On slow filter puzzles

Case number:845813-2000327
Topic:Game: Tools
Opened by:Timo van der Laan
Opened on:Wednesday, February 4, 2015 - 20:49
Last modified:Sunday, February 8, 2015 - 02:39

First of all a question:
- Are the game tools affected by any of the slow filters?
I really hope they dont, see my code in http://fold.it/portal/recipe/100662.
This would only work good if the aren't.

Then there are some extra Lua functions that would really help for coping with these kind of puzzles:
1) A function to get the total of maximum bonus for all slow filters
2) A function that computes and returns the current bonus of those filters even if those filters are turned off.

(Wed, 02/04/2015 - 20:49  |  2 comments)

LociOiling's picture
User offline. Last seen 6 hours 4 min ago. Offline
Joined: 12/27/2012

The link in Timo's post gives a 404 due to excess punctuation.

The correct link is: http://fold.it/portal/recipe/100662

MurloW's picture
User offline. Last seen 14 weeks 4 days ago. Offline
Joined: 11/21/2012

Ever since the filters were introduced (long time back, I know), I have noted a change in the Wiggle feature when filters are on or off.
Whenever the filters are on and CI=1.0, the Wiggle feature is "being held back" by the filters; changing CI to .95~.99 often (but not always) "fixes" this issue and lets you wiggle on.

In essence, the problem here is: when CI == 1.0, Wiggle will not allow a score drop; even when this drop is purely due to the filters.
In every puzzle where the filters are present (and I do mean every single one, at least in my experience so far), there will be a point (with filters on) where you can wiggle at CI 1.0 without any change whatsoever, turn off the filters and start that wiggle at the exact same point and see a gain of several points, up to dozens or even more; after which you turn off the filters and have one of them drop your score, below your score previous to that wiggle.

As far as I can tell, this means that whenever the filters are on, the Wiggle function is not functioning properly. (being "held back" as stated before.)

e.g. structure at 9500pts, all filters are happy and gaining:
filters on -> I wiggle (CI 1.0), no change, not even 0.001.
Turn filters off -> wiggle, immediate gain of several points -> turn filters off, one (or more) of the filters is unhappy (netting minus several hundred points or whatever that filter is set to) and needs fixing.
I know how to deal with this, fix the fragment or invalid residue or whatever; but many less experienced players have no idea why or how this is happening, nevertheless how to fix it.

I see this as a bug: the wiggle function should continue manipulating the backbone and everything else according to CI function (based on Van der Waal's) and not be affected by the filters.


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, Boehringer Ingelheim, RosettaCommons