Band_set_strength to 0 and band_disable() not giving same results

Case number:699969-990059
Topic:General
Opened by:bertro
Status:Open
Type:Bug
Opened on:Tuesday, July 26, 2011 - 15:52
Last modified:Wednesday, August 10, 2011 - 14:27

In a script, setting ALL bands to 0 using a loop with band_set_strength(band_index,0) does not gives the same scoring results after the same do_shake(x) than doing a band_disable().

(Tue, 07/26/2011 - 15:52  |  4 comments)


Joined: 11/10/2007
Groups: Window Group
Status: Open » Open

This is not too unexpected, as shake may not do the same thing every time it is run, regardless of band settings. Do you get the same score every time if you just use band_disable?

bertro's picture
User offline. Last seen 1 day 2 hours ago. Offline
Joined: 05/02/2011
Groups: Beta Folders

With the same starting point/score, running a few times the same script with all bands to band_set_strength(x,y) with y=0 was giving a result close to the starting score. Now, same script but replace the band_set_strength(x,y)loop with a band_disable() gives a totally different set of result score. The puzzle is a guide puzzle and the starting score was a very tight minima point. Results are never the same but the two sets where really apart.

bertro's picture
User offline. Last seen 1 day 2 hours ago. Offline
Joined: 05/02/2011
Groups: Beta Folders

I have a repro «solution» and script to test this bug that ,so far, affect both do_global_wiggle_backbone and do_global_wiggle_all.

I can copy the script here if you want.

How can I find the right «solution» to send you?

bertro's picture
User offline. Last seen 1 day 2 hours ago. Offline
Joined: 05/02/2011
Groups: Beta Folders

No feedback so far...

So, do you consider this a bug or not?

Sitemap

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