With matplotlib 1.4.3 it is possible to draw at least a random time series of 70,000 points on the Qt4Agg backend:
$ python -c 'import numpy as np; from matplotlib import pyplot as plt; plt.plot(np.random.rand(70000)); plt.show()'
With the current HEAD, an OverflowError is thrown:
OverflowError: In draw_path: Exceeded cell block limit
Perhaps it's just a transient regression, but I didn't see it mentioned anywhere.
This sounds very familiar, but I could have sworn it was something to do with the macosx backend and old limitations of quartz.
On Tue, May 5, 2015 at 1:58 PM, Antony Lee notifications@github.com wrote:
With matplotlib 1.4.3 it is possible to draw at least a random time series of 70,000 points on the Qt4Agg backend: $ python -c 'import numpy as np; from matplotlib import pyplot as plt; plt.plot(np.random.rand(70000)); plt.show()' With the current HEAD, an OverflowError is thrown: OverflowError: In draw_path: Exceeded cell block limit Perhaps it's just a transient regression, but I didn't see it mentioned anywhere. — Reply to this email directly or view it on GitHub #4407.
— Reply to this email directly or view it on GitHub #4407.
It seems odd that 1.4.3 and master are behaving differently; I think that with path simplification, both should be able to handle the test case.
Do you have any rcParams in effect (in your matplotlibrc for example)?
This may have to do with "upgrading" to Agg-svn from Agg-2.4.
I thought both would pick up the same matplotlibrc anyways but just to be sure I removed it, and see the same difference.
See #4408.
Activity
WeatherGod commented on May 5, 2015
This sounds very familiar, but I could have sworn it was something to do
with the macosx backend and old limitations of quartz.
On Tue, May 5, 2015 at 1:58 PM, Antony Lee notifications@github.com wrote:
efiring commented on May 5, 2015
It seems odd that 1.4.3 and master are behaving differently; I think that with path simplification, both should be able to handle the test case.
mdboom commented on May 5, 2015
Do you have any rcParams in effect (in your matplotlibrc for example)?
mdboom commented on May 5, 2015
This may have to do with "upgrading" to Agg-svn from Agg-2.4.
anntzer commented on May 5, 2015
I thought both would pick up the same matplotlibrc anyways but just to be sure I removed it, and see the same difference.
mdboom commented on May 5, 2015
See #4408.