Closed oliver-phet closed 6 years ago
This appears to be an oversight, as I thought the displacement already snapped to a whole number of degrees. @jonathanolson can you make this change? Once the angle is snappy, we can test again and see how well the period agrees.
Implemented, can you verify?
Looking good @jonathanolson!
This looks good to me, too!
Applied to the 1.0 branch, closing.
Reopening so @oliver-phet can eventually email the user.
As a consequence, you can now snap exactly to 180 degrees. This is causing a big problem where period times are wildly different:
Don't ask how in the world the pendulum started swinging, it just did... and no I did NOT have friction during this. These problems are also closely reminiscent #91 and #128. @arouinfar @jonathanolson we should probably not allow snapping to 180 degrees, or if we do there needs to be more discussion.
we should probably not allow snapping to 180 degrees
That would be perfectly fine with me @phet-steele @jonathanolson.
Applied fix to master and the release branch. @phet-steele and @arouinfar can you verify?
I like the new behavior @jonathanolson, let's hope @phet-steele isn't able to break it. :)
@oliver-phet can you please notify the user that we've published a maintenance release that addresses this issue?
Looks like change using Util.sign worked on master, but had an older copy of dot in its SHAs. @phet-steele, https://www.colorado.edu/physics/phet/dev/html/pendulum-lab/1.0.2-rc.1/pendulum-lab_en.html should presumably solve the issue on production?
Good to go JO, thanks for the quick turn around.
Deployed, http://phet.colorado.edu/sims/html/pendulum-lab/1.0.2/pendulum-lab_en.html should have fix in place.
Email to user sent. Closing for now.
The sim currently allows users to drag two masses to different angles, but the sim displays the same angle. For example, both of these pendulums were dragged to "40°"![screen shot 2017-09-25 at 2 29 19 pm](https://user-images.githubusercontent.com/5090275/30829457-0a339dc8-a1fe-11e7-9168-82fb41a54986.png)
The Flash version forced the pendulum to "snap" to discrete angles. Which we should try to emulate this behavior since the current implementation can be misleading.
This was brought up by a teacher who was getting different period times for identically configured pendulums.
Assigning to @arouinfar for now.