key differences in the Grand Challenge Puzzles

A few of our players have pointed out the palpable difference in the game play that emerged in the new puzzles.

One of the main reasons for the different feel and different rate of progress for the new puzzles is that
they are often the best solutions found by the automatic methods -- meaning that the starting point is a fairly strong local minimum (slight changes invariably lead to worse scores or little difference). 

We are trying to find out whether people know how to get out of these, and find a much better space of
solutions by making one or several key changes and then refining these until a
better solution is found.  Often making the big key change may initially make the score a lot worse, even though the score would eventually be a lot higher after some refinement.  So in many ways this is harder, and also one of the key reasons why computers cannot really do anything to improve the score.

Also, since the score initially is already very high (the best the computer can get), the final solution is not going to be significantly higher than the starting position.  In the Grand Challenge puzzles 100 or even 20 points makes a huge difference, a difference between a really good prediction and the many flawed ones that massive distributed computing efforts generate.

Also, one possible reason why the solutions may seem to require a lot of small tweaks at some point is that you may have reached a global solution (the native protein) to the puzzle and any subsequent tweaks are simply not possible.  At least we hope that is one of the reasons ;)

So keep all this in mind as you tackle the Grand Challenges.  They are harder, and every 10 point difference can make a huge impact.  You're at the bleeding edge of the current state-of-the-art in protein structure prediction.

( Posted by  zoran 73 1944  |  Fri, 12/26/2008 - 19:32  |  7 comments )
xiando's picture
User offline. Last seen 6 years 19 weeks ago. Offline
Joined: 10/19/2008
Groups: Oma Gawd
Thank you

Thank you for your comments Zoran. Although they present a few questions in my mind, it is helpful to know the basis upon which these challenges are presented.

My first question would be this:

In what specific way do you mean "different feel"? "Different rate I can guesstimate an answer to, but the first is just a bit vague for my fuzzy Post-Xmas mind... ;)

thanks and Happy New Year if you're still on break

Joined: 05/19/2008
Groups: None
Please help me to

Please help me to understand: At this moment one of the major science goals is to know how a human player can figure out the way from a local minimum.

If this is so, my thoughts are the following, mostly speculative though, because I don't have enough information:

- Squeezing out more and more points with local wiggles NOW is secondary, or even less important from scientific point of view. Local wiggles are not meant to change the structure significantly.
- Maybe the app should support this idea in a way that local wiggle is not giving as many points as it does recently.
- The app should support and encourage resetting the puzzle to be able to try new directions. For this the (ever requested and ignored) "New Best" function is absolutely necessary.
- Maybe the app should hint what the top solutions are and what are the major differences from my current solution. Probably Peekaboo and Oracle are meant to do this, but then they should not depend on a completely irrelevant function, the Duel.
- Maybe the app should - similarly to the previous - hint what other players found a dead end street so far. (Maybe a player should be able to declare a solve as dead and put into public for others' benefit.)

...but only thoughts.


Joined: 10/23/2008
Groups: None
Where are the Grand Challenge Puzzles ??

Where are these Grand Challenge Puzzles. Or are the latest ones, the Grand Challenge type ?? (like 117, for instance) ??

zoran's picture
User offline. Last seen 8 years 19 weeks ago. Offline
Joined: 11/10/2007
Groups: Window Group
where are they

we've relabeled them as Grand Challenge #, to make it clear.

Joined: 05/10/2008
I add my voice to

I add my voice to DisposableHeart's excellent comment. I have suggested in Feedback what a "new best" feature could look like from a user interface point of view - although there is certainly room for simplification.

Local wiggling is tedious but I guess it may be useful at discovering "unstable" areas to be exploited later on. An automated process could be implemented to achieve the same result (I wrote another feedback post about that).

For what it's worth, I find puzzle 118 much more "solvable" than the previous puzzle - in the sense that it is more friendly to gradual manual improvements.

zoran's picture
User offline. Last seen 8 years 19 weeks ago. Offline
Joined: 11/10/2007
Groups: Window Group
New Best tool

Seth just implemented the new best feature.  It should be coming to the pc near you with an update slated for later on this week.  feedback is welcome, as always.

zoran's picture
User offline. Last seen 8 years 19 weeks ago. Offline
Joined: 11/10/2007
Groups: Window Group
solvability of puzzles

just a warning that the puzzles in the Grand Challenge group may be easier that the current one, but they also may get significantly harder. we can't wait to find out if our suspicion about the source of superiority of human solutions shows up in the solutions.

User login
Download links:
  Windows    OSX    Linux  
(10.12 or later)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Social Media

Only search
Other Games: Mozak
Recommend Foldit
Top New Users

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons