DeborahLynn, I'm glad you were able to pass this level.
The bug in Flippin' Sheets, however, is still there so until we fix that I will make this as Ongoing:
Although it can be solved in 3 moves, resetting the puzzle and doing those 3 same moves will not always result in a score of 9300. Even using local wiggle doesn't seem to get past 9300 sometimes if the initial tweak wasn't good enough.
This bug is still relevant. I just downloaded FoldIt today, and was able to progress at a reasonable rate until this level. Ubuntu Linux 10.04 x64.
Please let me know if there is anything I can do as a user to help debug this.
Thanks for the reminder, Seth can you look into this ASAP?
Phyzome, I hope the above hints have helped you pass this level, sorry about this bug!
I finally managed to get this with some help from IRC – it does seem to be a local minimum issue. I did a shake'n'wiggle before flipping the sheet, and it worked.
Before that, I was able to get to 9291 by flip, shake, wiggle, then tweaking the tail and wiggling more – but that's all.
Perhaps the threshold just needs to be adjusted downwards a bit, since this puzzle is just for introducing a tool.
If you see the thing that says "the protein is falling apart, you should reset"-ignore it. You do NOT have to reset. It's in there to advise people who manage to completely unravel the protein, but it's over-sensitive.
That's my main piece of advice.
Strangly, I've never had problem with it. I wonder if it is machine specific problem.
IIRC this is one I despaired over for about three days in late August. At the time I was using an 8-year old Windows XP box with video that also needed that trick with turning down the video settings. I've wondered since if these two bumps in the road might indeed be killing off a lot of aspiring folders.
A possible compromise might be lowering the score threshold needed for solving this puzzle…
If you make it too easy, they won't learn the patience needed to do the real puzzles.
It isn't particularly hard.
Re: Brick
"[This puzzle] isn't particularly hard."
In theory, yes, as the correct solution only takes 3 moves as already mentioned. In practice, however the search space for tweaking is actually quite large because there are four beta strands to tweak. For example, assuming that each strand may be shifted for up to 2 moves in either direction, that gives you 5^4 = 625 possible options already. Sure, one can narrow the search space down to 5 when given the hint that the 2nd strand is the problem, but it turns out that even following LennStar's instruction perfectly (see see http://foldit.wikia.com/wiki/Tutorial_Puzzles ) is not always enough to break the 9300 cutoff.
[Note that the behavior tab and the rebuilding feature are disabled for this puzzle. Practically speaking, this means that fixing the loop that ends up getting a bad backbone as a result of the shift is out of the question. One would thus have to hope that the drag-then-wiggle trick would matter at the margin– but unfortunately, the result is often getting stuck in the 9290s.]