A Random Comment

[Laurell]

I don't have any fancy words, or even an opinion on RNG, to add to the subject but I do have 2 numbers:

[Yukiko]

I begin to understand...
The puzzle forms a picture...


Dear Hiroshi_Minagawa,

Can you please ask the main programmer “K” with which numbers they are calculating and how a "normal" player in 10 years should ever reach that? True, when counting with lets say 1.000.000.000.000.000.000.000.000 trys in total just "3" or "9" times in a row is nearly nothing...

My question now is how realistic those PRNG of yours is, i doubt the mathematic algorythm is made for human lifetime, its made for computers with higher number counts than actually "crafting" or similar small numbers!

Can you please test your generator on a 2 axis "worm" up, down, left, right and tell me in which direction he is "drifting"?

You confirmed to use a PRNG, and that a TRNG is causing too much server stress, but even then: why nearly everything ingame is involving a known unreliably system? At least please add a net for unlucky players!

thanks for listening.


Edit: At this rate i will not toss a coin and bet in gold saucer...

[Nyalia]

Random anecdote: mining Umbral Rocks, having six hits at 12% HQ chance, 95% success rate, I got five HQ in a row (well, six, but the sixth was free since I hit five times). Next time I went? 32% chance, 95% success rate, I got four NQ and had two misses. Random is random.

Also, I did a long-term analysis on my pre-2.3 melds for all eight crafter classes. I had streaks of bad luck, and streaks of good luck. I noticed the bad luck much more than the good luck, and thought I spent way more than the expected value on my melds due to numerous failures. Actual result, looking at the final numbers (actual money spent vs expected value of money spent), I spent significantly (10%) less than I should have due to good luck outweighing the bad. Did I notice the good luck? Nope. That's the bias we're talking about here. Succeed at a meld first try with an 8% chance to succeed? Barely registers as abnormal. Fail 10 times in a row with a 45% chance to succeed? Very noticeable.

[Yoohre_WildRiver]

Despite the fact those were just examples I'll go down the rabbit hole real fast. The time is a known variable yes, but it's only valid for a millisecond. Not enough time to do anything about it. You don't know the range they clamp the results from, you don't know if they random a few times before picking the final number, or any other such techniques to improve variability..

this, i dont think FFXIV programmers would go with something as simple as just the clock to generate the seed. i was talking about this with a member of my FC. i would use more than just one variable to generate the seed i would nest arrays of words to be chosen by random and mix that with the date and then generate the seed maybe use the characters position on the map X Y Z, the eorzea clock server clock client side clock. the more the better.

everytime i hear a FC member tell another FC member doing atma to do th jap method atma farming i laugh a little.. this method is so full of comfirmation bias i tell them to just use my method instead.. the yoohree method is easy.. be patient, accept the reality of pRNG and embrace it with a cold six pack... because even if you dont get the atma at least you had some good beers!

[Nyalia]

I don't have any fancy words, or even an opinion on RNG, to add to the subject but I do have 2 numbers:

Oh, cool! It looks like the max HQ rate for fishing is 20%. I had thought it was only 15%, like the other gathering classes. Maybe it is 15% and you just got really lucky? Your catch rate is 19.4% HQ.

[Galliano]

They've stated it's pure RNG. I know friends who've spent 60+ hours & are only a couple or so in.

RNG has a %. If for example atma drop rate was 5% and the computer rolled 1-100, then landing on 5 of those numbers would equal an atma drop, the other 95 would not. So you could do 5000 FATEs and never have the RNG land on those 5 numbers, at the same time someone else could do 5 FATEs and land on one of the 5 numbers every time.

(Basic example but easy to understand).

[Waliel]

there is a 0% chance that you *know* if a coin is going to be heads or tails next toss, it's true random

If you know all the forces and whatnot that affects the coin toss, you can calculate the result 100% accuracy. For starters, the heavier side is more likely to be facing ground. There are actually machines that will flip the coin with 100% accuracy.

[VitaminMT]

I do really appreciate how the opening post said something to the effect of "humans are want to see patterns where there aren't any, so just chill out.", only to be followed up by armchair mathematicians coming in saying "yeah well nuh uh because see this happened to me and...". Dang, man.

[Kaaylryn]

RNG is in every game to some extent, however, FFXIV seems to have randomness in almost every aspect of the game. I think too much RNG tends to wear on people after time.
Dungeon drops (mostly tank drops from T6-T7 for weeks now, just downed T8 and guess what, tank drops)
Atma drop rates or Alexandrite from fates
Harvesting nodes for HQ, both mining and botany, fishing can be the worst of the 3
Assimilating SB items for materia and praying for something useful
Transmutation by Mutamix
Affixing materia to anything from gear to Novus scroll
The desynthesys system for demimateria
Now gardening with the minions, grade 2 or 3 choco food, being random seeds. Seems the intercrossing does not guarantee you will get what was expected.

So yes, random is random, RNG is RNG, but overplayed it becomes very taxing.

[Giantbane]

RNG is in every game to some extent, however, FFXIV seems to have randomness in almost every aspect of the game. I think too much RNG tends to wear on people after time.
Dungeon drops (mostly tank drops from T6-T7 for weeks now, just downed T8 and guess what, tank drops)
Atma drop rates or Alexandrite from fates
Harvesting nodes for HQ, both mining and botany, fishing can be the worst of the 3
Assimilating SB items for materia and praying for something useful
Transmutation by Mutamix
Affixing materia to anything from gear to Novus scroll
The desynthesys system for demimateria
Now gardening with the minions, grade 2 or 3 choco food, being random seeds. Seems the intercrossing does not guarantee you will get what was expected.

So yes, random is random, RNG is RNG, but overplayed it becomes very taxing.

^
this

the problem isn't necessarily that FFXIV's random number generator doesn't work. It's that they use *pure* RNG too much. Atma & Advanced melding in particular.

Although dungeon/primal drops when their loot table starts to get too big becomes a nuisance also.

I'd say crafting (lots of ways to control your RNG outcome), Novus (getting the alexandrites via myth/quest/hunt OR RNG fate, not the pure RNG melding process) & Hunts (seals with a low chance for a log drop) are areas where they got RNG right. The above 3 (atma, advanced melding, loot tables that are too big) are areas where I believe it could use improvement.