Hmm, I guess we can't win. Maybe we should ask about this on the matplotlib mailing list.
You can see from those two examples that the math does look better with usetex. For example, notice how the > in |q0> is misaligned with the matplotlib rendering. Basically, any math that's more than one character is going to be off in some way with the matplotlib rendering, and it gets worse the more complicated the math is. And as you noticed, some math, such as matrices, is not supported at all. So ask on the matplotlib mailing list about this, and point us to the thread. I doubt they will be able to make the matplotlib math rendering better, but they might be able to make the bounding boxes around the usetex math a little bigger. (btw, the thing that looks *really* bad in all the examples are the off center vertical lines. Any idea how to fix that?) Aaron Meurer On Mon, Jul 29, 2013 at 10:40 PM, Ondřej Čertík <[email protected]> wrote: > On Mon, Jul 29, 2013 at 7:42 PM, Rick Muller <[email protected]> wrote: >> This is directed mostly at Aaron, but it's here for everyone in case anyone >> wants to comment. >> >> In a recent merge thread, Aaron said (regarding the use of >> "rc('text',usetex=True)"): >> >>> You probably should just do this automatically in circuitplot, since >>> matplotlib's latex is so bad anyway. Then you could probably even print >>> those little measurement icons using LaTeX itself. Maybe there is some LaTeX >>> library that includes it? >> >> >> The problem with this is that it degrades the quality of the plots, in my >> opinion. >> >> Here's an example using the normal flags. >> >> Here's an example using the rc('text',usetex=True). >> >> I think the box around H is sized better, and the fonts print a little >> thicker, in the version without the usetex flags. Any thoughts as to how >> this impacts the suggestion of always using the flags? > > I've noticed Aaron comment in the PR. However, I do agree with you Rick, > I also like the "normal flags" version better. > > I did sometimes notice that when the formula is more complicated, that > the latex version might look a little bit better. > > Ondrej > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sympy. > For more options, visit https://groups.google.com/groups/opt_out. > > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. For more options, visit https://groups.google.com/groups/opt_out.
