Nabble has removed Mailing-list integration.
Posts created here DO NOT GET SENT TO THE MAILING LIST.
Mailing-list emails DO NOT GET POSTED TO THE FORUM.
So basically the Forum is now out of date, we are looking into migrating the history.

For now you should send emails, people will see them, discuss@lists.openscad.org.

OpenSCAD release candidate: 2020.12-RC2

classic Classic list List threaded Threaded
33 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

Troberg
thehans wrote

> If you can't provide an example to reproduce it then there's not a lot of
> hope in fixing it.
>
> Before troubleshooting further, are you sure you didn't run out of RAM and
> hit swap space when the RC was run?  That could certainly cause such a
> slowdown.
> If you ran the previous release first(assuming it used a significant
> portion of your RAM) and left it open while testing the RC, I could see
> that possibly happening.
>
> If that's not the cause, then I would recommend trying to single out which
> part(s) of the design are taking much longer on RC, by systematically
> using
> disable modifiers * and/or a root modifier ! on different places in the
> script.
> https://en.wikibooks.org/wiki/OpenSCAD_User_Manual/Modifier_Characters#Disable_Modifier
>
> To speed up the process, you can try reducing $fn (or increasing $fs and
> $fa) wherever used.
> Also if there are loops of geometries involved, try reducing their loop
> count significantly.
> If it's possible  to reduce overall complexity until the previous release
> takes say 1 second to preview, then if the RC is really still 15x-20x
> slower in this case,  at least you should only have to wait 15-20 seconds,
> not 5 minutes, in between trying out disable modifiers etc.

I can provide an example, but it's a 10000 line project, libraries and all,
so it wouldn't be very useful.

Definitely didn't run out of RAM, I have 32 GB, and there was plenty to
spare.

As for simplifying, it's when I put everything together that stuff just
grinds to a halt. Hard to simplify...

No big loops, can't really simplify using $fn and the others, as I need the
precision I have. Either way, it shouldn't behave so differently.



--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

thehans

On Wed, Dec 9, 2020 at 5:59 AM Troberg <[hidden email]> wrote:
No big loops, can't really simplify using $fn and the others, as I need the
precision I have. Either way, it shouldn't behave so differently.

 I'm talking about doing things to help troubleshoot the issue, this would be temporary, not permanent changes to your project.
ie: make a complete copy of your project if you need to, then try making big drastic changes to narrow down where the slowdown comes from.  
For this purpose it doesn't matter if the rendered result looks absolutely nothing like your original design, as long as it helps towards a reproducible example of where RC is much slower than previous release.

If you don't want to bother troubleshooting anything then whatever, just don't expect anyone to bother fixing it.


_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

Troberg
thehans wrote
> On Wed, Dec 9, 2020 at 5:59 AM Troberg &lt;

> troberg.anders@

> &gt; wrote:
>
>> No big loops, can't really simplify using $fn and the others, as I need
>> the
>> precision I have. Either way, it shouldn't behave so differently.
>>
>
>  I'm talking about doing things to help troubleshoot the issue, this would
> be temporary, not permanent changes to your project.
> ie: make a complete copy of your project if you need to, then try making
> big drastic changes to narrow down where the slowdown comes from.
> For this purpose it doesn't matter if the rendered result looks absolutely
> nothing like your original design, as long as it helps towards a
> reproducible example of where RC is much slower than previous release.
>
> If you don't want to bother troubleshooting anything then whatever, just
> don't expect anyone to bother fixing it.

No need to be snarky about it, I've been working with software development
for 30 years. I know that my description isn't much help in finding the
problem.

My thought was that if I mention the problem, other people will watch out
for the problem, and hopefully, they'll run into it in a cleaner example
which means that I don't have to gut a 10 000 lines project (which will be
exceedingly tedious work, as every test will likely take five minutes...).



--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

nophead
Here is an example where the RC is 3 times slower than the release. https://github.com/nophead/NopSCADlib/blob/master/tests/pulleys.scad

OpenSCAD 2019.05
https://www.openscad.org/

Copyright (C) 2009-2019 The OpenSCAD Developers

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

Loaded design 'W:/CHRIS/Projects/RepRaptor/OpenSCADLibs/NopSCADlib/tests/pulleys.scad'.

Used file cache size: 18 files

Compiling design (CSG Tree generation)...

Compiling design (CSG Products generation)...

Geometries in cache: 259

Geometry cache size in bytes: 7707888

CGAL Polyhedrons in cache: 9

CGAL cache size in bytes: 5571112

Compiling design (CSG Products normalization)...

Normalized CSG tree has 36 elements

Compile and preview finished.

Total rendering time: 0 hours, 0 minutes, 4 seconds


Loaded design 'W:/CHRIS/Projects/RepRaptor/OpenSCADLibs/NopSCADlib/tests/pulleys.scad'.
OpenSCAD 2020.12-RC2
https://www.openscad.org/

Copyright (C) 2009-2020 The OpenSCAD Developers

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

Used file cache size: 18 files
Compiling design (CSG Tree generation)...
Compiling design (CSG Products generation)...
Geometries in cache: 245
Geometry cache size in bytes: 5961344
CGAL Polyhedrons in cache: 23
CGAL cache size in bytes: 21865784
Compiling design (CSG Products normalization)...
Normalized tree has 36 elements!
Compile and preview finished.
Total rendering time: 0:00:12.699


On Wed, 9 Dec 2020 at 16:48, Troberg <[hidden email]> wrote:
thehans wrote
> On Wed, Dec 9, 2020 at 5:59 AM Troberg &lt;

> troberg.anders@

> &gt; wrote:
>
>> No big loops, can't really simplify using $fn and the others, as I need
>> the
>> precision I have. Either way, it shouldn't behave so differently.
>>
>
>  I'm talking about doing things to help troubleshoot the issue, this would
> be temporary, not permanent changes to your project.
> ie: make a complete copy of your project if you need to, then try making
> big drastic changes to narrow down where the slowdown comes from.
> For this purpose it doesn't matter if the rendered result looks absolutely
> nothing like your original design, as long as it helps towards a
> reproducible example of where RC is much slower than previous release.
>
> If you don't want to bother troubleshooting anything then whatever, just
> don't expect anyone to bother fixing it.

No need to be snarky about it, I've been working with software development
for 30 years. I know that my description isn't much help in finding the
problem.

My thought was that if I mention the problem, other people will watch out
for the problem, and hopefully, they'll run into it in a cleaner example
which means that I don't have to gut a 10 000 lines project (which will be
exceedingly tedious work, as every test will likely take five minutes...).



--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

nophead
Most of my tests seem a bit faster with the RC. Text renders slightly differently.

dip_diff.png

On Wed, 9 Dec 2020 at 20:17, nop head <[hidden email]> wrote:
Here is an example where the RC is 3 times slower than the release. https://github.com/nophead/NopSCADlib/blob/master/tests/pulleys.scad

OpenSCAD 2019.05
https://www.openscad.org/

Copyright (C) 2009-2019 The OpenSCAD Developers

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

Loaded design 'W:/CHRIS/Projects/RepRaptor/OpenSCADLibs/NopSCADlib/tests/pulleys.scad'.

Used file cache size: 18 files

Compiling design (CSG Tree generation)...

Compiling design (CSG Products generation)...

Geometries in cache: 259

Geometry cache size in bytes: 7707888

CGAL Polyhedrons in cache: 9

CGAL cache size in bytes: 5571112

Compiling design (CSG Products normalization)...

Normalized CSG tree has 36 elements

Compile and preview finished.

Total rendering time: 0 hours, 0 minutes, 4 seconds


Loaded design 'W:/CHRIS/Projects/RepRaptor/OpenSCADLibs/NopSCADlib/tests/pulleys.scad'.
OpenSCAD 2020.12-RC2
https://www.openscad.org/

Copyright (C) 2009-2020 The OpenSCAD Developers

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

Used file cache size: 18 files
Compiling design (CSG Tree generation)...
Compiling design (CSG Products generation)...
Geometries in cache: 245
Geometry cache size in bytes: 5961344
CGAL Polyhedrons in cache: 23
CGAL cache size in bytes: 21865784
Compiling design (CSG Products normalization)...
Normalized tree has 36 elements!
Compile and preview finished.
Total rendering time: 0:00:12.699


On Wed, 9 Dec 2020 at 16:48, Troberg <[hidden email]> wrote:
thehans wrote
> On Wed, Dec 9, 2020 at 5:59 AM Troberg &lt;

> troberg.anders@

> &gt; wrote:
>
>> No big loops, can't really simplify using $fn and the others, as I need
>> the
>> precision I have. Either way, it shouldn't behave so differently.
>>
>
>  I'm talking about doing things to help troubleshoot the issue, this would
> be temporary, not permanent changes to your project.
> ie: make a complete copy of your project if you need to, then try making
> big drastic changes to narrow down where the slowdown comes from.
> For this purpose it doesn't matter if the rendered result looks absolutely
> nothing like your original design, as long as it helps towards a
> reproducible example of where RC is much slower than previous release.
>
> If you don't want to bother troubleshooting anything then whatever, just
> don't expect anyone to bother fixing it.

No need to be snarky about it, I've been working with software development
for 30 years. I know that my description isn't much help in finding the
problem.

My thought was that if I mention the problem, other people will watch out
for the problem, and hopefully, they'll run into it in a cleaner example
which means that I don't have to gut a 10 000 lines project (which will be
exceedingly tedious work, as every test will likely take five minutes...).



--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

MichaelAtOz
Administrator
nophead wrote
> Most of my tests seem a bit faster with the RC. Text renders slightly
> differently.

Differently re performance, or graphically?



-----
OpenSCAD Admin - email* me if you need anything,  or if I've done something stupid...

* on the Forum, click on my MichaelAtOz label, there is a link to email me.

Unless specifically shown otherwise above, my contribution is in the Public Domain; to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. Obviously inclusion of works of previous authors is not included in the above.

--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
OpenSCAD Admin - email* me if you need anything, or if I've done something stupid...
* on the Forum, click on my MichaelAtOz label, there is a link to email me.

Unless specifically shown otherwise above, my contribution is in the Public Domain;
to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work.
Obviously inclusion of works of previous authors is not included in the above.
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

nophead
Graphically. A few pixels around the edges differ as shown in the picture I posted. Not really a problem but I am curious why it is the case. I get similar differences in pixels around the edges of everything when I move to a different PC but with the same PC images are generally identical.

On Thu, 10 Dec 2020 at 11:46, MichaelAtOz <[hidden email]> wrote:
nophead wrote
> Most of my tests seem a bit faster with the RC. Text renders slightly
> differently.

Differently re performance, or graphically?



-----
OpenSCAD Admin - email* me if you need anything,  or if I've done something stupid...

* on the Forum, click on my MichaelAtOz label, there is a link to email me.

Unless specifically shown otherwise above, my contribution is in the Public Domain; to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. Obviously inclusion of works of previous authors is not included in the above.

--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

thehans
Probably due to a slight difference in the way sizing is applied to text now, from here:

Previously it relied on Freetype's funky (and very limited) fixed-point number format, now it scales the vertex positions with double precision floating point multiplication, after getting the glyph paths from Freetype.

On Thu, Dec 10, 2020 at 6:07 AM nop head <[hidden email]> wrote:
Graphically. A few pixels around the edges differ as shown in the picture I posted. Not really a problem but I am curious why it is the case. I get similar differences in pixels around the edges of everything when I move to a different PC but with the same PC images are generally identical.

On Thu, 10 Dec 2020 at 11:46, MichaelAtOz <[hidden email]> wrote:
nophead wrote
> Most of my tests seem a bit faster with the RC. Text renders slightly
> differently.

Differently re performance, or graphically?



-----
OpenSCAD Admin - email* me if you need anything,  or if I've done something stupid...

* on the Forum, click on my MichaelAtOz label, there is a link to email me.

Unless specifically shown otherwise above, my contribution is in the Public Domain; to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. Obviously inclusion of works of previous authors is not included in the above.

--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

thehans
In reply to this post by nophead

On Wed, Dec 9, 2020 at 2:18 PM nop head <[hidden email]> wrote:
Here is an example where the RC is 3 times slower than the release. https://github.com/nophead/NopSCADlib/blob/master/tests/pulleys.scad
...

Thank you nophead for the reproducible example.  I've opened an issue on github for further tracking. https://github.com/openscad/openscad/issues/3529

On Thu, Dec 10, 2020 at 5:04 AM nop head <[hidden email]> wrote:
Most of my tests seem a bit faster with the RC.
 
And thanks for acknowledging that it's not all doom and gloom either :)

Hans

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

nophead
No problem. OpenScad is the only program I can think of that mostly evolves to being faster.

On Fri, 11 Dec 2020 at 13:28, Hans L <[hidden email]> wrote:

On Wed, Dec 9, 2020 at 2:18 PM nop head <[hidden email]> wrote:
Here is an example where the RC is 3 times slower than the release. https://github.com/nophead/NopSCADlib/blob/master/tests/pulleys.scad
...

Thank you nophead for the reproducible example.  I've opened an issue on github for further tracking. https://github.com/openscad/openscad/issues/3529

On Thu, Dec 10, 2020 at 5:04 AM nop head <[hidden email]> wrote:
Most of my tests seem a bit faster with the RC.
 
And thanks for acknowledging that it's not all doom and gloom either :)

Hans
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

lar3ry
In reply to this post by nophead
nophead wrote

> Here is an example where the RC is 3 times slower than the release.
> https://github.com/nophead/NopSCADlib/blob/master/tests/pulleys.scad
>
> OpenSCAD 2019.05
> Geometries in cache: 259
> Geometry cache size in bytes: 7707888
> CGAL Polyhedrons in cache: 9
> CGAL cache size in bytes: 5571112
> Total rendering time: 0 hours, 0 minutes, 4 seconds
>
> OpenSCAD 2020.12-RC2
> Geometries in cache: 245
> Geometry cache size in bytes: 5961344
> CGAL Polyhedrons in cache: 23
> CGAL cache size in bytes: 21865784
> Total rendering time: 0:00:12.699

Yes, the rendering is slower, but I was more startled by the differences in
the lines quoted above.
I think I understand the differences in cache sizes, but I don't understand
the differences in geometries and CGAL Polyhedrons




--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

nophead
Yes I think that is a clue to the difference in speed but the end result looks the same when shown as a PNG.

On Fri, 11 Dec 2020 at 17:05, lar3ry <[hidden email]> wrote:
nophead wrote
> Here is an example where the RC is 3 times slower than the release.
> https://github.com/nophead/NopSCADlib/blob/master/tests/pulleys.scad
>
> OpenSCAD 2019.05
> Geometries in cache: 259
> Geometry cache size in bytes: 7707888
> CGAL Polyhedrons in cache: 9
> CGAL cache size in bytes: 5571112
> Total rendering time: 0 hours, 0 minutes, 4 seconds
>
> OpenSCAD 2020.12-RC2
> Geometries in cache: 245
> Geometry cache size in bytes: 5961344
> CGAL Polyhedrons in cache: 23
> CGAL cache size in bytes: 21865784
> Total rendering time: 0:00:12.699

Yes, the rendering is slower, but I was more startled by the differences in
the lines quoted above.
I think I understand the differences in cache sizes, but I don't understand
the differences in geometries and CGAL Polyhedrons




--
Sent from: http://forum.openscad.org/

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Reply | Threaded
Open this post in threaded view
|

Re: OpenSCAD release candidate: 2020.12-RC2

RevarBat
In reply to this post by softfoot
I’m on Mac OS Catalina 10.15.7, trying out OpenSCAD snapshot 2020.12.11.
I cannot seem to open files into a new tab.  My current existing tab always gets replaced by the new file tab.  This isn’t making a new window or anything like that.  When I do try to open into a new window that works fine.  It’s just opening a new file into a new tab that loses the old tab.
I see the same tab-replacement behavior when opening scad files from the command-line with the ‘open’ command.

- Revar



_______________________________________________
OpenSCAD mailing list
[hidden email]
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
12