View unanswered posts | View active topics It is currently Wed Feb 21, 2018 8:57 pm ← Back to the Calcudoku puzzle page

 Page 2 of 3 [ 24 posts ] Go to page Previous  1, 2, 3  Next
 Print view Previous topic | Next topic
more of which puzzle?
Author Message

Posted on: Thu Jan 12, 2012 2:28 pm

Posts: 698
Joined: Fri May 13, 2011 6:51 pm
Re: more of which puzzle?
pnm wrote:
sneaklyfox wrote:
Anyway, if the thought is that there are too many puzzles in a day why are we now talking about adding another one?

Not really. My question was another way of asking "what is your favourite puzzle?" ...

Timed puzzles: In case of adding an extra puzzle to the timed puzzles, other possibilities I may imagine are:

- An special 5x5 with negative numbers
- An special 6x6 with many operators (including mod, bitwise, exp ... )
- An 8x8 (as underlined by starling) with, let's say, 9, 7, 5, 5, 5 for the first five times and later, let's say, 3 points for every solver who obtains a correct solution within some specified time, i.e., 30 minutes, or 1 hour ... (this "limiting time" philosophy, in addition and not in opposition to the regular "24 hours" time, could be extended to other "extra" puzzles in the future, with different sizes, 9x9, 12x12, ...).

Posted on: Thu Jan 12, 2012 4:54 pm

Posts: 422
Joined: Fri May 13, 2011 2:43 am
Re: more of which puzzle?
pnm wrote:
Not really. My question was another way of asking "what is your favourite puzzle?"

Oh and nice work on the 28 second 6x6, and get well soon..

That's a different question. My favourite is probably 8x8 difficult because it's somewhat more logically challenging than the smaller ones and yet doesn't take as much time as some of the bigger ones.

Haaahaaa... I'm feeling a little better this morning after a good night's rest. We'll see what happens with the timed puzzles now haa...

Posted on: Thu Jan 12, 2012 7:29 pm

Posts: 175
Joined: Fri May 13, 2011 2:11 am
Re: more of which puzzle?
clm wrote:
- An 8x8 (as underlined by starling) with, let's say, 9, 7, 5, 5, 5 for the first five times and later, let's say, 3 points for every solver who obtains a correct solution within some specified time, i.e., 30 minutes, or 1 hour ... (this "limiting time" philosophy, in addition and not in opposition to the regular "24 hours" time, could be extended to other "extra" puzzles in the future, with different sizes, 9x9, 12x12, ...).

Personally, I'd prefer adding 3 8x8's in the same form as we have the 3 sizes right now, and doubling the points for places.

That said, I think that maybe a point for solving at all would be useful on the 8x8 timed, since unlike any of the smaller ones, I'd anticipate that an 8x8 timed would actually require some concerted effort to solve.

As for a "points for solving it within x amount of time" solution, I'd think even 10 minutes would be too much.

The peak of the 6x6's right now occurs around 30 seconds. My 24 second's a pretty solid outlier among peak scores, since all my other best times, and everyone else's among the really high level times, are closer to 28-30 seconds. Doubling that number since an 8x8 would have just short of twice the characters gives a cap around 55 seconds to a minute. This obviously doesn't work perfectly, since using 6 seconds for 4x4's projects to a 10 second 5x5 and a 12 second 6x6, and that ignores the 4.something to 5.5 times which sheldolina and I have.

Still, though, even if a minute and a half is about where the time cap occurs, then the average time would probably fall around 6-7 minutes. And by average, I refer to the number I expect to see other people around, which is about 1.5-2 minutes on 6x6. With that in mind, I'd think a 10 minute 8x8 would show what would equate to about a 3 minute 6x6, which I don't think should get 3 points.

Posted on: Thu Jan 12, 2012 11:44 pm

Posts: 698
Joined: Fri May 13, 2011 6:51 pm
Re: more of which puzzle?
starling wrote:
clm wrote:
- An 8x8 (as underlined by starling) with, let's say, 9, 7, 5, 5, 5 for the first five times and later, let's say, 3 points for every solver who obtains a correct solution within some specified time, i.e., 30 minutes, or 1 hour ... (this "limiting time" philosophy, in addition and not in opposition to the regular "24 hours" time, could be extended to other "extra" puzzles in the future, with different sizes, 9x9, 12x12, ...).

Personally, I'd prefer adding 3 8x8's in the same form as we have the 3 sizes right now, and doubling the points for places.

That said, I think that maybe a point for solving at all would be useful on the 8x8 timed, since unlike any of the smaller ones, I'd anticipate that an 8x8 timed would actually require some concerted effort to solve.

As for a "points for solving it within x amount of time" solution, I'd think even 10 minutes would be too much.

The peak of the 6x6's right now occurs around 30 seconds. My 24 second's a pretty solid outlier among peak scores, since all my other best times, and everyone else's among the really high level times, are closer to 28-30 seconds. Doubling that number since an 8x8 would have just short of twice the characters gives a cap around 55 seconds to a minute. This obviously doesn't work perfectly, since using 6 seconds for 4x4's projects to a 10 second 5x5 and a 12 second 6x6, and that ignores the 4.something to 5.5 times which sheldolina and I have.

Still, though, even if a minute and a half is about where the time cap occurs, then the average time would probably fall around 6-7 minutes. And by average, I refer to the number I expect to see other people around, which is about 1.5-2 minutes on 6x6. With that in mind, I'd think a 10 minute 8x8 would show what would equate to about a 3 minute 6x6, which I don't think should get 3 points.

Basically I would agree with your analysis, my reference to a specific time (30 minutes, 1 hour, ... ) was just tentative. The main idea was to suggest the concept of the "limiting time" (with some "fix prize" at the end) to compensate for a higher dedication and effort (from that minimum 6-7 minutes if only one 8x8 is necessary to those 18-21 minutes of "concentration without disruptions" if all three 8x8's are necessary).
It is perfectly valid to extrapolate the data we have accumulated till the moment with the 4x4's, 5x5's and 6x6's but, Would it be linear for everyone and for every size?, perhaps we would need to have some real experience with the 8x8's (or bigger, 9x9 ... ) (the "human factor" introduces the non-linearities) and later to make the adjustments.

Posted on: Fri Jan 13, 2012 2:53 am

Posts: 422
Joined: Fri May 13, 2011 2:43 am
Re: more of which puzzle?
If we really were to have an 8x8 timed puzzle, I think it would be more attractive to all users if x number of points were awarded just for finishing the puzzle regardless of time. I know it sounds ridiculous once you take more than a certain amount of time to complete a puzzle, but it's not like 4x4, 5x5, or 6x6. Even if you do 3 timed 4x4s and don't get any points for it, it's not biggie because they normally only award 1-2 points anyway... (2-3 points for 5x5s normally, 3-5 points for 6x6s normally.)

Speaking of timed points again, I'm still not highly in favour of the current awards system, but I'm not sure how it can be "fixed". Maybe it would be better to award points based on mean and standard deviation over all timed puzzles done by all users up to date? I'm not sure how difficult that would be to implement, and the actual points awarded would still be something to decide. I also understand the reasons for it not being too many points, but we don't have a decimal point system... unless say all the normal puzzles are awarded points times factor of 5 or 10 and then there could be more difference for the timed puzzles.

Posted on: Fri Jan 13, 2012 4:16 am

Posts: 175
Joined: Fri May 13, 2011 2:11 am
Re: more of which puzzle?
sneaklyfox wrote:
If we really were to have an 8x8 timed puzzle, I think it would be more attractive to all users if x number of points were awarded just for finishing the puzzle regardless of time. I know it sounds ridiculous once you take more than a certain amount of time to complete a puzzle, but it's not like 4x4, 5x5, or 6x6. Even if you do 3 timed 4x4s and don't get any points for it, it's not biggie because they normally only award 1-2 points anyway... (2-3 points for 5x5s normally, 3-5 points for 6x6s normally.)

Speaking of timed points again, I'm still not highly in favour of the current awards system, but I'm not sure how it can be "fixed". Maybe it would be better to award points based on mean and standard deviation over all timed puzzles done by all users up to date? I'm not sure how difficult that would be to implement, and the actual points awarded would still be something to decide. I also understand the reasons for it not being too many points, but we don't have a decimal point system... unless say all the normal puzzles are awarded points times factor of 5 or 10 and then there could be more difference for the timed puzzles.

I agree completely with both of these, but think there's some question as to whether we would need a "5 points for solving an 8x8 timed," or "2 points for each 8x8 timed you solve." The point structure we keep right now seems to say it'd be the 5 points per.

Posted on: Fri Jan 13, 2012 5:19 am

Posts: 422
Joined: Fri May 13, 2011 2:43 am
Re: more of which puzzle?
starling wrote:
I agree completely with both of these, but think there's some question as to whether we would need a "5 points for solving an 8x8 timed," or "2 points for each 8x8 timed you solve." The point structure we keep right now seems to say it'd be the 5 points per.

I don't see how the point structure we keep right now would say 5 points per solved timed 8x8. Right now, no points are awarded for any timed puzzles just for solving... you have to be among 20 fastest of the day to get anything.

Gotta say, starling, how glad I am I didn't get my 28-sec 6x6 solve today! Just highlights the problem I see with the award system for timed puzzles. Just imagine the timing for best (different user's) times of the day being something like this...
1. 28.013 sec
2. 28.282 sec
3. 28.505 sec
4. 49.004 sec
5. 54.898 sec...

And the poor guy (or gal) in 3rd place gets the same number of points as the one in 5th place while the one in 1st place gets two points more just for being half a second faster?

Of course though, in a race, as the current award system is based on, it doesn't matter if you beat the next guy by a nose or by a mile...

But consider also that we can't tell if an individual's race is as easy as the next guy's race. It's not the same course (puzzle).

Posted on: Fri Jan 13, 2012 6:32 am

Posts: 175
Joined: Fri May 13, 2011 2:11 am
Re: more of which puzzle?
sneaklyfox wrote:
starling wrote:
I agree completely with both of these, but think there's some question as to whether we would need a "5 points for solving an 8x8 timed," or "2 points for each 8x8 timed you solve." The point structure we keep right now seems to say it'd be the 5 points per.

I don't see how the point structure we keep right now would say 5 points per solved timed 8x8. Right now, no points are awarded for any timed puzzles just for solving... you have to be among 20 fastest of the day to get anything.

Gotta say, starling, how glad I am I didn't get my 28-sec 6x6 solve today! Just highlights the problem I see with the award system for timed puzzles. Just imagine the timing for best (different user's) times of the day being something like this...
1. 28.013 sec
2. 28.282 sec
3. 28.505 sec
4. 49.004 sec
5. 54.898 sec...

And the poor guy (or gal) in 3rd place gets the same number of points as the one in 5th place while the one in 1st place gets two points more just for being half a second faster?

Of course though, in a race, as the current award system is based on, it doesn't matter if you beat the next guy by a nose or by a mile...

But consider also that we can't tell if an individual's race is as easy as the next guy's race. It's not the same course (puzzle).

Oh no, not per puzzle, for solving one puzzle, and solving three gets you no more. That's why I'm saying the current point structure suggests that, since only solving one gets you anything.

But yeah, I feel like having major gaps in the times is the problem with static, relative numbers of points by places.

There was that one day I got 4 for a 6.1 second 4x4, back when this was like the 3rd fastest time we had ever done. Sheldolina got the 2nd fastest as of then that same day.

Posted on: Fri Jan 13, 2012 7:37 am

Posts: 98
Joined: Fri May 13, 2011 12:48 am
Re: more of which puzzle?
At one point you had a 7x7 that incorporated all the specialty operations. I think it was one of the bonus puzzles, but it was fairly easy. I think something like that could get people more used to the alternative operators.

Posted on: Fri Jan 13, 2012 11:37 am

Posts: 2246
Joined: Thu May 12, 2011 11:58 pm
Re: more of which puzzle?
picklepep wrote:
At one point you had a 7x7 that incorporated all the specialty operations. I think it was one of the bonus puzzles, but it was fairly easy. I think something like that could get people more used to the alternative operators.

Yes, "7x7 ManyOp" (on Mondays), I ended up dropping the bitwise OR operator from that one..
Patrick

Display posts from previous:  Sort by
 Page 2 of 3 [ 24 posts ] Go to page Previous  1, 2, 3  Next

 You cannot post new topics in this forumYou cannot reply to topics in this forumYou cannot edit your posts in this forumYou cannot delete your posts in this forum

Search for:
 Jump to:  Select a forum ------------------ English    Announcements    Calcudoku General    Other number stuff    Solving strategies and tips    Specific puzzles / your own puzzles    Timed Puzzles    Bugs and errors    Killer Sudoku    Sudoku Nederlands    Aankondigingen    Calcudoku Algemeen    Oplostips en strategieën Italiano    Calcudoku Generale    Strategie e consigli per risolvere Español    Avisos    Calcudoku - General    Estrategias de solución y aspectos relevantes
All forum contents © Patrick Min, and by the post authors.

Forum software phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by STSoftware.