when the magnitude of the point fluctuates up and down in each year the magnitude is not consistent for the same year

[tropicaldev]

In the original example (globe folder) the magnitudes are constantly increasing over the years so the magnitudes are rendered correctly

with magnitude fluctuating the magnitude is shown incorrectly ( 1992 retains the same amplitude as 1991)

[
["1990",[0,0,0.02]]
,["1991",[0,0,0.05]]
,["1992",[0,0,0.02]]
]

magnitudes are rendered correctly with magnitudes constantly increasing

[
["1990",[0,0,0.02]]
,["1991",[0,0,0.05]]
,["1992",[0,0,0.07]]
]

[ezequias]

OK @tropicaldev I will check that but don't you think is's a strange behavior?

[tropicaldev]

if you are asking me if its strange behavior populatin decreasing...yes its strange.
if you are asking if something else is strange...i think its a bug :)

if there was country (maybe there isn't in the globe folder example) that the population decreased from one year to another and then increased again (or vice versa) the magnitude would not be shown correctly

for easy confirmation just replace the json file with
[
["1990",[0,0,0.02]]
,["1991",[0,0,0.05]]
,["1992",[0,0,0.02]]
]

[ezequias]

@tropicaldev I meant for multiple purposes. I know population decreasing is quite difficult but I am using for other purpose.

Anyway, I just check out another version and this json (increasing and decreasing) is working well. Not bad.

See my json:

[
[
"2013",
[
-22.738889,
-47.330833,
0.002,
-10.91,
-37.07,
0.002
]
],
[
"2014",
[
-22.738881,
-47.330832,
0.010,
-10.910001,
-37.070004,
0.020
]
],
[
"2015",
[
-22.738882,
-47.330835,
0.030,
-10.910004,
-37.070005,
0.002
]
]
]

[tropicaldev]

I just tested with your json and the amplitude for the 2nd point in 2015 is not the same as 2013 (it should be 0.002 for both years)...it's retaining the 2014 value for 2015

I tested the json with the examples in the "master" download

[BuiltInParris]

I have also noticed this problem. Just wanted to throw in my support for this being fixed. I had the same issue as @tropicaldev. It renders fine until you tell it to go from a larger number to a smaller one. Since this has use cases other than just increasing population density over time, this definitely needs to be fixed.

@ezequias your case only "appears" to work. What happens is the data moves upward, then on the second switch it moves upward THEN downward back to its original position in the second state.

[BuiltInParris]

It appears it may just only go down to a certain threshold. An example of this can be seen in the diabetes example here: http://www.creativedigitalgallery.org/ImpactofDiabetes/index.html.
In the middle of Europe, there is a data point that goes down then pops back up to the same position again. http://prntscr.com/93xg8r