939ers: Try Clicking the Center Box FIRST For a Better Payoff in Prize Box Bonuts.

"KrustyBrand;1552397" wrote:

"Juicer-19;1552376" wrote:
So...it is not random. The bonuts rotation is a specific sequence. It is not RNG.

I remember, vaguely, back when there was a glitch and you could spam bonuts almost infinitely, that the simple sequence was figured out.

It always just shifts 1 to the left. So if you got 3 donuts in box 2, next time the 3 donuts will be in box 1, and then the next time it will be in box 3.

The tricky part is remembering the last box you clicked.

Nope. Don’t believe this. If this glitch ever existed as you describe, it doesn’t seem to exist now. And if it does, I wanna know what the putative sequence is, so I can test it. Second, your description of the sequence conflicts with the descriptions of others who describe “patterns” in which max donuts appear at a fixed location over successive days rather than alternating/shifting. Third, such a pattern as you describe begs the question — why would a programmer go through the trouble of storing the previous donut locations rather than just program them anew using some pseudo-random process?

I’m more than willing to admit I’m wrong in the face of evidence to contrary, but so far I haven’t seen it.

Edit: My original post was from a different device on a different account...I am the same guy, just posted this under a different account apparently.

There was a glitch a long time ago that you could force the bonus donut selection to keep reappearing.

Once it was working, you would select your box...if you got 3 donuts, you would accept and the menu to select would reappear. The process was always to select the box to the left of the previous selection. You would get 3 every time. You would continue that as long as you wanted or the glitch stopped working.

It is the process that I go by. I don't pay any attention to this anymore, so it could have been changed with an update. Who knows? This was a thing before and the way I still select my donuts. I got this information on this very forum, so I bet with some digging, you could find the archived thread that explained it all.

The idea that it would be more efficient using a random number generator is false. Unless you use some sort of data structure to reduce the time complexity, generating random and different numbers just to fill the choices has a time complexity of O(infinite). This is because of the possibility that the same number is generated every time, leaving you in an infinite loop. You would need to generate some sort of data structure (dynamic array, perhaps) which would allow you to fill the arrays with different numbers and then randomly select an index to match with whatever array you are iterating through. You would then delete that pointer in the array so that you guarantee that the next selection is different.

Say you fill a dynamic array donuts so that the data in the array is equal to the index + 1.

Then you iterate through box and randomly generate a number to match to the donuts array.

set a variable count = 3

for(int n=0; n<3; n++)
box = donuts
delete donuts

So the process of making the time complexity usable requires creating arrays. Why trouble the programmer with generating a random number, creating multiple arrays in the function, just to make the selection random?

You could simply store 1 array Box and store the data as 1 2 3 or 3 1 2 or whatever. And every time the prompt for bonus donuts has been completed, you shift the values in the array to the left. Less code, better time complexity, at the cost of device storage. Not server storage. So it costs EA $0 to store the data and the game itself has less functions and loops to iterate through.

I could also be wrong, I don't look at this game's code at all. And I don't pay attention to any patterns for donuts. This was a thing a long time ago. They may have changed the function to be random in some update. I am sure if some other coder needed to fix a glitch, they probably rewrote the entire function...and it is possible that it was given seeded randomness.

But I promise you, selecting 1 to left used to work.

"juicer420;1552532" wrote:


The idea that it would be more efficient using a random number generator is false. Unless you use some sort of data structure to reduce the time complexity, generating random and different numbers just to fill the choices has a time complexity of O(infinite). This is because of the possibility that the same number is generated every time, leaving you in an infinite loop. You would need to generate some sort of data structure (dynamic array, perhaps) which would allow you to fill the arrays with different numbers and then randomly select an index to match with whatever array you are iterating through. You would then delete that pointer in the array so that you guarantee that the next selection is different.

That's only if you populate the three boxes in advance. If you trigger the random generation on the click of the user it becomes trivial. First click rand(3), second click rand(2), third click remaining option.

Another, easier alternative, is to populate all the possibilities into arrays (pre-programmed) and then just randomly select one of the arrays. THIS is the most likely avenue they chose, considering that that's exactly how the vegas games worked, as well as the scratch tickets. There's a pre-set number of combinations and you randomly get one of the combinations.
So for the bonuts:

Possibilities are:
1,2,3
1,3,2
2,1,3
2,3,1
3,1,2
3,2,1

Then just a rand(6) and voila-- done. Most efficient solution.

"alapalme;1552584" wrote:

"juicer420;1552532" wrote:


The idea that it would be more efficient using a random number generator is false. Unless you use some sort of data structure to reduce the time complexity, generating random and different numbers just to fill the choices has a time complexity of O(infinite). This is because of the possibility that the same number is generated every time, leaving you in an infinite loop. You would need to generate some sort of data structure (dynamic array, perhaps) which would allow you to fill the arrays with different numbers and then randomly select an index to match with whatever array you are iterating through. You would then delete that pointer in the array so that you guarantee that the next selection is different.

That's only if you populate the three boxes in advance. If you trigger the random generation on the click of the user it becomes trivial. First click rand(3), second click rand(2), third click remaining option.

Another, easier alternative, is to populate all the possibilities into arrays (pre-programmed) and then just randomly select one of the arrays. THIS is the most likely avenue they chose, considering that that's exactly how the vegas games worked, as well as the scratch tickets. There's a pre-set number of combinations and you randomly get one of the combinations.
So for the bonuts:

Possibilities are:
1,2,3
1,3,2
2,1,3
2,3,1
3,1,2
3,2,1

Then just a rand(6) and voila-- done. Most efficient solution

Yeah, I was going to say something about coding along these same lines.

On the other hand, @juicer420 ’s comments on the nature of the glitch do ring a bell — i.e., I do recall discussion of it from quite some time back. But, iirc, it was patched. And current experience with the game certainly shows that any pattern such as the one described is no longer in effect.

That theory is disproved easily. If it's rotating through the 6 possibilities, you would never get the same amount of donuts in the same box more than twice in a row. I've had random occasions of getting 3 donuts in the same box 3, 4+ times in a row. Likewise for 2 and 1. I've been collecting donut boxes daily, now at a rate of 5 to 6 rounds every 4 hours-- and I have yet to find a reliable pattern. If it's not random it's terribly close to it!

It's random.

It's random. But I like the programming talk.

Lost in this discussion is the fact that I am sure most "939ers" are like me and have a millions and millions in cash. Even if I only collect from buildings and I don't get the donuts until the 3rd box every time, it's mathematically impossible for me to run out of money. I don't need a pattern, because I'm game rich.

"Juicer-19;1552617" wrote:

"KrustyBrand;1552590" wrote:

"alapalme;1552584" wrote:

"juicer420;1552532" wrote:


The idea that it would be more efficient using a random number generator is false. Unless you use some sort of data structure to reduce the time complexity, generating random and different numbers just to fill the choices has a time complexity of O(infinite). This is because of the possibility that the same number is generated every time, leaving you in an infinite loop. You would need to generate some sort of data structure (dynamic array, perhaps) which would allow you to fill the arrays with different numbers and then randomly select an index to match with whatever array you are iterating through. You would then delete that pointer in the array so that you guarantee that the next selection is different.

That's only if you populate the three boxes in advance. If you trigger the random generation on the click of the user it becomes trivial. First click rand(3), second click rand(2), third click remaining option.

Another, easier alternative, is to populate all the possibilities into arrays (pre-programmed) and then just randomly select one of the arrays. THIS is the most likely avenue they chose, considering that that's exactly how the vegas games worked, as well as the scratch tickets. There's a pre-set number of combinations and you randomly get one of the combinations.
So for the bonuts:

Possibilities are:
1,2,3
1,3,2
2,1,3
2,3,1
3,1,2
3,2,1

Then just a rand(6) and voila-- done. Most efficient solution

Yeah, I was going to say something about coding along these same lines.

On the other hand, @juicer420 ’s comments on the nature of the glitch do ring a bell — i.e., I do recall discussion of it from quite some time back. But, iirc, it was patched. And current experience with the game certainly shows that any pattern such as the one described is no longer in effect.

The glitch was patched for sure. I guess my point was that from that glitch a sequence was figured out, iirc.

For the longest time...I was grinding for donuts and I used that method to save some money because it was flawless. I had a sticky memo on my desktop that always told me my next box. I used to play primarily through bluestacks.

Now that I don't grind for donuts and I play primarily on my phone...I don't pay as much attention. But I do try to remember my last box. I feel like it stills works.

As the poster just above said, it is likely that they initialize the data sets and randomly pick one of those.

It could also be that they randomly ordered those data sets and stored them in an array, which they just sequence through instead of randomly selecting. I could have had a sequence that is stored such as this...

1, 2, 3
1, 3, 2
3, 1, 2
2, 1, 3
2, 3, 1
3, 2, 1

This could be tested by anyone. Just test every 6th bonus, if the boxes are the same, it is just going through a specific sequence that is unique to the user. 1st try, 7th try, 13th try, etc.

Has anyone even tried that? Collected any sort of data on this?

It is just as difficult to disprove a theory without evidence as it is to prove a theory without evidence.

I recall in the past that someone actually saw the set of possibilities in the code and, at the time, it was conjectured that this was the ordered sequence that was used. Turned out that it was most likely just a look-up-table, from which the box contents were selected (pseudo-)randomly because nobody could reliably get the pattern to repeat. Yes, there were a few times where it happened to look like it, but it would always deviate.

There was quite a bit of searching for patterns done when the endless-Bob and forever-fan glitches were happening. I had hundreds of results written down with no discernible pattern.