############################################
# Remaining Work
########################
• For certain values of screen brightness in the two lasers panel,
if there are two lasers of different colors that are both at maximum intensity,
the pattern on the screen is asymmetric (this caused considerable confusion for student,
and was much more noticeable in the interview than it is in this screen shot).
>>This is from reflection on the boundary. In Wave Interference sim,
I extend the lattice beyond the visible region;
here it could also be done, but very time consuming to implement,
and would slow down the sim (scales with the size of the boundary).
>>I checked the x locations of the wave sources, and they are symmetric.
Bug: Mandel panel photons all have same wavelength (though different color)
The reason we didn't discover this before is that it's hard to see with "time-averaged intensity mode" on.
Mandel Panel should have both visualization styles
>>This proves difficult with the current setup. I estimate 2+ hours to add these features. For right now, I've reverted to intensity-view only, and wavelength is changed with color.
>>see MandelSchrodingerPanel::createAccessor() for debugging this.
Reset should reset wavelength, velocity,
>>This proves to be very difficult right now (2+ hours?). After refactoring, should be piece of cake.
I've figured out what the problem is with the asymmetry in the Two
Lasers panel in QWI: the center of the left laser is 14 pixels from the
left edge of the box and the center of the right laser is only 13 pixels
from the edge of the box. My guess is that you're measuring the
distance from the center of the box, but it's rounding in some weird way
because there are an odd numbers of pixels (45) in the box. Can you
fix this?
Also, I did another interview on Saturday and there were more problems
with asymmetry in the other two panels. If you make the simulation full
screen so it's easy to see the individual pixels, you can see that
changing the width and separation of the slits often changes things in
very weird asymmetric ways.
___
Remember when we were working on units in QWI and there was a mysterious
factor of 10 that I couldn't figure out, so I told you to just roll it
in and not worry about it? Well, I finally figured out where it came
from! The units on the ruler for matter particles are off by a factor
of 10. Rather than being 0.45nm wide, the box should be 4.5nm wide, so
the units on the ruler should be 10 times as large. To compensate for
this, we made the units per clock tick 10x too small.
The current units per clock tick are:
Electrons 0.005 fs
Neutrons 0.01 ps
He Atoms 0.05 ps
(At medium resolution, divide these numbers by 4, at high, by 16)
The correct units per clock tick are:
Electrons 0.05 fs
Neutrons 0.1 ps
He Atoms 0.5 ps
(At medium resolution, divide these numbers by 4, at high, by 16)
___
Just found a small bug: If I resize the simulation, the close and rotate
buttons on the ruler move off the ruler. Also, if I put the ruler too
far to the right and make the simulation size smaller, the ruler goes
off the screen and I can't get it back without resizing.
I should mention that the bug is only there when the ruler is rotated
vertically.
I can't get the save/load feature to work in QWI. I thought maybe it
had gotten broken somewhere along the way, but even if I go back to old
versions, it doesn't work. It creates a file that contains nothing but
the following text:
When I load this file, there are no error messages, but nothing happens.
Sam
############################################
# Lower Priority
########################
If I check and uncheck “slit absorption” with no slits and no barriers, it creates little glitches in the beam.
High Intensity electrons with absorption and one detector gave blank answer in north region
(although intensity pattern appeared on screen).
When I move the vertical position of the slits, the response is very jerky and discontinuous. Can this be any smoother?
o Wave function norm is incorrect. A detector that covers the entire box only registers 80-95%, rather than 100%.
-this is partly due to the particle leaking outside the box
Allow user to move slit barrier up and down by clicking on it and dragging.
The “slit absorption” checkbox should have no effect if there are no slits or barriers.
o There is an abnormally bright region just beneath the slits in intensity mode with detector(s) on.
Ruler should revise size on resolution change or wave area size change.
Sizes should be relative:
Slits
Wave velocity (hard to do this with a local propagator)
Gun should shrink for small screen size (gun size is now incorrectly dependent on screen size).
Add separate control for box physical size (independent of lattice size)?
Center the Reset All and Clear Wave buttons.
>>This is difficult, given the "all control panels should have the same width" constraint.
Add a close button for the stopwatch.
Modelslider labels have a bug on mac.
Bug: In all panels, if you move the wavelength slider all the way to the left, funny little brown squares flash all over the wave:
>>I Can't reproduce this.
In Java version 1.4.2_03, the slider leaves a little trail when you move it
and has a green rectangle behind the tick marks (see below).
This problem does not appear in Java version 1.5.0_04 or 1.5.0_06.
>>Very difficult ( interface not exposed, internal bug in Java?).
Initial Wiggle me is hard to see:
If it were a font that is about twice as large, bold,
colored (red maybe) and without a bubble I think it’d be more prominent.
It should enter from the top of the screen and descend
down to the location that it sits in now.
>>Is this still a problem
In Two Lasers panel, the left gun controls go off the screen at 1024x768 resolution:
>>This problem has minimal visual impact, doesn't interfere with functionality and wouldn't be easy to fix. I will wait on solving this bug.