Incorrect finish times in all time high scores for levels


#1

Some of the all time high scores for single player missions are way off and when you watch their solution and view their code they’re not doing anything special.

Is it possible for these to be corrected such as re simulating the level to fix the finish time.
Another solution to this problem for random levels is to simulate the level with a few seperate seeds and post the average time taken.

It feels like there is less incentive to program a decent solution to try for top scores when most of the all time top scores are incorrect.

You don’t have to re simulate every score just the ones that make it onto the high scores, I think it would be a nice improvement to the game and is the only problem I have noticed so far.

Thanks, Sartek.


#2

I don’t normally pay attention to high scores on levels that are not multiplayer but I’ve noticed another issue: In some cases the high scores might come from older versions of the level. For example, in Precision Kicking the top two or so scores on the leader board come from an older version of the level that doesn’t take as long to complete.
What I find more serious is how some players have very high scores in the repeatable levels. Did they cheat or something?


#3

I read that the strange high scores was caused by a bug allowing people to finish a level even after making changes in it, which was discussed here. It seems like the issue is gone but the fishy high scores haven’t been removed yet.

Also I’m a bit curious if people were rewarded gems using this bug, considering the exponential component I suppose they would’ve gotten pretty rich in the process in that case.


#4

I suspected it might have had something to do with that. The linked thread suggests it’s been fixed but I haven’t tested it to be sure.
I think the people that exploited it did gain lots of exp and gems.
Look at this guy: http://codecombat.com/user/5501cb8920c73acd0663df77
(ethangriggs - can be found in theSlayers)
He is level 30 without completing many different levels. He did pass Kithgard Brawl over 50 times though.

Perhaps the leader boards for levels can be reset every time it is updated. The people who care about high scores can just resubmit a solution.


#5

Any plans to solve this problem? @nick


#6

Yeah, I have plans, but not time! (Still busy taking care of the newborn, but I’ll have a bunch more time for hacking starting next month.)


#7

Thanks, Also in the future what is the best place to report bugs or game issues, github?


#8

In general:

  • Search both forum and github
  • For “issues”, if nothing post in forum.
    (I’d say cheaters are an “issue” unless you know how then it is a bug) :smile:
  • For actually bugs in CoCo code, post in github. (“linking” to the forum discussion, if any.)
    (Or both, linking to each other)

If it is a problem you are experiencing include any relevant javascript console errors, or lack thereof.