Plotly.py: slight change in surface opacity changes image completely

[paddyroddy]

I need to adjust the opacity of a surface, so I can overlay two surfaces on top of each other (not sure how else to do that). But on a very small change from 1 to slightly below the look of the plot changes a lot.

opacity=1

opacity=0.99999

import numpy as np
from plotly.graph_objs import Figure, Surface

x = np.array(
    [
        [0, 0, 0, 0, 0, 0, 0, 0, 0],
        [
            7.07106781e-01,
            0.5,
            4.32978028e-17,
            -0.5,
            -7.07106781e-01,
            -0.5,
            -1.29893408e-16,
            0.5,
            7.07106781e-01,
        ],
        [
            1,
            7.07106781e-01,
            6.12323400e-17,
            -7.07106781e-01,
            -1,
            -7.07106781e-01,
            -1.83697020e-16,
            7.07106781e-01,
            1,
        ],
        [
            7.07106781e-01,
            0.5,
            4.32978028e-17,
            -0.5,
            -7.07106781e-01,
            -0.5,
            -1.29893408e-16,
            0.5,
            7.07106781e-01,
        ],
        [
            1.22464680e-16,
            8.65956056e-17,
            7.49879891e-33,
            -8.65956056e-17,
            -1.22464680e-16,
            -8.65956056e-17,
            -2.24963967e-32,
            8.65956056e-17,
            1.22464680e-16,
        ],
    ]
)
y = np.array(
    [
        [0, 0, 0, 0, 0, -0, -0, -0, 0],
        [0, 0.5, 7.07106781e-01, 0.5, 8.65956056e-17, -0.5, -7.07106781e-01, -0.5, 0],
        [
            0,
            7.07106781e-01,
            1,
            7.07106781e-01,
            1.22464680e-16,
            -7.07106781e-01,
            -1,
            -7.07106781e-01,
            0,
        ],
        [0, 0.5, 7.07106781e-01, 0.5, 8.65956056e-17, -0.5, -7.07106781e-01, -0.5, 0],
        [
            0,
            8.65956056e-17,
            1.22464680e-16,
            8.65956056e-17,
            1.49975978e-32,
            -8.65956056e-17,
            -1.22464680e-16,
            -8.65956056e-17,
            0,
        ],
    ]
)
z = np.array(
    [
        [1, 1, 1, 1, 1, 1, 1, 1, 1],
        [
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
            7.07106781e-01,
        ],
        [
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
            6.12323400e-17,
        ],
        [
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
            -7.07106781e-01,
        ],
        [-1, -1, -1, -1, -1, -1, -1, -1, -1],
    ]
)
data = [Surface(x=x, y=y, z=z, surfacecolor=np.ones(x.shape), opacity=0.99999)]
fig = Figure(data=data)
fig.show()

sorry about the numbers, they've been generated by some external library

my actual use case is like this where there are weird lines appearing on the plot by setting opacity

[nicolaskruchten]

@archmoj can you comment please?

[archmoj]

@paddyroddy could you please set the opacity to 0.25 or 0.5 and share new screenshots?

[paddyroddy]

opacity=0.5

and in my more complicated use-case with opacity=0.5

[archmoj]

Here is a demo in codepen.
Wondering if you tried using mesh3d instead of surface plot for that kind of geometry?

[archmoj]

This could possibly be fixed/improved by https://github.com/plotly/plotly.js/pull/4643.
But at the moment I don't see that PR being merged in near feature.

[paddyroddy]

Here is a demo in codepen.
Wondering if you tried using mesh3d instead of surface plot for that kind of geometry?

thanks, no I haven't. My actual data is spherical (the example was low resolution) and isn't one uniform colour I have a surfacecolor passed to it. So not sure whether mesh3d would achieve the same thing