Just to add some fuel...

[Shibi]

Recent SE Japanese post has a lot of details. They are clearly stating it's not a fault in the random function.

They also suggest that for the Zero-Win bug (there are a few bugs) the winning number is in the database and the plot is actually allocated to the winner, it's just not displayed on the client. So, those winners are locked in, and existing winners were chosen correctly. Zero is client side only and doesn't reflect the lottery results or the state of the RNG.

the winning number is displayed as 0 even though the internal lottery is correctly drawn and a winner exists.

https://pastebin.com/jE8icAJZ

Let's put it's a fault in RNG to rest now.

[PharisHanasaki]

It amazes me to see how everyone suddenly became a programmer in light of this bug.

As other people have said, it's not just a matter of counting from 0 or 1, because there were occurrences of the last ticket winning. This wouldn't be possible if the issue was counting from array[0] instead of array[0+1]. Assuming that was even the way it was coded.

Even so, this would have easily been pinpointed while testing, which I'm sure they did because they had to make sure money was given back to losers.

No one knows for sure what happened, but it has been said that some of these losing bets were placed before the maintenance. Were somehow these tickets lost? Maybe the number of participants was stored separately as an incremental value? All of this is speculation until a response is provided. But you can't say winning tickets were not all valid given the information we have either.

[PharisHanasaki]

Recent SE Japanese post has a lot of details. They are clearly stating it's not a fault in the random function.

They also suggest that for the Zero-Win bug (there are a few bugs) the winning number is in the database and the plot is actually allocated to the winner, it's just not displayed on the client. So, those winners are locked in, and existing winners were chosen correctly. Zero is client side only and doesn't reflect the lottery results or the state of the RNG.


https://pastebin.com/jE8icAJZ

Let's put it's a fault in RNG to rest now.

Yeah it was definitely not a random function bug or array count, this would have been caught easily. But it's nice to finally see some light. Thanks.

[AluminumSiren]

On top of losing auctions as the only bidder, not being able to get your refund, and the completely unfair lack of moderation on bidding (You can bid until you run out of money, although you can only own one house, effectively having the ability to win multiple houses in one week and cut however many other people out of a week's worth of waiting. There is no 'runner up' so those lots sit empty until the next bidding cycle, a freaking week later.), there's something I haven't seen anyone mention yet.

Wait what? I was only able to bid on one house during the lotto period. Am I missing something?

[rance-sama]

On top of losing auctions as the only bidder, not being able to get your refund, and the completely unfair lack of moderation on bidding (You can bid until you run out of money, although you can only own one house, effectively having the ability to win multiple houses in one week and cut however many other people out of a week's worth of waiting. There is no 'runner up' so those lots sit empty until the next bidding cycle, a freaking week later.), there's something I haven't seen anyone mention yet.

Every single person who won this week's lottery was NOT the actual winner. How to explain in layman's terms...

When you generate a random number in code, it looks something like random(100), meaning a random number between 0 - 99, NOT 1 - 100. Computers start at zero by default in almost every language. The problem we ran into is a rookie mistake that a first year student wouldn't even make after week 3. To compensate, you return the number generated PLUS 1. Somebody forgot the plus 1. It's really that simple. Yes, maybe other mistakes were made in other parts of the system as well, but there wouldn't be so many number 0 winners had that very simple part of the code been correct. Even if the number was written to a database before the award code read it, I highly doubt anyone would've accidentally subtracted a 1 before writing or after reading the database. The fact remains we got zeros and the numbers started at 1. Anyway...

The meaning behind this is that every single person who won this week was not the actual winner, it was the next person in line.

Let that sink in, and know they will do NOTHING about it. Every single one of those houses should be refunded and put back up for sale.

Have a nice day.

You could code an array that includes numbers from 1 to 100, then use a random number picker that grabs one from said array without having to pick one and summing plus one.
This is Japanese programming we're talking about, not Western pseudo-coders.

[PharisHanasaki]

Wait what? I was only able to bid on one house during the lotto period. Am I missing something?

FCs could have multiple members bid on one, or many plots. Or you could bid with alts.

[Kacho_Nacho]

You could code an array that includes numbers from 1 to 100, then use a random number picker that grabs one from said array without having to pick one and summing plus one.
This is Japanese programming we're talking about, not Western pseudo-coders.

Was that png necessary to make your point?