bug report

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

bug report

eagleapex
I'm using release 2011.06 on my MacBook and I'm getting some odd display problems in the viewport when compiling with OpenCSG
Check out this screenshot http://dl.dropbox.com/u/1201442/Screen%20shot%202011-06-30%20at%2011.11.47%20PM.jpg
There is a horizontal line through the exact middle of the viewport. Above this boundary the rendering is wrong does not show the proper forms. If I rotate the model around, the boundary stays, but the proper features on the model will be visible below the boundary.

Rendering with CGAL produces an accurate model.

Do you have a fix?

Christopher Thompson
~~~~
Visit me at http://eagleapex.com
Reply | Threaded
Open this post in threaded view
|

Re: bug report

kintel
Administrator
Hi,

I haven't seen this before. Could you provide the scad file in question?
Also, which MacBook and which version of Mac OS X are you using?
If you export the info from the OpenGL Driver Monitor (Monitors -> Renderer info), it might also be helpful.

 -Marius

On Jul 1, 2011, at 06:55 AM, Christopher Thompson wrote:

> I'm using release 2011.06 on my MacBook and I'm getting some odd display problems in the viewport when compiling with OpenCSG
> Check out this screenshot http://dl.dropbox.com/u/1201442/Screen%20shot%202011-06-30%20at%2011.11.47%20PM.jpg
> There is a horizontal line through the exact middle of the viewport. Above this boundary the rendering is wrong does not show the proper forms. If I rotate the model around, the boundary stays, but the proper features on the model will be visible below the boundary.
>
> Rendering with CGAL produces an accurate model.
>
> Do you have a fix?
>
> Christopher Thompson
> ~~~~
> Visit me at http://eagleapex.com
> _______________________________________________
> OpenSCAD mailing list
> [hidden email]
> http://rocklinux.net/mailman/listinfo/openscad


Reply | Threaded
Open this post in threaded view
|

Re: bug report

eagleapex
MacBook Intel Core 2 Duo 2 GHz
NVIDIA GeForce 9400M
Mac OS X 10.6.8 (10K540)

Of course now that I try to do it, the problem is gone. 

I also can't find the monitors>render info you ask for.


Christopher Thompson
~~~~

On Jul 5, 2011, at 9:52 PM, Marius Kintel wrote:

Hi,

I haven't seen this before. Could you provide the scad file in question?
Also, which MacBook and which version of Mac OS X are you using?
If you export the info from the OpenGL Driver Monitor (Monitors -> Renderer info), it might also be helpful.

-Marius

On Jul 1, 2011, at 06:55 AM, Christopher Thompson wrote:

I'm using release 2011.06 on my MacBook and I'm getting some odd display problems in the viewport when compiling with OpenCSG
Check out this screenshot http://dl.dropbox.com/u/1201442/Screen%20shot%202011-06-30%20at%2011.11.47%20PM.jpg
There is a horizontal line through the exact middle of the viewport. Above this boundary the rendering is wrong does not show the proper forms. If I rotate the model around, the boundary stays, but the proper features on the model will be visible below the boundary.

Rendering with CGAL produces an accurate model.

Do you have a fix?

Christopher Thompson
~~~~
Visit me at http://eagleapex.com
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

Reply | Threaded
Open this post in threaded view
|

Re: bug report

kintel
Administrator
On Jul 6, 2011, at 05:24 AM, Christopher Thompson wrote:
>
> Of course now that I try to do it, the problem is gone.
>
It could be a freak bug. If it happens again, please get in touch - perhaps try to figure out in what circumstances it happens.

> I also can't find the monitors>render info you ask for.
>
Start "OpenGL  Driver Monitor.app". In the "Monitors" menu, there should be a "Renderer Info". This will open a new Window. Select NVIDIA 9400M and click "Save As Text".
..but don't worry - now that I know your graphics card I can find this info myself.

 -Marius


Reply | Threaded
Open this post in threaded view
|

Re: bug report

Laszlo KREKACS
Hey,

It does appear for me too (Intel 4500MHD), it happens when the two
base model has a touching face.
So I make always extra sure when doing difference one will be a bit
taller then the other, because this rendering error will appear.
(putting a box inside an another box at same Z (so the bottom face is
the exact same) this rendering error does show)

I always thought it is normal.


Best,
 Laszlo


On Wed, Jul 6, 2011 at 2:55 PM, Marius Kintel <[hidden email]> wrote:

> On Jul 6, 2011, at 05:24 AM, Christopher Thompson wrote:
>>
>> Of course now that I try to do it, the problem is gone.
>>
> It could be a freak bug. If it happens again, please get in touch - perhaps try to figure out in what circumstances it happens.
>
>> I also can't find the monitors>render info you ask for.
>>
> Start "OpenGL  Driver Monitor.app". In the "Monitors" menu, there should be a "Renderer Info". This will open a new Window. Select NVIDIA 9400M and click "Save As Text".
> ..but don't worry - now that I know your graphics card I can find this info myself.
>
>  -Marius
>
> _______________________________________________
> OpenSCAD mailing list
> [hidden email]
> http://rocklinux.net/mailman/listinfo/openscad
>

Reply | Threaded
Open this post in threaded view
|

Re: bug report

kintel
Administrator
On Jul 17, 2011, at 22:16 PM, Laszlo KREKACS wrote:

> So I make always extra sure when doing difference one will be a bit
> taller then the other, because this rendering error will appear.
> (putting a box inside an another box at same Z (so the bottom face is
> the exact same) this rendering error does show)
>
> I always thought it is normal.
>
Ah, ok, if it's just z buffer tearing of coincident positive and negative surfaces, that's a known visual artifact.
I didn't quite catch that from the original report.

 -Marius


Reply | Threaded
Open this post in threaded view
|

Re: bug report

Laszlo KREKACS
On Mon, Jul 18, 2011 at 12:18 AM, Marius Kintel <[hidden email]> wrote:
> Ah, ok, if it's just z buffer tearing of coincident positive and negative surfaces, that's a known visual artifact.
> I didn't quite catch that from the original report.

Nope, you are right. The original report is not about the z buffer
tearing. I somehow wrongly observed the screenshot and slipped over
the actual bug!

Laszlo