Yes. Working fine now and I agree that the diff between cairo and gdi32 is no longer very noticeable.
Very nice!! On Wed, Mar 16, 2011 at 6:17 PM, bill lam <[email protected]> wrote: > Thanks Ric, fixed and committed for pacman. > > The new dll should now work for both cairo and gdi32 backend under windows. > > The apple-for-apple comparison between cairo and gdi32 under window shows > the performance of cairo is not that bad. > > Срд, 16 Мар 2011, Ric Tikkanz писал(а): >> The problem seems to be the old spaces-in-the-path problem. >> LIBGLCMDS_jglcanvas_ needs to be dquoted in case it contains spaces. >> >> On Tue, Mar 15, 2011 at 8:22 AM, Ric Sherlock <[email protected]> wrote: >> > Will try on Windows again tonight. Meantime it is working well in >> > 64-bit Linux - very nice to have responsive plots again! >> > >> > >> > On Tue, Mar 15, 2011 at 3:04 AM, bill lam <[email protected]> wrote: >> >> I got no real window, only tested on kvm and wine. What did the cder'' >> >> tell? >> >> What is the value of LIBGLCMDS_jglcanvas_? Will it be a permission >> >> probllem? >> >> >> >> Втр, 15 Мар 2011, Ric Tikkanz писал(а): >> >>> Sounds great, but doesn't run for me on 32-bit Win7 :( >> >>> >> >>> load 'plot' >> >>> plot i. 3 4 >> >>> |domain error: cd >> >>> | (LIBGLCMDS,' Glcmds_gdi32 > + i x *x *x *c *x x')&cd >> >>> HDC;xpar;ipar;(utf8 PROFONT_jgl2_);(,y);#y >> >>> callback run error: expose_event_3_ >> >>> |domain error: cd >> >>> | (LIBGLCMDS,' Glcmds_gdi32 > + i x *x *x *c *x x')&cd >> >>> HDC;xpar;ipar;(utf8 PROFONT_jgl2_);(,y);#y >> >>> >> >>> JVERSION >> >>> Engine: j701/2011-01-10/11:25 >> >>> Library: 7.01.048 >> >>> Platform: Win 32 >> >>> Installer: j701a_win.exe >> >>> InstallPath: c:/program files/j701 >> >>> load 'pacman' >> >>> 'search' jpkgv 'plot gtk gl2 gdi' >> >>> ┌──────────────────┬──────┬──────┬────────────┐ >> >>> │api/gdi32 │1.0.0 │1.0.0 │gdi32 api │ >> >>> │api/gdiplus │1.0.0 │1.0.0 │gdiplus api │ >> >>> │api/gtkinclude │1.0.0 │1.0.0 │GTK C Header│ >> >>> │convert/gtkdocdict│1.0.0 │1.0.0 │gtkdocdict │ >> >>> │convert/gtkdoctag │1.0.0 │1.0.0 │gtkdoctag │ >> >>> │demos/gtk │1.0.16│1.0.16│gtk demos │ >> >>> │demos/plot │1.0.3 │1.0.3 │Plot demos │ >> >>> │docs/gtk │ │1.0.2 │GTK API │ >> >>> │graphics/gl2 │1.0.7 │1.0.7 │gl2 │ >> >>> │graphics/plot │1.0.9 │1.0.9 │Plot │ >> >>> │gui/gtk │1.0.61│1.0.61│GTK API │ >> >>> │ide/gtk │1.0.69│1.0.69│GTK IDE │ >> >>> └──────────────────┴──────┴──────┴────────────┘ >> >>> >> >>> >> >>> On Tue, Mar 15, 2011 at 12:04 AM, bill lam <[email protected]> wrote: >> >>> > I added glcmds shared library for linux cairo and windows gdi32, the >> >>> > speed >> >>> > can now catch up with J602 wd. The suface plot timing, >> >>> > windows(under kvm) first run 0.921s subsequent runs 0.078 >> >>> > linux first run 0.393s subsequent runs 0.303 >> >>> > >> >>> > not sure why windows can become much faster in the 2nd run. >> >>> > >> >>> > Please upgrade those graphics packages. However you are not encouraged >> >>> > to >> >>> > use that glcmds dll/so for writing new codes. >> >>> > >> >>> > Срд, 09 Мар 2011, bill lam писал(а): >> >>> >> If the 0.11377 second is timed on Mac + jwd, then it suggested there >> >>> >> is >> >>> >> room of improvement. >> >>> >> >> >>> >> Втр, 08 Мар 2011, Joey K Tuttle писал(а): >> >>> >> > At 8:38 PM -0800 11/03/08, Joey K Tuttle wrote: >> >>> >> > > >> >>> >> > >I can now to the sine plot, it takes noticeably longer (more than 3 >> >>> >> > >times as long) than j602 producing the plot.pdf file. The window is >> >>> >> > >VERY slow to resize (using 100% of one of the 4 available cores) - >> >>> >> > >much longer than "zoom" on the pdf plot and making the window >> >>> >> > >bigger. >> >>> >> > > >> >>> >> > >> >>> >> > When I said "more than 3 times" as long - I hadn't measured it. To >> >>> >> > do - >> >>> >> > >> >>> >> > 'surface' plot +/~ 1 o. i: 6j99 >> >>> >> > >> >>> >> > Takes 56 times as much cpu time in j701 as in j602 - 6.4631 seconds >> >>> >> > versus 0.11377 seconds... This doesn't include all the mouse >> >>> >> > clicking to free the jconsole session again. >> >>> >> > >> >>> >> > But at least I can plot something and get my session back (although >> >>> >> > I >> >>> >> > get to "live with" the orphan process...). >> >>> >> > ---------------------------------------------------------------------- >> >>> >> > For information about J forums see >> >>> >> > http://www.jsoftware.com/forums.htm >> >>> >> >> >>> >> -- >> >>> >> regards, >> >>> >> ==================================================== >> >>> >> GPG key 1024D/4434BAB3 2008-08-24 >> >>> >> gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 >> >>> > >> >>> > -- >> >>> > regards, >> >>> > ==================================================== >> >>> > GPG key 1024D/4434BAB3 2008-08-24 >> >>> > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 >> >>> > ---------------------------------------------------------------------- >> >>> > For information about J forums see http://www.jsoftware.com/forums.htm >> >>> ---------------------------------------------------------------------- >> >>> For information about J forums see http://www.jsoftware.com/forums.htm >> >> >> >> -- >> >> regards, >> >> ==================================================== >> >> GPG key 1024D/4434BAB3 2008-08-24 >> >> gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 >> >> ---------------------------------------------------------------------- >> >> For information about J forums see http://www.jsoftware.com/forums.htm >> > >> ---------------------------------------------------------------------- >> For information about J forums see http://www.jsoftware.com/forums.htm > > -- > regards, > ==================================================== > GPG key 1024D/4434BAB3 2008-08-24 > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
