Re-roll bug: I can't roll a 23

25

Guys, I'm considering quitting for good.

I spent millions of credits + crafting materials just now in order to try and get +23% ranged damage on my Gravgun.

The range is FIFTEEN TO TWENTY-THREE.

That's FIFTEEN.

TO.

TWENTY-THREE.

THAT'S:

15

16

17

18

19

20

21

22

23

That's, add 'em up: 1, 2, 3, 4, 5, 6, 7, 8, 9. NINE possible values.

9.

NINE.

Not 1 billion.

Not 5 trillion.

But NINE (9) possibilities.

How is it that after 100-200 tries, I didn't get 23?

Is there some sadistic bit of code your programmers add to ensure that the max value pops up only after you've been playing 2000 hours and have tried re-rolling 1000 times?

Guys, I like your game, but I'm going to be brutally honest now.

You need to get new programmers (and a new QA team).

Man, this game pisses me off at times.

This post was edited 4 years 350 days ago by treecargarage
Store Page
Share
25 comments
Re-roll bug: I can't roll a 23
Your Thoughts? Please login to place your opinion. Not a member yet? Register here and now!
4 years 188 days ago
Hi Neocore, any updates on this bug? I thought it may have been fixed this last patch, but I tried a few times, ended up rolling a 22, and figure it was still there.


Was it fixed or not? If it was, I will try and re-roll some 23s.

4 years 352 days ago
+1
Posted by Marcopolocs 4 years 352 days ago
It is a bug, currently the 23 cannot roll on the enchant. We will fix it but as it is not a priority on our end it most likely will come later. @TREECARGARAGE‍ 
Ok, just let me know when it's done, and I'll start re-rolling again to perfect the last of my items.
4 years 352 days ago
+1
treecargarage
It is a bug, currently the 23 cannot roll on the enchant. We will fix it but as it is not a priority on our end it most likely will come later. @TREECARGARAGE‍ 
4 years 354 days ago
Posted by treecargarage 4 years 356 days ago
Marco, are you able to confirm whether it may not be possible to roll a 23 for enchants that contain the range 15-23 (are there others that end in 23?)?
Bump...

Any word on whether the X-23 and X-38  enchants are bugged and cannot ever roll a 23 and 38, respectively?

Because these are the only two that refuse to roll a perfect value.

And I've tried a LOT for the 23...the 38, not as much, but it's still reached the point of "statistical virtual impossibility".

4 years 356 days ago
Marco, are you able to confirm whether it may not be possible to roll a 23 for enchants that contain the range 15-23 (are there others that end in 23?)?
4 years 357 days ago
+1
Update 3: I have a hypothesis...I believe that the "23" bug may literally be the "23" bug, in the sense that ranges that end with 23 cannot roll a 23. At least it seems to be that way for ME. The reason I say this is that I've been perfecting my rolls, and I just hit another roadblock. The roadblock being the enchant +% damage to X (I tried with X = Chaos marines and X = regular enemies). For both, I could not roll a 23. I did get 22s, though.
4 years 357 days ago
+1

Update 2: I re-rolled some more perfects just now. Your RNG seems to work, overall. I'm 99.9%+ sure the 23 issue is a bug now.

This comment was edited 4 years 357 days ago by treecargarage
4 years 358 days ago
Update: I just did some re-rolling on my other items. I got several perfect rolls. This means that what my 23 problem may actually be a bug (I highly suspect that it is). The bigger problem is, is it wide-spread? How many other items is it affecting? I guess the only way to get a decent idea of whether the particular item/enchant is bugged is if you cannot get a perfect roll after an obscene amount of tries.
4 years 358 days ago
Posted by Kudja 4 years 358 days ago

One person wrote about the problem, but everyone is trying to fill in the necessary parameters. Will you compensate everyone for the money?

I would be surprised if everyone is trying to get a unicorn number that doesn't seem possible to attain.

I got frustrated relatively quickly, made a post, tried again a day or two, then gave up.

IF there is a person in a similar position that has been trying to get a "23" for months, then I both applaud and pity this person.

This comment was edited 4 years 358 days ago by treecargarage
4 years 358 days ago

One person wrote about the problem, but everyone is trying to fill in the necessary parameters. Will you compensate everyone for the money?

4 years 358 days ago
-2

Just went through another 1.5 million credits.

Still nothing.

I think I'm going to uninstall.

There's no point in playing anymore, since the only reason I'm still playing is to optimize my build.

I'll maybe come back if you guys fix your most serious bugs (RNG/re-rolling being one of the huge ones).

This comment was edited 4 years 358 days ago by treecargarage
4 years 358 days ago
+1
Posted by Marcopolocs 4 years 359 days ago
We will check the code why the 23 could not roll onto the enchant. Sorry for the trouble! I compensated your credit loss.

Hi Marco, I just went through 4.2 million of your 5 million credits.

I still haven't gotten 23.

I've probably re-rolled 250 times by now...

Dunno what else to say.

Also, I discovered a bug. I was re-rolling, and after a while, it no longer showed the re-roll counter decreasing. I kept clicking, but it kept showing 2 re-rolls left. I figured it may be a bug, so I logged out and logged back in. I saw that my credit count had gone down, and the re-roll counter was at 0. So I guess it was re-rolling, but the information wasn't being updated. Anyway, getting tired of this kind of stuff, but thought you guys should know.

This comment was edited 4 years 358 days ago by treecargarage
4 years 359 days ago
Posted by LoneKharnivore 4 years 359 days ago
I think we disagree there. Getting something perfect or terrible should be less likely than getting something average. Given that you'll probably not notice the difference between +20 and +23 in-game it's not that important in effect.


In addition, as most values include a decimal place, nine possibilities would become ninety. I'm not sure about your specific example though.

It's not a decimal, so it's just 9.

As for the perfection, if it doesn't make a big diff, then a perfect roll shouldn't be treated as special, no?

RNG is RNG. If they're making it artificially difficult to get perfect rolls, then I think you'd agree they should let us know.

As for the decimal ranges...I'm probably going to have nightmares trying to perfect those enchants.

This comment was edited 4 years 359 days ago by treecargarage
4 years 359 days ago
Posted by treecargarage 4 years 359 days ago
Yes, I was considering this as well. It would be a diabolical system, but I would have some satisfaction knowing that this is the reason for my frustration.
I think we disagree there. Getting something perfect or terrible should be less likely than getting something average. Given that you'll probably not notice the difference between +20 and +23 in-game it's not that important in effect.


In addition, as most values include a decimal place, nine possibilities would become ninety. I'm not sure about your specific example though.

4 years 359 days ago
+1
Posted by Marcopolocs 4 years 359 days ago
We will check the code why the 23 could not roll onto the enchant. Sorry for the trouble! I compensated your credit loss.
Thanks Marco. Keep in mind that I'm not upset with you, but with your programmers.

Thanks for the compensation, but realistically, what will happens is I will go through all the credits and not roll a 23.

Then I will contemplate the meaning of life.

4 years 359 days ago
+2
Posted by LoneKharnivore 4 years 359 days ago
I suspect it's a bell curve, personally. I roll mid-table more often than high or low.


Getting whiny and throwing insults about is not the solution, however.

Yes, I was considering this as well. It would be a diabolical system, but I would have some satisfaction knowing that this is the reason for my frustration.
4 years 359 days ago
Posted by Dr BAMF PhD MMT 4 years 359 days ago
I agree that the roll values are frustrating and the "random" stuff seems purpose built to piss you off sometimes, I totally agree this is a big problem. 


But it isnt a Martyr problem, its an every-game-ever-made problem. Ever played destiny? Or eso? Or basically any looter ever? The reason for the endemic nature of this problem is that while the more or less standard system that all these games use in one form or another is referred to as RNG, true random-ness is an organic concept that isnt possible in a binary system.

A computer cant simply summon an unpredictably chosen number out of nowhere, it needs to get it from somewhere. To solve this problem back in the early days of computing,  someone made a model whereby, when asked to produce a value from nothing, the cpu would sample the value of the picoseconds internal clock it uses to order and syncronize its operations, truncate that to within the requested range and provide it as the requested value. This specific system is what we know as RNG.

It was simple, easy, and not even close to actually random. But it worked well enough with the scope of programs at the time, and it used almost no proccessing time unlike virtually every other possibility for generating not-actually-random numbers like the fibonnaci sequence and Pi, so it ended up becoming hardwired into every computer on earth.

The problem with using a verbose expression of time to present the illusion of randomness however, is that picosecond decimals are a pattern, returning to zero at the end of a second. Put that into a gargantuan system of patterns, in a world based around patterns, being lived by people who generate patterns, sooner or later those patterns will begin to align and find an equilibrium that destroys any attempt at randomness.

For example, human neuromuscular reaction time means you can only press a button a few times a second, and will tend to attempt to press it in regular intervals. Factor in the latency of your machine, the internet, their servers and whatever controller you use, and what you get is a complex system of interlocking exclusions and probabilities that renders certain ranges of values statistically much more or much less probable to recieve. What ranges those are depend on an enormous set of variables that are impossible to calculate while they remain in effect, but the net result is a system with significant and noticable discriminations in probability at any given time.

This is a big problem to video games of all kinds, but none moreso than online looter games, which depend on RNG for fair generation of loot, and have much more complex and inscrutable systems than most. In short, a developer setting a flat 20% drop rate for something, cant realistically expect to actually get it 20 times out of a hundred once the system it runs on gets large enough. This is a mathematical limitation of how computers are made, that isnt changing anytime soon, and would require a better understanding of both maths and chaos theory than i have to fix.

The solution is simple. You keep track of what values have been randomly generated, and you put pressure on the ones that aren't being selected.

If the 23 hasn't popped up after 8 re-rolls, pressure is applied.

If it hasn't popped up after a 9th re-roll, more pressure is applied.

More and more pressure is applied until a threshold is reached, and the baloon pops.

The 23 will then be forced to show up, no matter what.

4 years 359 days ago
+1
I suspect it's a bell curve, personally. I roll mid-table more often than high or low.


Getting whiny and throwing insults about is not the solution, however.

4 years 359 days ago
We will check the code why the 23 could not roll onto the enchant. Sorry for the trouble! I compensated your credit loss.
4 years 359 days ago
+2
I agree that the roll values are frustrating and the "random" stuff seems purpose built to piss you off sometimes, I totally agree this is a big problem. 


But it isnt a Martyr problem, its an every-game-ever-made problem. Ever played destiny? Or eso? Or basically any looter ever? The reason for the endemic nature of this problem is that while the more or less standard system that all these games use in one form or another is referred to as RNG, true random-ness is an organic concept that isnt possible in a binary system.

A computer cant simply summon an unpredictably chosen number out of nowhere, it needs to get it from somewhere. To solve this problem back in the early days of computing,  someone made a model whereby, when asked to produce a value from nothing, the cpu would sample the value of the picoseconds internal clock it uses to order and syncronize its operations, truncate that to within the requested range and provide it as the requested value. This specific system is what we know as RNG.

It was simple, easy, and not even close to actually random. But it worked well enough with the scope of programs at the time, and it used almost no proccessing time unlike virtually every other possibility for generating not-actually-random numbers like the fibonnaci sequence and Pi, so it ended up becoming hardwired into every computer on earth.

The problem with using a verbose expression of time to present the illusion of randomness however, is that picosecond decimals are a pattern, returning to zero at the end of a second. Put that into a gargantuan system of patterns, in a world based around patterns, being lived by people who generate patterns, sooner or later those patterns will begin to align and find an equilibrium that destroys any attempt at randomness.

For example, human neuromuscular reaction time means you can only press a button a few times a second, and will tend to attempt to press it in regular intervals. Factor in the latency of your machine, the internet, their servers and whatever controller you use, and what you get is a complex system of interlocking exclusions and probabilities that renders certain ranges of values statistically much more or much less probable to recieve. What ranges those are depend on an enormous set of variables that are impossible to calculate while they remain in effect, but the net result is a system with significant and noticable discriminations in probability at any given time.

This is a big problem to video games of all kinds, but none moreso than online looter games, which depend on RNG for fair generation of loot, and have much more complex and inscrutable systems than most. In short, a developer setting a flat 20% drop rate for something, cant realistically expect to actually get it 20 times out of a hundred once the system it runs on gets large enough. This is a mathematical limitation of how computers are made, that isnt changing anytime soon, and would require a better understanding of both maths and chaos theory than i have to fix.

4 years 359 days ago
-1

lolol

Just did a couple of missions to try and re-roll some more...

20 more re-rolls, 20 more fails.

Look away everyone! Nothing to see here! Not a bug! Not shoddy programming! It's just bad luck!


I would love to be as optimistic as you guys.

You should think about becoming slaves. You would be utterly magnificent at it.

0 standards.

Nothing could phase you.

Getting whipped? No problem, just bad luck. Things could be worse.

No food? No problem, just bad luck. Things could be worse.

RNG gives you a perfect roll once every 1000 re-rolls? No problem, just bad luck. Things could be worse. It could take 2000 re-rolls!

This comment was edited 4 years 359 days ago by treecargarage
4 years 359 days ago
-1
Posted by Rigg 4 years 359 days ago

this is standard random function. values may be repeated. you're just unlucky.

...will you be saying that a year from now when I still haven't gotten 23?


I re-rolled around 150 times, let's say.

I SHOULD be getting the 23 every 9 re-rolls.

150/9 = 16.66 repeating, let's say it = ~17

That means that I've been having "bad luck" for 17 iterations???????

Total BS.


As I said, I will keep updating this post.

I HOPE I get 23, and I HOPE that it's my "bad luck", but I wouldn't be surprised if this is yet another bug, and it's not even POSSIBLE for me to roll an Fin 23.

We will see...


I guess when I get a 23 (IF I get a 23) after I end up re-rolling 600 times and spending 1 billion credits, you guys can post "See?? It IS possible! The code is good!"


lol...

This comment was edited 4 years 359 days ago by treecargarage
4 years 359 days ago
-1
Posted by Rigg 4 years 359 days ago

this is standard random function. values may be repeated. you're just unlucky.

Yup. I can usually get max value, or at least close enough, after maybe half a dozen rolls. Sometimes RNGesus just frowns on you.
4 years 359 days ago
+1

this is standard random function. values may be repeated. you're just unlucky.

4 years 359 days ago
-1

And to put things into perspective, here is what happened tonight:

1) I played through an entire Ebony Void Crusade. Took me an hour+.

2) Got Intels for credits. Ended up with around 10 million. Rest of the items were useless, and so I sold/salvaged them.

3) I started re-rolling my items.

4) I ended up wasting most of my credits on this 15-23% re-roll.


So, what you did, Neocore, is you essentially TOTALLY WASTED 1-2 HOURS OF MY TIME.

What's the point of even playing the game and trying to optimize your gear if your RNG is GARBAGE and does not work?


This comment was edited 4 years 359 days ago by treecargarage