BuzzerBeater Forums

Bugs, bugs, bugs > Identical games

Identical games

Set priority
Show messages by
This Post:
00
179505.15 in reply to 179505.4
Date: 4/14/2011 9:27:33 PM
Overall Posts Rated:
587587
I'm amused by the fact that although the stats are *exactly* the same, the PP100 for the Φοινικας Καλλιπολης Center was better in 324 compared to 303.

Indeed. I think what we are seeing here is the same teams, same lineups, and same tactics, with all other differences (in this case only player experience and team enthusiasm, since the games were played with identical gameshape and skills) so small that the games happen to run the same given the same random seed. There is a tiny difference however, which I guess could have changed the outcome at some point. This difference is shown by the rounding of the PP100 figure. Another explanation for it would IMO be a bug in the rounding, which is highly unlikely.

I seem to remember the random seed issue was fixed when people were seeing identical games the last time. Is this not the case then? I think this should be a pretty high priority.

This Post:
00
179505.16 in reply to 179505.7
Date: 4/14/2011 9:28:47 PM
Overall Posts Rated:
587587
Hard to imagine what kind of software bugs could be related to these "identical games". (earlier commented by BB)

Using the same random seed is the easiest explanation.

This Post:
00
179505.17 in reply to 179505.11
Date: 4/14/2011 9:33:38 PM
Overall Posts Rated:
587587
this is a big fail for this game and i don't think the problem is the lack of random..
one of those games in the link, (32629852) and (31835798), have been played three weeks after the other one..

Well, that's pretty interesting. Both teams are however bots. Is there a difference in bot team processing?

This Post:
00
179505.18 in reply to 179505.16
Date: 4/15/2011 1:13:02 AM
Overall Posts Rated:
986986
Yes and then it is a bug in game engine design, which is even worse than just a bug in one line of code, i think.

This Post:
00
179505.19 in reply to 179505.13
Date: 4/15/2011 2:49:20 AM
Młoty Stargard
PLK
Overall Posts Rated:
840840
Talking about numbers and pseudo-random does not make sense.....
My calculator generates pseudo-random numbers, and they are never the same.
The engine numbers are generated BB, for example, 0.87 .... 0.65 .... 0.43 .... 0.13 ... 0.99
In the next game the same teams will be generated the same counts in the same order:)
Do you know what are the chances?
It's impossible

This Post:
33
179505.20 in reply to 179505.19
Date: 4/15/2011 3:29:58 AM
Overall Posts Rated:
3333
Once you know the identity of the BB RNG, it'll all start to make sense...

http://www.random.org/analysis/dilbert.jpg

This Post:
00
179505.21 in reply to 179505.20
Date: 4/15/2011 8:43:04 AM
TrenseRI
II.1
Overall Posts Rated:
36023602
Second Team:
ChiLeaders
I love the cartoon

Anyway, this problem is our top priority. Currently I don't know why it happens, but to presume it happens because the second game was generated exactly the same is a big stretch. The problem is likely much simpler...

This Post:
00
179505.22 in reply to 179505.18
Date: 4/16/2011 12:16:19 PM
Overall Posts Rated:
587587
If the problem really is the random seed, it is not serious in the sense that you would need to fix a lot of the game engine code. Nothing here suggests the game engine design is flawed.