Calcudoku puzzle forum
https://www.calcudoku.org/forum/

Timed Puzzle Bonus
https://www.calcudoku.org/forum/viewtopic.php?f=17&t=45
Page 10 of 13

Author:  pnm  [ Sun Jan 29, 2012 1:17 pm ]
Post subject:  Re: Timed Puzzle Bonus

sneaklyfox wrote:
I just like to think of the loopholes for fun.

And I have to think about many loopholes to ensure "trust in the system"..
(there's no value in a timed rankings list if it's too easy to cheat)

Patrick

Author:  clm  [ Sun Jan 29, 2012 1:35 pm ]
Post subject:  Re: Timed Puzzle Bonus

pnm wrote:
clm wrote:
... Sure, you are a very good and elegant player and, as you are very smart and good observer, probably you have already noted what, in a pair of cases (at least apparently and coming from long time ago), looks like "twin players" [smile]. Who knows ...

There are ways for me to (automatically) detect this, of course.
...
Patrick


Ok. To have various mailboxes is very common (and easy) so the same player could have various usernames. From this point I think there are some options:

-- To permit "any" (or a limited) number of users to the same player (no restrictions, either for the regular or for the timed puzzles). In case of regular puzzles it represents some additional work to the players to enter twice all puzzles everyday but some players could find this valuable for having more representation of a determined country, just for having different names, of for other reasons ...
-- To permit "any" (or a limited) number of users for regular but only one for timed, for instance.
-- To restrict that number to only one for regular and timed, for instance (perhaps too hard restriction, in this case if someone wanted to modify the username and keep the points should ask you via e-mail and the change published in some way...).
...

Some sites stablish restrictions, i.e., "www.playchess.com" suspends a player if they detect (and they have the means of detecting that) the player is using computers to play against other players since there are different "coffee rooms", person-to-person, person-to-machine, machine-to-machine, ... that you can access. But they permit two different usernames from the same computer (and they have in the order of 220,000 registered players).

Now, if you detect a "twin player" here in http://www.calcudoku.org, what to do with this non "sportsmanlike" attitude?. Difficult decision (dilemma), certainly, Patrick. But the situation is in some way similar to the use of software to solve puzzles as commented in the past in other thread (also detectable).

Author:  starling  [ Tue Jan 31, 2012 1:28 am ]
Post subject:  Re: Timed Puzzle Bonus

Also thought I should point out, since it seems relevant, my past three days:

1/29: 12.871 5x5
1/30: 4.845 4x4
1/31: 12.806 5x5

The other thing, which makes me feel bad for sneaklyfox, was 1/30's 4x4 rankings:

1st place: 4.845 seconds (starling)
2nd place: 5.078 seconds (sheldolina)
3rd place: 6.178 seconds (sneaklyfox)
4th place: 9.175 seconds (pnm)

And meanwhile, the best times rankings. That is, the best times that any given individual has ever gotten:

1st place: 4.737 seconds (sheldolina)
2nd place: 4.845 seconds (starling)
3rd Place: 5.169 seconds (sneaklyfox)
4th place: 6.192 seconds (hopingcc)

The significance to this is that sneaklyfox posted a time that literally only two people have ever beaten. And then those two people beat her (Technically not chronologically, since I do my puzzles before she does, but still).

Author:  sneaklyfox  [ Tue Jan 31, 2012 5:02 am ]
Post subject:  Re: Timed Puzzle Bonus

starling wrote:
Also thought I should point out, since it seems relevant, my past three days:

1/29: 12.871 5x5
1/30: 4.845 4x4
1/31: 12.806 5x5

The other thing, which makes me feel bad for sneaklyfox, was 1/30's 4x4 rankings:

1st place: 4.845 seconds (starling)
2nd place: 5.078 seconds (sheldolina)
3rd place: 6.178 seconds (sneaklyfox)
4th place: 9.175 seconds (pnm)

And meanwhile, the best times rankings. That is, the best times that any given individual has ever gotten:

1st place: 4.737 seconds (sheldolina)
2nd place: 4.845 seconds (starling)
3rd Place: 5.169 seconds (sneaklyfox)
4th place: 6.192 seconds (hopingcc)

The significance to this is that sneaklyfox posted a time that literally only two people have ever beaten. And then those two people beat her (Technically not chronologically, since I do my puzzles before she does, but still).


This is exactly why I am probably most against the current awards system. It's very annoying having good times and only being 3rd and getting the same points as the next two people (but by the same token, I laugh when I have not so good times and still end up 5th getting the same points as the two people with better times). Btw, nice job on the sub-5-sec 4x4, starling as well as the 12-sec 5x5s.

Author:  jaek  [ Tue Jan 31, 2012 5:27 pm ]
Post subject:  Re: Timed Puzzle Bonus

sneaklyfox wrote:
I just like to think of the loopholes for fun.

In the online puzzle community this is called 'white hat calcudoku hacking'. Okay, that's a lie.

Author:  pnm  [ Tue Jan 31, 2012 10:47 pm ]
Post subject:  Re: Timed Puzzle Bonus

sneaklyfox wrote:
This is exactly why I am probably most against the current awards system. It's very annoying having good times and only being 3rd and getting the same points as the next two people (but by the same token, I laugh when I have not so good times and still end up 5th getting the same points as the two people with better times).

Ok, ok, once I find some time I'll see if I can improve the awards setup..

Patrick

Author:  clm  [ Fri Feb 03, 2012 11:48 pm ]
Post subject:  Re: Timed Puzzle Bonus

sneaklyfox wrote:
clm wrote:
Sure, you are a very good and elegant player and, as you are very smart and good observer, probably you have already noted what, in a pair of cases (at least apparently and coming from long time ago), looks like "twin players" [smile]. Who knows ...


I don't think I'm the observant type. So no, I haven't noted any apparent "twin players" in the past. Agree with Patrick that it would be unsportsmanlike ...


Ok, these would be the "conditions" for the "twin players":
--They would appear and disappear (active puzzlers) practically at the same moments and with the same flag of course.
--They would have everyday practically the same performance in timed puzzles (4x4's, 5x5's and 6x6's) and also almost identical "Best time", "Average of 3 best times" and "Average of 65% best times" along the time, because this factor is very "personal".
--They would have the "Last 7 Days" points identical, gold star or not, going "on vacation" or coming "from vacation", practically everyday, etc.
--They would have very similar "All time" points (the difference perhaps due to the total number of days sending solutions, since the creation of the usernames could have a delay).

"I don't say that the masses atract the other masses, I only say that they behave like if they would atract ..." (I. Newton, not literal).

So, in this case, paraphrasing Newton, I can not say, observing two lines, that they are in fact parallel, but they behave like if they were parallel [smile].

Author:  sneaklyfox  [ Sat Feb 04, 2012 5:25 am ]
Post subject:  Re: Timed Puzzle Bonus

clm wrote:
Ok, these would be the "conditions" for the "twin players":
--They would appear and disappear (active puzzlers) practically at the same moments and with the same flag of course.
--They would have everyday practically the same performance in timed puzzles (4x4's, 5x5's and 6x6's) and also almost identical "Best time", "Average of 3 best times" and "Average of 65% best times" along the time, because this factor is very "personal".
--They would have the "Last 7 Days" points identical, gold star or not, going "on vacation" or coming "from vacation", practically everyday, etc.
--They would have very similar "All time" points (the difference perhaps due to the total number of days sending solutions, since the creation of the usernames could have a delay).

"I don't say that the masses atract the other masses, I only say that they behave like if they would atract ..." (I. Newton, not literal).

So, in this case, paraphrasing Newton, I can not say, observing two lines, that they are in fact parallel, but they behave like if they were parallel [smile].


Yes, it probably is like that as you say, but have you noticed any suspicious "twin players"? Because there are a lot of users. I only notice the other people who are in my range (or especially if they get more points than me when they shouldn't which was how we noticed the bug where you get another bonus puzzle if you hit the target points exactly.)

Author:  clm  [ Sat Feb 04, 2012 10:05 am ]
Post subject:  Re: Timed Puzzle Bonus

sneaklyfox wrote:
clm wrote:
Ok, these would be the "conditions" for the "twin players":
--They would appear and disappear (active puzzlers) practically at the same moments and with the same flag ...


Yes, it probably is like that as you say, but have you noticed any suspicious "twin players"? Because there are a lot of users. I only notice the other people who are in my range (or especially if they get more points than me when they shouldn't which was how we noticed the bug where you get another bonus puzzle if you hit the target points exactly.)


Yes, I feel that presently there are two pairs of "suspicious" "twin players" but I could be wrong (circumstantial facts) and in this case (this is not a bug in the software) I can "say the sin but not the sinner". It's just a curiosity ... anyway the very important thing is to solve the puzzles and get fun with this wonderful site provided by Patrick (additionally being clean and honest is a personal option) (if, after some time, you were still interested in this subject, I could tell you my suspicions and comments in a PM).

Author:  sneaklyfox  [ Sat Feb 04, 2012 5:09 pm ]
Post subject:  Re: Timed Puzzle Bonus

clm wrote:
Yes, I feel that presently there are two pairs of "suspicious" "twin players" but I could be wrong (circumstantial facts) and in this case (this is not a bug in the software) I can "say the sin but not the sinner". It's just a curiosity ... anyway the very important thing is to solve the puzzles and get fun with this wonderful site provided by Patrick (additionally being clean and honest is a personal option) (if, after some time, you were still interested in this subject, I could tell you my suspicions and comments in a PM).


No, I don't think I need to know who these possible "twin players" are. (If Patrick wanted to know, you could tell him and he could probably check if they came from the same computer, but then what? I don't think he would actually do anything about it. It could be a coincidence that two people on the same computer have about the same level of intelligence.) I guess it would only bother me if they had more points than me ha. I was only surprised that you could be so observant to see that they had similar points and timings, and see that they got on and off at the same time or were both on vacation at the same time and check that it was like that across the whole board all the time. Anyway, I don't see much point in submitting solutions for the regular puzzles twice everyday. It's just a waste of time. What would be the purpose? Timed puzzles is another thing because each user gets different puzzles so any "twin player" could practice more by creating another user but only for timed, not the regular.

Page 10 of 13 All times are UTC + 1 hour [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/