[NewChapter] Wiggle Accuracy Slider

Case number:954892-996664
Topic:Developer Preview
Opened by:brow42
Opened on:Thursday, January 16, 2014 - 04:21
Last modified:Saturday, January 18, 2014 - 01:35

I would like to see the return of the Wiggle Accuracy slider in both UI and Lua (removed fold.it/portal/node/991932 ). Assuming the hooks into mini-rosetta still exist, this would let us a) globally work faster for hand-folding or when intentionally creating clashes (blue fuze part 1) where we don't need to find the minimum, or run longer to find the minimum in the endgame or when evaluating poses ( blue fuze part 2). Scripts may do this automatically in, for example, blue fuze subroutines.

The return of these sliders should make clear the following:

one side is MUCH faster but lower scoring (e.g. 0)
one side is MUCH slower but higher scoring (e.g. 7)
pre-slider behavior is #x (e.g. 7, but not applicable if introduced concurrent with new chapter)
a reasonable default, such as 3.

Definitely do not leave it as a social experiment to see if players invent for the slider without guidance! /fold.it/portal/node/990943#comment-13841

See also fold.it/portal/node/991938 and again http://fold.it/portal/node/990943#comment-13853 for name suggestions.

I have had to remove all links because of the spam filter.

(Thu, 01/16/2014 - 04:21  |  2 comments)

Joined: 11/10/2007
Groups: Window Group

Yes, we are looking in to adding a similar feature to give more control over wiggle in newchapter.

Joined: 11/10/2007
Groups: Window Group

We're releasing an update to newchapter shortly to try a new control for wiggle: Wiggle Power. Currently, this option controls how much wiggle will be allowed to try to idealize the structure: low power does some idealization, medium power does a bit, and high power does a lot. These should roughly trade off between runtime and how many points wiggle can find.


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