Because, people are making the argument that they do not see a specific piece of needed stuff and the response is that that is how RNG works.
I am saying RNG is only one element in play. Frequency is also governed by how some items are weighted by the coding. If someone is not seeing something they are looking to get, RNG does not adequately account for it.
Imagine a lookup table like this…
001-075 : Sword of Normality
076-090 : Sword of Superbness
091-099 : Sword of Awesomesauce
100-100 : Sword of a Thousand Truths
The game rolls a 100-sided die, checks the lookup table, and gives you the item from the list that matches the range the die roll is within. Clearly, it’s still RNG as to which number you get on the 100-sided die, but you’re going to get 75 times as many Sword of Normality than Sword of a Thousand Truths because of the weighting of the table.
1 Like
That is exactly what I am saying. Excellent analogy.
It is, and that’s RNG, although in game there’s even more levels of RNG involved.
Others and I asked before how RNG is RNG ? Every time there is always someone who said RNG is RNG. Clearly it’s a misleading answer meant to confuse not clarify.
Because none of use can clarify it. It’s like trying to explain why a dice fell on the number it did. There are to many variables to accurately explain that so we just call it random.
Why are you trying to override what Foxwalker said ? The fact is he is trying to clarify and you are trying to confuse.
Because he’s wrong? RNG in no shape or form implies everything on the table has the same drop weight.
How is it any clear when he uses phrases such as;
This quote here. How lines of code, a random number generator would “assume” anything? Have I missed something? There are excel sheets out there with average bloodshards and upgrades to get certain items, which differ from each other as well. That’s the quirk of the system.
Game doesn’t owe you a favor to reveal its odds. Similar thing applies to legendary gem upgrades as Urshi can show 60-70% success but watch you fail 4-5 times in a row at one go. Even with 60% success rate, it’s a rigged coin toss that fails 40% of the time.
If player is not seeing what he seeks for then example size must be questioned first, not the lines of codes. A pseudo random number generator has always been coded pretty standard to replicate results from a total random number generator. If you have any doubts you can curb your curiosity by taking a larger example sizes or checking the excel and google sheets out there.
That’s why i said it’s misleading.
So, me correcting someone that made that argument makes me the one that’s misleading? You have to walk me trough that logic.
You can’t just get technical on the term RNG alone, doing so is misleading. You need to tie it in the context of D3 gameplay, loot table, drop weight whatsoever.
2 Likes
But we can’t, we don’t have that information. We can guess, but that’s all it is, a guess.
Purposely misunderstanding what I am saying. Some of you bandy about RNG as if it is the magic answer to why you don’t see the thing you are waiting to drop. A casual player may not understand that the randomness of the drops are modified by the behind scenes manipulation. The Randomness is not just a coin toss, so the thing you are looking for may not appear only due to that factor.
If you want to deliberately misunderstand what I am trying to point out that is on you. Honestly I think we are all saying pretty much the same thing. I was only pointing out that the Randomness was skewed by those behind the scenes mechanics.
Thanks,
My standing point is that you said;
No such thing. RNG doesn’t assume anything as it’s a piece of code, player does; if it has equal chance, then it has equal chance, if it doesn’t then it won’t appear as much in a small example size. Repeating rolls are always a problem and game developers use something called “bagging” to deal with such thing but sometimes game can just decide it’s not necessary or some stutter in the connection happens. If you check Game Developer Conferences at the previous years you’ll come across to some speaks about randomization and how it’s handled by algorithms.
Gem upgrades is a coin toss, it heads or tails, success or fail; the trick is that its set to succeed or fail for a certain amount of percentage in a large example. It’s more of a roulette spin that you can always bet for red or black with ever changing rates.
The smaller the sample size more likely a player will think they’re tricked in that model. Because if there’s a chance to fail then Murphy’s Laws will execute, else player will enter the Gambler’s Fallacy. However, as example sample size expands the odds just meet the announced odds. There coding tricks behind the lines but neither I, nor you are eligible to tell.
Why do you assume you’ll get tricked by a roullette spin that you can not control the outcome of anyway? For any gem upgrade, random number generator could set a series of non-consecutive numbers from 1 to 100 for success ratio and others for fail; creating a roullette table.
I can pick 30, 80, 60 or 50 non-consecutive numbers out of a number series that goes to 1 to 100 and set them to fail or success, at each spin. Creating a rate of black or red non-consecutive random numbers for each roll in a repeating series, doesn’t mean randomness is skewed. That only means there are pitfalls but rates are more or less still same.
For affix rolls and enchanting it’s multiple dices rolled for multiple times, player doesn’t have to know that either. It’s a pretty standard system with smart loot anyway; you’ll get a 4 primary and 2 secondary affix item always at level70, tailored to your class usually. Affix combination you get maybe trash but it has no pitfalls.
There could be 7 legendary items and 30 set items for one slot; if your dice roll haven’t surpassed a threshold you’ll get 1 out of 7 Legendary items for that slot in a drop. If it did, you’d get a Set item. At a second dice roll, game determines an Ancient or Primal quality of either. If your initial magic find dice ain’t rolled anything high then you’ll only get rare, white or magical items and roll stops. You can’t exactly tell while playing but your build of choice and power gaps between items create the pitfalls in a very linear progress for characters.
There are reddit threads about gem upgrade odds with mathcraft and possibilities listed. There google docs and excel sheets to tell you that gamble and Kanai’s Cube odds are at different rates with minimum spent required for an item to be achievable. There are google docs that you can look up for affix reroll possibilities to calculate the odds of combinations. How does knowing any of this exactly helps you in the game but curb down your bias? By the demand and increasing power leap, some items are bound to be hard to achieve anyway.
All games usually handle their interaction with pseudo random generators (usually a Mersenne Twister as far as I’m concerned), so what you’re pointing is the obvious. Game can randomly decide to reward you or not; as you don’t have any piece of its source code, you can’t tell anything for sure. You can’t tell how developers decide to deal with repeating odds either. There’s no skewed randomization at all, as every game out there with random ranges handle the numbers the same way.
- Sample size, not example.
- The stated odds, e.g. 60% chance to upgrade a gem, is what’s known as the Expected Value from the Law of Large Numbers.
I’d like to give a thanks to everyone who responded. Mid-way through season 24, I watched a video where Raxxanterax had stated that he was still trying to get the Flying Dragon Affix on his weapon. Given that he plays as much as he does - and he still hadn’t found it - really made it clear to me that it’s just the way it is. After that, I’ve not thought about it since. Again, thanks to everyone for your answers. Best, J.
1 Like
Exactly. At the time I watched that video I had one equipped and I know I found at least one other. And I played less than 100 hours in season 24.