From: Rafael L. <rla...@us...> - 2005-01-30 20:51:05
|
* Alan W. Irwin <ir...@be...> [2005-01-30 10:18]: > Here are the requested screenshots. gv and xpdf were executed with no > special options and no fiddling with options inside the GUI except to turn > the magnification to 200 per cent in both cases. I also checked that > default options meant antialising was turned on in both cases. If I > deliberately turn it off in both cases, the result looks much worse (except > for xpdf where the lines in the embedded graphic remain the same, i.e. > antialiasing makes no difference to the lines in the plot for xpdf while it > does make a difference for gv). Attached below are two screenshots made in my system. I deliberately decreased the size of the windows to save bandwith. screenshot1.png is from gv while viewing the first page of test.ps with magnification of 200% and antialiasing on. The characters look fine for me. screenshot1.png is from xpdf while viewing part of the figure at 400%. At that magnification, we see the difference of line widths. At 200% the lines do seem to have the same width. Do you confirm these results in your system? > * Did you display test.ps and test.pdf on your Debian testing system or > Debian unstable system? It should make no difference for xpdf, but the gv > versions are quite different between the two. Unstable here. I have gv_3.6.1-4. > * Did you do the comparisons at 200 per cent magnification as in the > screenshots. See above. > * What is the resolution of your display? Mine is 1024x768. Its possible > the deficiencies of the gv rasterizer compared to that used by xpdf are not > so obvious if the resolution of your display is a lot higher than mine. Mine is also 1024x768. -- Rafael |