Some members of our group are reporting that they see large gains on the score board while running a script, but the script (which uses restore best or quicksaves) is operating far below that score. This best score configuration cannot be retrieved by load best score in undo. They'll post their experiences below.
In my own experience, I just started a new client to try to work on 513 in the last hours. None of my 3 tracks had a best score anywhere near my scoreboard score. Also, my best recorded score autosave claims 9711 is my best score, but when loaded, says 9709. The board says 9776. I haven't run multiple clients on this puzzle.
Same problem for me when loading an autosave- the save says one score, when I load it, it's less.
I've seen it running a script that had already put me into my best score (an overnight run). The script calls for very best many times over, I can see that from the output - but is registering that "very best" 1 or 2 points lower than what is on the board. When the script is stopped, and restore very best button is clicked, the score does go to the board score. I've seen it a few times, but always in the wrong client to grab a screenshot.
Which script were you running when these problems occurred?
Are you attempting to evolve when this problem happens?
I'm not sure who you were addressing, but no, I was working solo, and I didn't record what script I was running.
The conversation seems to have moved to http://fold.it/portal/node/991851 so perhaps this one should be closed.
I've just had this happen to me - my scoreboard solo is 2 points higher than my absolute best.
However, another member of my team had evolved my solo by 2 points earliar today…..so is it possible that the evolver scores from other team members are showing up on the original player's solo scoreboard somehow ??
this post is not about structure it is about score - would a mod please adjust, we all have problems with obtaining best structure that is the game.
If anyone has steps for reproducing this, please let us know. For now we're limited to looking through the code for obvious bugs, but if we can get a reproduce-able case, we'll be able to debug it much faster.
