsthand experience to know and understand the OP’s post.
This is not just bad design, but poor implementation. The game will fixate on one affix. You have 3 options, and it will continuously give one of them repeatedly. Just not the one you want. Masterworking can show the same effect. You want the Stat affix and you get the resource cost reduction. But you do not need Resource, but it will chose that 4 out of 5 tries.
If you used a realistic sample size according to the game being 5 rolls that’s be 100% fail rate yea? I understand bigger sample size equals more realistic findings, however it’s a video game y’all act like it would be hard to add a line of code that doesn’t reroll the same stat twice. 3000 rolls of temper are multiple seasons of d4 for most people so yea if it takes that long to even out I doubt it will feel fair to players
Ok, so its just that your sad you didnt get what you wanted. You can absolutely roll what you want multiple times in a row as well.
95% of what fails for me is when im greedy and want a better roll.
Its no big deal, just find another item. If tempering is so aweful ask for drops to be like S0-S3 where all stats are available on drop. All tempering does is complete the item with much higher odds getting what you want.
This was done by pathological in S4. He was a hardcore troll hating everything D4 so I actual believe him. If failure rate was actually as high as some of these posters youd see a lot more people taking the hours/days replicating pathological.
I keep seeing that its not weighted, but it sure as heck feels that way (like with masterworking). Consider bringing some decoy items to temper/MW to eat the undesirables results. Not sure if it actually works, but it makes me feel better
Also, if you feel like you’re getting screwed by the RNG Gods you can always re-log.
In the unlikely event that the RNG that Blizzard uses is bugged, this will outright reset it back to baseline. You could probably also re-launch a session but eh.
I do this sometimes when I feel like the RNGods have been unreasonable too long for just about any game.
Currently, I have no data to back it up, but it seems to me that the code is randomized for each individual affix percent so the likelihood of getting a crit damage 85% is the same as a crit damage 76%, is the same as an overpower 155%, etc. This is why I personally don’t think it is weighted.
What a middle ground would be between what we have now and just choosing the affix, is a weighted system to first roll a 33% for 3 tempers or 25% for 4 tempers and then roll the percent on the affix. It would make it really difficult to brick unless you’re going for a higher percent, I think you could take away the scroll of restoration if you did this. It would be similar to gear in D III, it first would roll what piece you would get (ring) then roll what ring you would get.
The change I would make is smaller ranges for affixes and 3 per temper manual.
No, but he is not the only one is he. When enough people dislike something about your product it definatly is an objectivly bad thing.
I don’t know about a “SINGLE” person complaining about the number of stats on an item.
there was plenty complaining about “Dmg on Monday 9 o’clock” stats, but i have not seen a single person asking to remove number of stats. And i spend a lot of time in this forum.
well, you can buy GA items, but you cant buy an item with the tempers you want. Pretty obvious isnt int :P?
just like im not the only one liking the system. Does that make the system objectively good?
oh, there were tons of complaints because getting a 5/5 item was pretty rough to get to drop. If anything, especially the zommer playerbase, should be absolutely grateful tempering exists, as it leads to far earlier build enabling builds.
To the OP: that is precisely why I have been screaming since S4 to just let us choose the affixes already and leave all the RNG into the results of the rolls.
Lower the floor on most affixes so the average roll is worse to balance it out, with the real risk in min-maxing instead of praying you get the thing you wanted in the first place.
And I’d said this before as well, but **Who the hell goes to someone to ask for a custom job but the results can be random?! **
Well yeah but he went on a mission to validate weighted claims. Hes the only one that has to date. Ppl normally temper 1 item and have hurt feelings so they need to tell everyone about it. The only people that I could see having hurt feelings is if they swiped there credit card expecting bis.
Its just part of the process. Even when im not being greedy I will maybe, maybe use 1 or 2 scrolls a season.
But in that same problem, if a person only does 50 tempers in a season and uses the same test it would be said the test variable needs more time to even out … If that’s all he does in a season there is no evening out and the test should be done on the real parameters of a season yea? What I see is he did the test until he got the outcome he was looking for.
If your out to prove a data set it needs to be big enough to do just that. 50 is to small of a sample size imo.
To truely collect the data in proving if tempering is weighted you would need at least 500-1000 items. Now, idk anyone who cares enough to take that much time to do that outside of that one user. Most people wont go down that path becausr youre more likely to get what you want then not.
Im not saying you dont roll suboptimal as i myself use 1-2 reset scrolls a season. More if Im being greedy and gambling for a higher roll but at that point if it doesnt work out it was my choice to push.
But we need a sample size that reflects a season yea? Or is the test for multiple seasons? I didn’t even do 50 tempers this season just going by my personal experience
First off how can there be any test variables for a single use? What are you testing it against when there is multiple choices? Point again I’m making is no one is tempering that many times in a season, do a test with an avg say 150 tempers? Seems prob avg for most players
I hate how tempering is done when it comes to enchanting but I don’t believe there is any weighting for tempering. RNG is just sometimes working funny in this game…OK, it is not funny when it happens.
I already commented that a couple of times but since the argument keeps coming back, I believe I have to repeat myself. The 3000 tempering tests were done by someone called Pathologic if I remember well. He looked at the results, saw that they were close enough and concluded that there was no weighting.
There is a mathematical test called Chi-Square Independence Test that checks if variables are related. If you use it on the 3000 tempering results, the conclusion is that they are unlikely to be independent, suggesting possible weighting. The 3000 tempering test was great, the conclusion was wrong.
I said “unlikely” because we are again talking probability and extremely rare events can never be excluded. The conclusion was still wrong.
Anyway, I would just explain that by funky D4 RNG. Weighting would have resulted in an even higher difference between the results, hence me saying that tempering is not weighted.