absolutebest not equal to creditbest score

Case number:845813-994750
Topic:Game: Tools
Opened by:bertro
Status:Open
Type:Bug
Opened on:Monday, March 18, 2013 - 01:08
Last modified:Thursday, March 28, 2013 - 22:48

Hi,

Got this solution saved for Myself on 687 under the name: Bug Very Best vs Credit Best (@score 9531)

If you load it and do nothing else than run this recipe:

creditbest.Restore()
print(current.GetScore())
recentbest.Restore()
print(current.GetScore())
absolutebest.Restore()
print(current.GetScore())
recentbest.Restore()
print(current.GetScore())

The output will be:
9531.072897263
9531.072897263
9132.9813264145
9531.072897263

You can also click the two hearts in the Undo window for the same result.

absolutebest should allways be equal to (or higher than) creditbest on a single track like this one that generated the creditbest score in the first place. Even if I Wiggle Sidechains to increase the creditbest score, the absolutebest score still returns the lower value. Only changing tracks did update the absolutebest value.

I did use the Disable slow filters checkbox at one time for a recipe.

I don't think it should happen :)

(Mon, 03/18/2013 - 01:08  |  5 comments)


tamirh's picture
User offline. Last seen 6 years 13 weeks ago. Offline
Joined: 05/11/2012

Thanks, I'll look into this.

spdenne's picture
User offline. Last seen 36 weeks 3 days ago. Offline
Joined: 10/01/2011
Groups: Void Crushers

I'd seen this too, when the absolute best was obtained with slow filters disabled. With them re-enabled, the score of the absolute best pose is lower than the score of the credit best pose.

Joined: 09/24/2012
Groups: Go Science
Status: Open » Closed

same as here:

http://fold.it/portal/node/994817#comment-22328

=> I close this duplicate

bertro's picture
User offline. Last seen 2 weeks 3 days ago. Offline
Joined: 05/02/2011
Groups: Beta Folders
Status: Closed » Open

@ bruno.kestemont:

I would greatly appreciate it if you did not close feedback that is not your own and that devs are actively looking into at the moment.

I am pretty sure that tamirh will close it in time with proper feedback when he has completed his investigation.

Thank you

tamirh's picture
User offline. Last seen 6 years 13 weeks ago. Offline
Joined: 05/11/2012

reassigning, tjbertram has been looking into these issues.

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