These are chat archives for symengine/symengine

Apr 2016
Ralf Stephan
Apr 25 2016 03:13
Too much copying necessary?
Shivam Vats
Apr 25 2016 12:53
@isuruf The benchmarks @irislq posted, showed our implementation as being slower. The Newton iteration based expansion algorithms we use are quite fast (though not the fastest), which made me wonder if Piranha uses some better algos. I wasn't aware it doesn't have series expansion.
Isuru Fernando
Apr 25 2016 13:27
@shivamvats, all series expansion code uses same code for UPSeriesPiranha and UnivariateSeries. Speed difference should be because of the speed difference in low level methods
Charles Chen
Apr 25 2016 18:05
@isuruf I see that you have OpenMP constructs in dense_matrix.cpp. How do you configure it so compiler will recognize it?