Re: [Geeqie-devel] Speed Comparisons

2009-08-26 Thread Vladimir Nadvornik
On Wednesday 26 August 2009 14:24:04 Arnd wrote: > Omari Stephens [08/26/2009 08:07 AM] > > > Arnd, were you running something >=r1799? > > No, the beta2 release from July 2nd. > > I tried the same test as before (1.2), and it is indeed faster, from > 10.7s (confirmed with the old beta2) down to 9.

Re: [Geeqie-devel] Speed Comparisons

2009-08-26 Thread Vladimir Nadvornik
On Wednesday 26 August 2009 12:46:28 Arnd wrote: > Vladimir Nadvornik [08/26/2009 10:55 AM] > > > Just a note that geeqie flushes the draw queue every 50ms, that means > > 20 updates if the loading takes 1s and 200 updates if it takes 10s. > > 20 updates where each update draws 1/20 of the image ar

Re: [Geeqie-devel] Speed Comparisons

2009-08-26 Thread Arnd
Omari Stephens [08/26/2009 08:07 AM] > Arnd, were you running something >=r1799? No, the beta2 release from July 2nd. I tried the same test as before (1.2), and it is indeed faster, from 10.7s (confirmed with the old beta2) down to 9.5s (svn-rev1824). Still slower than the others, but on the r

Re: [Geeqie-devel] Speed Comparisons

2009-08-26 Thread Arnd
Sabin Iacob [08/26/2009 01:00 AM] > test images, config and scripts, please :P (I am willing to try it on my > mighty Atom EeePC). Images & configs here: http://rapidshare.com/files/271693521/geeqiespeedtest.tar I didn't use a script, I just launched "time " 6 times, and took the average of th

Re: [Geeqie-devel] Speed Comparisons

2009-08-26 Thread Arnd
Vladimir Nadvornik [08/26/2009 10:55 AM] > Just a note that geeqie flushes the draw queue every 50ms, that means > 20 updates if the loading takes 1s and 200 updates if it takes 10s. > 20 updates where each update draws 1/20 of the image area is faster than > 200 updates by 1/200 of the area. > >

Re: [Geeqie-devel] Speed Comparisons

2009-08-26 Thread Vladimir Nadvornik
Hi Arnd, Thanks for the testing. On út 25. srpna 2009, Arnd wrote: > Hi! > > Thank you for continuing development of Geeqie. I am a long time GQview > user. I use some other viewers as well. > > Here are some speed comparisons which might be interesting. I've done > them on an old P3-600 single C

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Omari Stephens
Vladimir Nadvornik wrote: > On Tuesday 25 August 2009 20:33:14 Omari Stephens wrote: >> [NOTE: I just realized that my tree is patched with the multi-threaded >> thumbnail loaders patch. I'm not sure what this may have changed.] >> >> So, the vast majority of the time is spent in: >> static void v

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Omari Stephens
Sabin Iacob wrote: ::snip? SNIP!:: > P.S. is it me, or do Nikon NEF files really render faster than JPEGs? :-/ They should render as fast as JPEGs of the same quality. What you see is actually an embedded (iirc) ~75% quality JPEG preview that lives in every NEF. It's not actually processing the

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Sabin Iacob
Arnd wrote: > Hi! > > Thank you for continuing development of Geeqie. I am a long time GQview > user. I use some other viewers as well. > > Here are some speed comparisons which might be interesting. I've done > them on an old P3-600 single CPU computer, which demonstrates the > differences (abs

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Vladimir Nadvornik
On Tuesday 25 August 2009 20:33:14 Omari Stephens wrote: > [NOTE: I just realized that my tree is patched with the multi-threaded > thumbnail loaders patch. I'm not sure what this may have changed.] > > So, the vast majority of the time is spent in: > static void vflist_name_cell_data_cb(...) > wh

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Omari Stephens
[NOTE: I just realized that my tree is patched with the multi-threaded thumbnail loaders patch. I'm not sure what this may have changed.] So, the vast majority of the time is spent in: static void vflist_name_cell_data_cb(...) which lives around line 1771 of view_file_list.c The backtrace gene

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Laurent Monin
Klaus Ethgen a écrit : > Hi, > Hi Klaus, > Am Di den 25. Aug 2009 um 16:18 schrieb Laurent Monin: > > Thanks for the numbers ! > > Yea, thanks. It shows in numbers what I just felt. > > > It would be interesting to have such numbers for recent hardware too (n > > CPUs/more RAM). > > I think it woul

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Klaus Ethgen
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Hi, Am Di den 25. Aug 2009 um 16:18 schrieb Laurent Monin: > Thanks for the numbers ! Yea, thanks. It shows in numbers what I just felt. > It would be interesting to have such numbers for recent hardware too (n > CPUs/more RAM). I think it would

Re: [Geeqie-devel] Speed Comparisons

2009-08-25 Thread Laurent Monin
Arnd a écrit : > Hi! > > Thank you for continuing development of Geeqie. I am a long time GQview > user. I use some other viewers as well. > > Here are some speed comparisons which might be interesting. I've done > them on an old P3-600 single CPU computer, which demonstrates the > differences (

[Geeqie-devel] Speed Comparisons

2009-08-25 Thread Arnd
Hi! Thank you for continuing development of Geeqie. I am a long time GQview user. I use some other viewers as well. Here are some speed comparisons which might be interesting. I've done them on an old P3-600 single CPU computer, which demonstrates the differences (absolute values are not signi