I think we discussed this many times and it all came down to "someone
needs to implement it".
If I'm not imagining things the consensus back then was that the units
should be the user's
design units, so if the user had Imperial measures it would remain
0,1-inch but if the user had
Metric measures
Hi folks,
Attached is a patch that adds a patch that helps wxWidgets build on
the latest macOS 10.13 High Sierra.
Additionally, when folks build wxWidgets, they'll need to set the
CPPFLAG of D__ASSERT_MACROS_DEFINE_VERSIONS_WITHOUT_UNDERSCORES=1.
I'm working on updating the packaging scripts so
Nick,
Good point. I'll raise a small patch later on that looks for IGES if no
STEP is found
On Tue, Nov 7, 2017 at 10:41 AM, Nick Østergaard wrote:
> What about iges files? Should't it try to export iges files if no step is
> found?
>
> 2017-11-07 0:32 GMT+01:00 Wayne
On 11/3/2017 9:08 AM, Maciej Sumiński wrote:
> On 11/03/2017 01:38 PM, Carsten Schoenert wrote:
>> Hello Marvin,
>>
>> Am 02.11.2017 um 21:58 schrieb Marvin Schmidt:
>>> Just a set of small code cleanups to fix a bunch of warnings
>>> and remove dead code.
>>> Other than that the 'build:' commits
STEP simply has no provision for entities required for photo-realistic
rendering.
Even VRML is somewhat limited in that respect but it's leagues ahead of STEP.
- Cirilo
On Mon, Nov 6, 2017 at 9:27 PM, David Novak wrote:
> Please enlighten me. I've always thought of STEP
Oliver,
I merged your patch with Orson's suggestion along with fixing the commit
message to include the launchpad bug number. Thank you for your
contribution to KiCad.
Cheers,
Wayne
On 11/4/2017 8:47 AM, Oliver Walters wrote:
> The attached patch implements the feature suggested here
> :
Ohh, I have never seen that. I guess this is because I always use the
default colors for wires which is the same green as the junction dot. Thank
you for the clarificaiton :)
2017-11-06 19:01 GMT+01:00 Seth Hillbrand :
> Hi Nick-
>
> Attached are two images, showing the
What about iges files? Should't it try to export iges files if no step is
found?
2017-11-07 0:32 GMT+01:00 Wayne Stambaugh :
> Hey Orson,
>
> I've got it now. Go ahead and commit it if you have it ready to go. If
> not, I will do it in the next day or so.
>
> Cheers,
>
>
Hey Orson,
I've got it now. Go ahead and commit it if you have it ready to go. If
not, I will do it in the next day or so.
Cheers,
Wayne
On 11/6/2017 6:02 PM, Maciej Suminski wrote:
> Hi Wayne,
>
> I think it only affects the STEP exporter (kicad2step executable), which
> cannot handle WRL
Apparently I did not read the patch correctly.
On 11/6/2017 3:30 PM, Oliver Walters wrote:
> Wayne,
>
>
> /It would make more//
> sense to me that kicad2step would attempt to substitute the wrl file
> links withe step files when converting wrl to step./
>
> This is exactly what it does. I'm
Oliver,
if you create by script or code or manually an array of single STEP
models, the resulting array could be not fine in geometric terms also if
the single model is good...
Please have some deep insight to FreeCAD forum or even OCC forum to see
how mechanical boolean operation can be
Hi Wayne,
I think it only affects the STEP exporter (kicad2step executable), which
cannot handle WRL models at all. With the patch applied, the exporter
looks for a model with the same file name, but .step extension. In case
there one is found - it is during the STEP export. I believe there is
Orson,
Yes a decompression plugin would be a good addition to the
"download-as-needed" that Maurice mentions above.
On Tue, Nov 7, 2017 at 9:54 AM, Maciej Suminski
wrote:
> On 11/06/2017 11:43 PM, Oliver Walters wrote:
> > Orson,
> >
> > That's a good idea, and I am
Maurice,
This patch does *not* generate any models on the fly. It provides the
ability for users to re-use existing model data (which they are NOT forced
to use by any means). In the example of pin headers or terminal blocks,
where each unit is identical, there is no need to re-validate each
On 11/06/2017 11:43 PM, Oliver Walters wrote:
> Orson,
>
> That's a good idea, and I am actively investigating that. Check out
> https://kicad.github.io - currently only symbols and footprints are there
> but models are on their way. The scripts that generate that data use a very
> heavy 7z
Oliver,
just an other point against it IMO:
If you develop a build on the fly 3D library, in case of update on OCE
release, you would need to check and probably review all the models
generated by your scripts...
As I already pointed out on GitHub discussion, STEP are not just
rendering
Maurice,
I really like the idea of downloading models on an as-needed basis. This
will be a lot of work to implement (but I think it could be done).
This patch is not intended to be the final solution for the libraries
issue. I would still support a move to download models as needed
On Tue, Nov
Orson,
That's a good idea, and I am actively investigating that. Check out
https://kicad.github.io - currently only symbols and footprints are there
but models are on their way. The scripts that generate that data use a very
heavy 7z compression.
However, is this a reason to *not* add the array
>
> I do like to use VRML models in the 3D viewer of pcbnew as I can use
> textures and materials with them so they look good when I take a screen
> shot for use on a web page. The CAD software I use can import many model
> file formats including VRML and STEP. It doesn't matter to me which type
Hi Oliver,
What do you think about providing the models as compressed files? As any
text format, one can expect quite high compression ratio. Perhaps that
would be sufficient to solve the huge libraries problem.
Cheers,
Orson
On 11/06/2017 03:01 PM, Oliver Walters wrote:
> To provide an option
Seth, for the STEP export that would be a better idea, I can look into how
to implement that in the kicad2step tool.
Wayne, if you are completely against this as a feature, then I shan't spend
any more time on it, but:
1. I think it is a useful feature to have even if it is not the "right" way
I think it may be preferable to utilize the STEP file's inherent
array_placement_group function to generate an array of the underlying
structure.
-S
On Mon, Nov 6, 2017 at 1:01 PM, Kevin Cozens wrote:
> On 2017-11-06 09:01 AM, Oliver Walters wrote:
>
>> To provide an option to
Please enlighten me. I've always thought of STEP as the best end game,
but haven't had much time to investigate advantages and disadvantages.
What prevents STEP models from looking just as good as VRML models?
Best Regards,
David
On 11/6/2017 4:17 PM, Cirilo Bernardo wrote:
I wouldn't worry
STEPs don't need to look ugly, a "simple" extension of the step format
could be a mtl file that indexes vrml materials to STEP colors, it would be
ignored by MCAD, but kicad could pick that up to significantly prettify the
step files in the 3d viewer without having to generate a redundant WRL file
On 2017-11-06 01:03 PM, Wayne Stambaugh wrote:
The second issue is that I imagine most users who export to step files
don't want to deal with a large array of single pin 3D models when they
are using the board 3D model for mechanical layout inside a 3D modeling
application such as FreeCAD.
On 2017-11-06 03:30 PM, Oliver Walters wrote:
Wayne,
 /It would make more//
sense to me that kicad2step would attempt to substitute the wrl file
links withe step files when converting wrl to step./
This is exactly what it does. I'm not sure what you mean with embedded step
file references.Â
I wouldn't worry about the STEP file having numerous small parts;
this happens all the time anyway unless the model can be fused.
I don't see us abandoning VRML. I wrote the new parser but never
use VRML except when testing the VRML parser. However, many
people out there like to have VRML to make
What is happening is that kicad2step will now take any file name with
*.wrl and attempt to open a similarly named stp/step file instead.
kicad2step will never support importing any actual wrl data since it
is not valid solid model data.
- Cirilo
On Mon, Nov 6, 2017 at 4:07 PM, Wayne Stambaugh
On 2017-11-06 09:01 AM, Oliver Walters wrote:
To provide an option to reduce the size of the 3D model library, I have
implemented an "array" feature for 3D models. A module (footprint) can
reference a single model multiple times, with a dimensional offset between
each copy.
Sounds
On 2017-11-06 09:01 AM, Oliver Walters wrote:
To provide an option to reduce the size of the 3D model library, I have
implemented an "array" feature for 3D models. A module (footprint) can
reference a single model multiple times, with a dimensional offset between
each copy.
Sounds
Wayne,
*It would make more*
*sense to me that kicad2step would attempt to substitute the wrl filelinks
withe step files when converting wrl to step.*
This is exactly what it does. I'm not sure what you mean with embedded step
file references.
On 7 Nov 2017 03:07, "Wayne Stambaugh"
Hey Oliver,
I don't think this is the way to go. I see two primary issues. The
first is the impact on the code clarity. If this only stopped at your
changes, it might be fine but how many other array or grouping
permutations will happen over time making this code a huge mess? I just
see this
Hi Nick-
Attached are two images, showing the issue and the resolution. The
attached schematic should also show it on your machine. Note that this is
captured in Eldar's bug report as well.
-Seth
On Mon, Nov 6, 2017 at 4:37 AM, Nick Østergaard wrote:
> What does the issue
Hi,
referring to:
"even if it is "only" for the pin headers, this is close to half of the
3d library size as of now"
you probably missed the 3d library status...please have a look at recent
PRs of the 3D library
There are contributors of new models that are afraid to contribute
because of
The only thing i really have against this is that it would make a pretty
messy step file because the component would be broken up in tiny parts.
Wouldn't there be a space saving with just ditching wrl altogether? the
next stable won't need them
On Mon, Nov 6, 2017 at 9:53 AM, Kristoffer Ödmark <
If I'm reading this patch correctly, exported wrl files will always have
step file references in them even when they are found. If that is the
case, then I am opposed to this change. AFAIK, opening the wrl board
file will fail with the embedded step file links. It would make more
sense to me
To be fair, as it stands now, even if it is "only" for the pin headers,
this is close to half of the 3d library size as of now.
The pin headers can now be of an arbitrary size as well. Someone can now
create a 4x6 pin header for example, and the current footprints can be
converted to this
Hi Oliver,
I'm sorry not to be on your side for this option...
1) for which kind of modules this array is applicable?
I see only pin-headers straight and angled...
for example box headers are not easily done unless you consider to
manage by the code the box for each model
2) the problem
I think that the bug is only present in the way that the offset is
displayed in the footprint preview window. I will not break compatibility :)
On Mon, Nov 6, 2017 at 11:59 PM, Wayne Stambaugh
wrote:
> Be careful here. If the 3d model offset is currently in some units
>
The reason that it is breaking is that I am currently generating a unique
download link for each individual 3D model group. A "group" contains .wrl
and .step pairs.
If I do not go this far, and provide a single archive for each "folder" of
files then the generator still works fine..
For now at
at some point changing the models to be mm rather than 0.1" or whatever it is
now needs to be seriously looked at
> On 7/11/2017, at 01:59, Wayne Stambaugh wrote:
>
> Be careful here. If the 3d model offset is currently in some units
> other than mm and you change this,
Be careful here. If the 3d model offset is currently in some units
other than mm and you change this, all existing 3d model offsets will be
broken.
On 11/6/2017 7:56 AM, Oliver Walters wrote:
> That's what I expected. It then appears that there is a bug in the 3D
> code with regard to model
That's what I expected. It then appears that there is a bug in the 3D code
with regard to model offset. I'm working on a related patch set that will
fix this, I just wanted to be sure. Thanks Wayne
On Mon, Nov 6, 2017 at 11:53 PM, Wayne Stambaugh
wrote:
> It should be mm
It should be mm like all of the other coordinates.
On 11/6/2017 7:32 AM, Oliver Walters wrote:
> With regard to the 3D model offset paramater
>
> e.g.
>
> (at (xyz a b c))
>
> Are the units stored in the file inches or mm?
>
> Cheers,
> Oliver
>
>
>
I'm not sure what to do about this. I'm not opposed adding this to the
kicad website. Do we even have enough server space on the kicad website
to handle this? Anyone else have any ideas?
Who knew the kicad project would reach these kinds of limits so quickly.
Kudos to the library devs for all
What does the issue look like visually? Do you have a demo file to verify
this patch? This is nice to know if someone reports this issue in stable.
2017-11-03 21:08 GMT+01:00 Kevin Cozens :
> On 2017-11-03 02:00 PM, Seth Hillbrand wrote:
>
>> ​Kevin made a ​request to have
With regard to the 3D model offset paramater
e.g.
(at (xyz a b c))
Are the units stored in the file inches or mm?
Cheers,
Oliver
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe
Orson,
Literally 10 seconds after submitting the patch I realised that .Lower()
was the superior solution :)
Thanks,
Oliver
On Mon, Nov 6, 2017 at 11:17 PM, Maciej Sumiński
wrote:
> Hi Oliver,
>
> If I may suggest one thing, then I would turn:
>
> +if( ext ==
Hi Oliver,
If I may suggest one thing, then I would turn:
+if( ext == "wrl" || ext == "WRL" || ext == "Wrl" )
+return FMT_WRL;
into:
+if( ext.Lower() == "wrl" )
+return FMT_WRL;
Apart from that, the patch seems fine just by reading the contents. No
need to send another
49 matches
Mail list logo