plus startup speedup
9 views
Skip to first unread message
edgar....@web.de
May 3, 2012, 5:30:15 PM5/3/12
to openjum...@googlegroups.com
i just committed some changes to trunk and am under the impression that the latest PLUS snapshot startup is again much faster especially on network shares.
especially people having OJ installed on network shares might profit from these changes, but everybody else should also notice a significant speed up. i might have broken some things, so please download the latest PLUS snapshot e.g.
http://sourceforge.net/projects/jump-pilot/files/OpenJUMP_snapshots/OpenJUMP-20120503-r2851-PLUS.zip/download
and come back with remarks on performance and stability please.
thanks ede
Eric Grosso
May 4, 2012, 9:15:52 AM5/4/12
to openjum...@googlegroups.com
Hi Ede,
Startup time divided by two under a Linux/Debian 64-bits environment, around 7-8 seconds (before 14-15 seconds -- OpenJUMP-1.5.1-r2721-PLUS version --).
No particular problem of stability have been notice.
Eric
Startup time divided by two under a Linux/Debian 64-bits environment, around 7-8 seconds (before 14-15 seconds -- OpenJUMP-1.5.1-r2721-PLUS version --).
No particular problem of stability have been notice.
Eric
-- You received this message because you are subscribed to the Google Groups openjump-users group. To post to this group, send email to openjum...@googlegroups.com. To unsubscribe from this group, send email to openjump-user...@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/openjump-users?hl=en
andre tanoe
May 6, 2012, 10:01:23 AM5/6/12
to openjum...@googlegroups.com
Helo Ede,
I found out that during editing, its a little bit slower (probably about less than a second) than v 2721. when I want to cut or add a polygon, i wait for maybe 0.5 second longer.
Everything else seems fine. The start up is definitely faster than v 2721.
thanks.
I found out that during editing, its a little bit slower (probably about less than a second) than v 2721. when I want to cut or add a polygon, i wait for maybe 0.5 second longer.
Everything else seems fine. The start up is definitely faster than v 2721.
thanks.
edgar....@web.de May 03 11:30PM +0200
edgar....@web.de
May 6, 2012, 10:17:35 AM5/6/12
to openjum...@googlegroups.com
How do measure editing speed or even notice the difference of less than a second?
In case that's not due to momentary lags in your operating system, meaning consistently reproducible when running both revisions multiple times, could you give us a step by step routine to reproduce it?
If it's a problem it was most likely introduced before 2851 as the changes in the PluginManager most likely won't affect usage performance in any way except of breaking some extension completely ;).
thanks for the reply and come back with more details if your issue persists in newer versions. ..ede
On 06.05.2012 16:01, andre tanoe wrote:
> Helo Ede,
>
> I found out that during editing, its a little bit slower (probably about
> less than a second) than v 2721. when I want to cut or add a polygon, i
> wait for maybe 0.5 second longer.
> Everything else seems fine. The start up is definitely faster than v 2721.
>
> thanks.
>
>
>
>> plus startup speedup<http://groups.google.com/group/openjump-users/t/48648ba1321f5899>
In case that's not due to momentary lags in your operating system, meaning consistently reproducible when running both revisions multiple times, could you give us a step by step routine to reproduce it?
If it's a problem it was most likely introduced before 2851 as the changes in the PluginManager most likely won't affect usage performance in any way except of breaking some extension completely ;).
thanks for the reply and come back with more details if your issue persists in newer versions. ..ede
On 06.05.2012 16:01, andre tanoe wrote:
> Helo Ede,
>
> I found out that during editing, its a little bit slower (probably about
> less than a second) than v 2721. when I want to cut or add a polygon, i
> wait for maybe 0.5 second longer.
> Everything else seems fine. The start up is definitely faster than v 2721.
>
> thanks.
>
>
>
>>
>> edgar....@web.de May 03 11:30PM +0200
>>
>> i just committed some changes to trunk and am under the impression
>> that the latest PLUS snapshot startup is again much faster especially on
>> network shares.
>>
>> especially people having OJ installed on network shares might profit
>> from these changes, but everybody else should also notice a significant
>> speed up. i might have broken some things, so please download the latest
>> PLUS snapshot e.g.
>>
>> http://sourceforge.net/projects/jump-pilot/files/OpenJUMP_snapshots/OpenJUMP-20120503-r2851-PLUS.zip/download
>> and come back with remarks on performance and stability please.
>>
>> thanks ede
>>
>>
>>
>>
>
>> edgar....@web.de May 03 11:30PM +0200
>>
>> i just committed some changes to trunk and am under the impression
>> that the latest PLUS snapshot startup is again much faster especially on
>> network shares.
>>
>> especially people having OJ installed on network shares might profit
>> from these changes, but everybody else should also notice a significant
>> speed up. i might have broken some things, so please download the latest
>> PLUS snapshot e.g.
>>
>> http://sourceforge.net/projects/jump-pilot/files/OpenJUMP_snapshots/OpenJUMP-20120503-r2851-PLUS.zip/download
>> and come back with remarks on performance and stability please.
>>
>> thanks ede
>>
>>
>>
>>
>
andre tanoe
May 6, 2012, 10:22:32 PM5/6/12
to openjum...@googlegroups.com
Helo Ede,
I open several shape files contains points, lines, and polygons. Then
I tried to edit my polygons by using the "create cookie cut" and "auto
complete polygon". I opened the same shapefiles with v2721 and 2851.
I notice that there is a delay in 2851 longer than 2721 when I do
editing to my shapefiles. because everytime after i click my mouse to
end the editing, i have to wait longer (i feel that it is not more
than a second, probably wrong)
This delay was not counted by clock, but I can feel it, that after i
press the mouse, i have to wait a little bit longer (momentary lag).
It is correct that even the 2721 has this momentary lag, but the 2851
gives a little bit longer.
I dont think the operating system cause it since i use the same
computer. I am using windows 7 with 6.31 java.
I think this delay is not really important, because its based on my
feeling rather than counting the clock.
i am still happy using it, especially with the faster start up.
thank you for the improvement.
I open several shape files contains points, lines, and polygons. Then
I tried to edit my polygons by using the "create cookie cut" and "auto
complete polygon". I opened the same shapefiles with v2721 and 2851.
I notice that there is a delay in 2851 longer than 2721 when I do
editing to my shapefiles. because everytime after i click my mouse to
end the editing, i have to wait longer (i feel that it is not more
than a second, probably wrong)
This delay was not counted by clock, but I can feel it, that after i
press the mouse, i have to wait a little bit longer (momentary lag).
It is correct that even the 2721 has this momentary lag, but the 2851
gives a little bit longer.
I dont think the operating system cause it since i use the same
computer. I am using windows 7 with 6.31 java.
I think this delay is not really important, because its based on my
feeling rather than counting the clock.
i am still happy using it, especially with the faster start up.
thank you for the improvement.
Michaël Michaud
May 7, 2012, 2:37:02 AM5/7/12
to openjum...@googlegroups.com
Hi Andre,
To test it, I used a shapefile with overcomplicated polygons and tried
both tools
"create cookie cut" and "auto complete polygon" on the same object.
It lasted more than 5 s in both cases (no difference easily measurable).
Note that the speed of these tools are probably dependent on
1) polygon complexity (overlay calculation on this polygon)
2) shapefile size (screen refresh at the end of operation)
3) total amount of loaded data compared with max memory available
(garbage collection activity is higher and pauses are more frequent if
you're near the max)
Micha�l
>
> This delay was not counted by clock, but I can feel it, that after i
> press the mouse, i have to wait a little bit longer (momentary lag).
> It is correct that even the 2721 has this momentary lag, but the 2851
> gives a little bit longer.
> I dont think the operating system cause it since i use the same
> computer. I am using windows 7 with 6.31 java.
>
> I think this delay is not really important, because its based on my
> feeling rather than counting the clock.
> i am still happy using it, especially with the faster start up.
>
> thank you for the improvement.
>
>
>> From: edgar....@web.de
>> Date: May 06 04:17PM +0200
>> Url: http://groups.google.com/group/openjump-users/msg/9a613531c0e024ed
>>
>> How do measure editing speed or even notice the difference of less than a
>> second?
>>
>> In case that's not due to momentary lags in your operating system, meaning
>> consistently reproducible when running both revisions multiple times, could
>> you give us a step by step routine to reproduce it?
>>
>> If it's a problem it was most likely introduced before 2851 as the changes
>> in the PluginManager most likely won't affect usage performance in any way
>> except of breaking some extension completely ;).
>>
>> thanks for the reply and come back with more details if your issue persists
>> in newer versions. ..ede
>>
>>
> I open several shape files contains points, lines, and polygons. Then
> I tried to edit my polygons by using the "create cookie cut" and "auto
> complete polygon". I opened the same shapefiles with v2721 and 2851.
> I notice that there is a delay in 2851 longer than 2721 when I do
> editing to my shapefiles. because everytime after i click my mouse to
> end the editing, i have to wait longer (i feel that it is not more
> than a second, probably wrong)
I could not measure this slowdown.
> I tried to edit my polygons by using the "create cookie cut" and "auto
> complete polygon". I opened the same shapefiles with v2721 and 2851.
> I notice that there is a delay in 2851 longer than 2721 when I do
> editing to my shapefiles. because everytime after i click my mouse to
> end the editing, i have to wait longer (i feel that it is not more
> than a second, probably wrong)
To test it, I used a shapefile with overcomplicated polygons and tried
both tools
"create cookie cut" and "auto complete polygon" on the same object.
It lasted more than 5 s in both cases (no difference easily measurable).
Note that the speed of these tools are probably dependent on
1) polygon complexity (overlay calculation on this polygon)
2) shapefile size (screen refresh at the end of operation)
3) total amount of loaded data compared with max memory available
(garbage collection activity is higher and pauses are more frequent if
you're near the max)
Micha�l
>
> This delay was not counted by clock, but I can feel it, that after i
> press the mouse, i have to wait a little bit longer (momentary lag).
> It is correct that even the 2721 has this momentary lag, but the 2851
> gives a little bit longer.
> I dont think the operating system cause it since i use the same
> computer. I am using windows 7 with 6.31 java.
>
> I think this delay is not really important, because its based on my
> feeling rather than counting the clock.
> i am still happy using it, especially with the faster start up.
>
> thank you for the improvement.
>
>
>> From: edgar....@web.de
>> Date: May 06 04:17PM +0200
>> Url: http://groups.google.com/group/openjump-users/msg/9a613531c0e024ed
>>
>> How do measure editing speed or even notice the difference of less than a
>> second?
>>
>> In case that's not due to momentary lags in your operating system, meaning
>> consistently reproducible when running both revisions multiple times, could
>> you give us a step by step routine to reproduce it?
>>
>> If it's a problem it was most likely introduced before 2851 as the changes
>> in the PluginManager most likely won't affect usage performance in any way
>> except of breaking some extension completely ;).
>>
>> thanks for the reply and come back with more details if your issue persists
>> in newer versions. ..ede
>>
>>
Reply all
Reply to author
Forward
0 new messages