A Random Comment

[Rktsci]

The only thing I'll say about this is the HQ% on crafting is most certainly not linear. If I jump on CUL to make buttons in a blanket with 347 craftsmanship and 330 control and use basic touch, I'll accrue X quality per usage and my % increase jumps from say 3 to 15. But subsequent uses of the same ability (and quality) increase that % chance in "chunks," which I've never liked. If other systems in the game do something similar, it's a small wonder people are skeptical of the RNG gods. How do you go from 3% HQ to 15 and then to 45 on the SAME number? You don't, because that wouldn't make sense. In a way, the system does absolutely favor failing you.

The crafting quality system is non-linear. That has nothing to do with the random number generator. It's set up to reward multiple quality increase attempts and hence building up CP on your character and understanding how cross-job skills can interact to allow more quality increase attempts.

[Umero]

they need a better system that beat RNG system for loot like Atma or Alexandrite from FATE

because using True RNG system doesn't 100% provide progression for some(a lot of) playersk.

Suggestion: Increase the drop rate by 1% every failure to obtain the loot so if unlucky people keep working on it, they'll eventually get their reward.

The problem with the current system is that the true RNG system is not a good practice for games that supposed to make you feel rewarding as you progress.

[white_mage]

Got it? No? That’s okay.

I know right? If there's a problem, they will fix it. People need to stop asking for a change in the ATMA drop rates. There's a reason its an arduous process. The reason why there's so much negativity surrounding it is because you don't hear all the people with good luck complaining on every public forum. It's mainly the bad experiences that are the most vocal.

[Sylkis]

For those asking for increase in success after failing:
Basically you are asking the system to be less random or not random at all so that everyone can win. If it is truly random, then expect some people to instantly win and some people to almost never win.
If you want a system that increases your success rate after failing, just ask for a system so that everyone can win in a certain amount of time. No need to talk about RNG.

To that RNG seed discussion:
System time is stored as the number of seconds since 1 Jan 1970, so you never get the same number. But I doubt this is used as the seed anyway, because its pretty simple to do.
Its possible to attach some sensor to a computer to read in randomness from the atmosphere. But I think thats unnecessary for a game like this.
Anyway, this place isn't the place to discuss this because its not only about FFXIV. Anyone looking to argue about that here is just looking to win an argument.

[Kaonis]

Aaaaand the problem the devs fail to miss is it's not just a bias we're believing we are seeing, that "Oh crap I failed on a 90% 3 times in a row!" isn't a one time deal, it happens...a lot....far too often fora real RNG. To put it in perspective, to fail a 90% chance action 3 times in a row is a 1/1000 odd. When this happens once it sucks, when it happens twice, you have bad luck, when it happens consistently there's a clear bias in the program. I can't even explain how many times on BTN I failed at a node 3-4 times with an 80%. There seems to be a very clear pattern where their algorithm likes to keep failures going in a row.

[Cyrus-Wallace]

I'm one of these people who failed 5 attempts in a row with 90% of success chance or got 5 HQ gathering items in a row with 15% of success chance. Also, this happened today:



I know these things are fully random, when you always face the same circunstances, like the same drops in Coil or the examples I mentioned above, it makes you wonder.

[IndigoHawk]

For those asking for increase in success after failing:
Basically you are asking the system to be less random or not random at all so that everyone can win. If it is truly random, then expect some people to instantly win and some people to almost never win.
If you want a system that increases your success rate after failing, just ask for a system so that everyone can win in a certain amount of time. No need to talk about RNG.

You talk about that like it's a bad thing. Being unlucky isn't fun.

If you're running a lottery, the results should be random, because you can't afford everyone to win. But if you're paying $15/mo as a gamer, why are you paying for unfun RNG when you could instead pay $15/mo for fun pseudo RNG?

There's no problem with helping unlucky people, and there's no reason to take your suggestion and get rid of random to turn everything into tokens.
Instead use pseudo random to keep it a bit random but still fun.

[IndigoHawk]

I always found this comparison funny, as if it has nothing to do with there being millions of cars/drivers and only thousands of airplanes/pilots. Of course the chances of being in a car accident are higher xD

You find this funny because you don't understand that safety is determined by rate of death/injury and not by quantity.

There are more car trips than plane trips, so you are not surprised that more people die from cars than planes.
More trips is not what makes cars more dangerous. What makes cars more dangerous is the percentage of car trips that result in deaths, compared to the percentage of air travel that results in deaths.

The percentage of death in air travel is far lower than cars. So, if someone said, "A plane crashed yesterday. Is it safer to take a plane or a car?" ... the answer is ... the plane, because they still have a lower percentage of fatal trips.

Cars are more dangerous to you, both because you use them more frequently AND because they have a higher injury rate.

[therpgfanatic]

There is nearly always disconnect between programmers and players over issues of randomness.

*From the player's perspective, the random is not equal because they spent 14 hours trying to get an Atma to drop, and someone else got it on the first FATE in that map.

*From the programmer's perspective it is a consistent random formula because when you look at ALL PLAYERS drop rates, there is a clear pattern and enough drops are happening per hour (or whatever criteria they are using as the barometer) that the devs feel the system is balanced.


In my limited experience working in a game studio, programmers tend to be the absolute worst people at designing any kind of content involving randomness because they look at it purely from a mathematical perspective, not from the individual user experience level.

This is because of the unique way engineers think about problems; they tend to take a top-down approach rather than a bottom-up approach. They tend to start with "this is how our numbers will work" rather than "this is how the user experience should be".

Based on that forum post explanation, I'm pretty sure that designer is never going to understand the problem because he's not starting with the user experience; he's just looking at the numbers. He was tasked with making a random drop rate for the Atma's and that's what he did.

Someone should have told him it needed to be done in such a way that it won't take 36 hours of grinding to get a single orb to drop.

I did the Atma farming already, almost done with my Animus upgrade. But I will never, ever do it again for my alt Jobs. This content was designed to demand too much of a time investment for too little of a stat boost over the weapons that drop in raids.

[Kyros]

The issue here isn't that the RNG isn't random enough, it is that it is.

The system implemented here is complete a utter RNG, without any algorithms in place to remove outliers. Outliers is what make people feel bad, because it feels (And is) unfair. That one guy that took 60+ hours to get an Atma is going to feel like he got treated unfairly no matter how good his understanding of RNG is.

This is why games nowadays implement smart RNG. A common example: League of Legends. In league they implemented a system so that critical hits wouldn't behave outside the norm (Basically eliminate outliers). If you had a 5% chance, you wouldn't for example get 3 critical hits in a row, even if it is completely possible mathematically. On the same token, if you had 90% crit, you wouldn't see streaks of 15 shots without a crit (Which again, is completely possible on a true RNG).

We, the players, want smart RNG. Because it is unfair both to those who are hindered and those who are favored by the RNG.