[vtk-developers] TestSplitViewportStereoHorizontal

Fri Feb 22 16:46:36 EST 2013

On Fri, Feb 22, 2013 at 4:28 PM, Bill Lorensen <bill.lorensen at gmail.com> wrote:
> Aashish,
>
> I merged your topic and checked in a baseline.

You are quick.!

>
> Thanks for fixing this long standing issue.

Thanks to you for bringing this up. I will keep an eye on the
dashboards for any other possible issues.

>
> Bill
>
>
> On Fri, Feb 22, 2013 at 3:27 PM, Bill Lorensen <bill.lorensen at gmail.com>
> wrote:
>>
>> Aashish,
>>
>> When I reenable the Mac test on my Mac, the image looks the same as the
>> image in your gerrit build.
>>
>> Very encouraging. If the Ubuntu looks good, I suggest you check in a new
>> baseline and merge. Then I'll submit a patch to reenable the Mac test.
>>
>> Thanks
>>
>> Bill
>>
>>
>> On Fri, Feb 22, 2013 at 3:01 PM, Aashish Chaudhary
>> <aashish.chaudhary at kitware.com> wrote:
>>>
>>> > I was thinking of moving the object in front (and that should make eye
>>> > separation larger anyways). I will push something soon.
>>>
>>> Ok, I pushed the change. OffAxis will fail but the image should be
>>> consistent across. If someone has red-blue glasses (I lost mine), then
>>> they should be able
>>> to confirm if the cone appears in front of the sphere.
>>>
>>>
>>> >
>>> >
>>> >>
>>> >>
>>> >>
>>> >> On Fri, Feb 22, 2013 at 1:15 PM, Bill Lorensen
>>> >> <bill.lorensen at gmail.com>
>>> >> wrote:
>>> >>>
>>> >>> Aashish,
>>> >>>
>>> >>> I think you are right. The eye separation should be larger. If you
>>> >>> follow
>>> >>> the 1/30th rule: "the interaxial separation should only be 1/30th of
>>> >>> the
>>> >>> distance from your camera to the closest subject"
>>> >>> then a separation of .4 to .5 should give a reasonable image. And it
>>> >>> does.
>>> >>>
>>> >>> Bill
>>> >>>
>>> >>>
>>> >>> On Fri, Feb 22, 2013 at 12:27 PM, Aashish Chaudhary
>>> >>> <aashish.chaudhary at kitware.com> wrote:
>>> >>>>
>>> >>>> On Fri, Feb 22, 2013 at 12:08 PM, Bill Lorensen
>>> >>>> <bill.lorensen at gmail.com>
>>> >>>> wrote:
>>> >>>> > Your recent patch fixes the SplitView test, but breaks the OffAxis
>>> >>>> > test. Any
>>> >>>> > ideas?
>>> >>>>
>>> >>>> I looked into the parameters, I think the baseline image is
>>> >>>> incorrect.
>>> >>>> I have to verify this but I am 80% sure. The spheres in the test are
>>> >>>> behind the projected wall and should not have this much eye
>>> >>>> separation.
>>> >>>>
>>> >>>> >
>>> >>>> >
>>> >>>> > On Thu, Feb 21, 2013 at 10:46 AM, Aashish Chaudhary
>>> >>>> > <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >>
>>> >>>> >> On Thu, Feb 21, 2013 at 10:41 AM, Bill Lorensen
>>> >>>> >> <bill.lorensen at gmail.com>
>>> >>>> >> wrote:
>>> >>>> >> > I did some experiments last night and I don't think these
>>> >>>> >> > changes
>>> >>>> >> > will
>>> >>>> >> > help.
>>> >>>> >> > Aashish any luck? I noticed you removed the old baselines. I
>>> >>>> >> > think
>>> >>>> >> > you
>>> >>>> >> > should have, just temporarily, added a new baseline for your
>>> >>>> >> > changes
>>> >>>> >> > that
>>> >>>> >> > are not yet merged. Now we have even more failing tests.
>>> >>>> >>
>>> >>>> >> Yes, after I fixed window to image filter, at least on my two
>>> >>>> >> systems
>>> >>>> >> the results were consistent. I will push a branch later in the
>>> >>>> >> day.
>>> >>>> >>
>>> >>>> >>
>>> >>>> >> >
>>> >>>> >> >
>>> >>>> >> > On Wed, Feb 20, 2013 at 5:34 PM, David Gobbi
>>> >>>> >> > <david.gobbi at gmail.com>
>>> >>>> >> > wrote:
>>> >>>> >> >>
>>> >>>> >> >> I will be overjoyed if this also fixes volTM2DRotateClip as a
>>> >>>> >> >> side-effect.
>>> >>>> >> >>
>>> >>>> >> >>  - David
>>> >>>> >> >>
>>> >>>> >> >> On Wed, Feb 20, 2013 at 3:26 PM, Bill Lorensen
>>> >>>> >> >> <bill.lorensen at gmail.com>
>>> >>>> >> >> wrote:
>>> >>>> >> >> > Nice catch!
>>> >>>> >> >> >
>>> >>>> >> >> > On Wed, Feb 20, 2013 at 5:25 PM, Aashish Chaudhary
>>> >>>> >> >> > <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >> To be specific the code is here:
>>> >>>> >> >> >>
>>> >>>> >> >> >> void vtkWindowToImageFilter::RequestData
>>> >>>> >> >> >>
>>> >>>> >> >> >> Which gets executed by vtkTesting
>>> >>>> >> >> >>
>>> >>>> >> >> >> Thanks,
>>> >>>> >> >> >>
>>> >>>> >> >> >>
>>> >>>> >> >> >> On Wed, Feb 20, 2013 at 5:23 PM, Aashish Chaudhary
>>> >>>> >> >> >> <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >>> Ok, I found the issue..it was the vtkTesting, Basically
>>> >>>> >> >> >>> there
>>> >>>> >> >> >>> it
>>> >>>> >> >> >>> creates a new camera and line by line copies the member
>>> >>>> >> >> >>> variables
>>> >>>> >> >> >>> (I
>>> >>>> >> >> >>> don't know why we just deep copy the cameras).
>>> >>>> >> >> >>>
>>> >>>> >> >> >>> There it was missing copying off screen parameters. I will
>>> >>>> >> >> >>> push
>>> >>>> >> >> >>> a
>>> >>>> >> >> >>> fix
>>> >>>> >> >> >>> tomorrow (need to leave very soon). Along with that I will
>>> >>>> >> >> >>> push
>>> >>>> >> >> >>> fixes
>>> >>>> >> >> >>> to shallow copy as well.
>>> >>>> >> >> >>>
>>> >>>> >> >> >>> Thanks
>>> >>>> >> >> >>>
>>> >>>> >> >> >>>
>>> >>>> >> >> >>> On Wed, Feb 20, 2013 at 4:45 PM, Aashish Chaudhary
>>> >>>> >> >> >>> <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >>>> Thanks David. PartialCopy has issues but it seems like
>>> >>>> >> >> >>>> that's
>>> >>>> >> >> >>>> not
>>> >>>> >> >> >>>> the
>>> >>>> >> >> >>>> problem. Basically all of the camera parameters are same
>>> >>>> >> >> >>>> before
>>> >>>> >> >> >>>> and
>>> >>>> >> >> >>>> after.
>>> >>>> >> >> >>>>
>>> >>>> >> >> >>>> Thanks,
>>> >>>> >> >> >>>>
>>> >>>> >> >> >>>>
>>> >>>> >> >> >>>> On Wed, Feb 20, 2013 at 4:12 PM, David Gobbi
>>> >>>> >> >> >>>> <david.gobbi at gmail.com>
>>> >>>> >> >> >>>> wrote:
>>> >>>> >> >> >>>>> Look at PartialCopy(), make sure it isn't missing
>>> >>>> >> >> >>>>> anything.
>>> >>>> >> >> >>>>>
>>> >>>> >> >> >>>>>  - David
>>> >>>> >> >> >>>>>
>>> >>>> >> >> >>>>>
>>> >>>> >> >> >>>>> On Wed, Feb 20, 2013 at 2:02 PM, Aashish Chaudhary
>>> >>>> >> >> >>>>> <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >>>>>> Ok, I found something interesting, these settings are
>>> >>>> >> >> >>>>>> gone
>>> >>>> >> >> >>>>>> inside
>>> >>>> >> >> >>>>>> the
>>> >>>> >> >> >>>>>> call to vtkRegressionTestImageThreshold
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> camera->SetScreenBottomLeft(bottomLeft);
>>> >>>> >> >> >>>>>> camera->SetScreenBottomRight(bottomRight);
>>> >>>> >> >> >>>>>> camera->SetScreenTopRight(topRight);
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> And that's causing it create different images. I don't
>>> >>>> >> >> >>>>>> know
>>> >>>> >> >> >>>>>> the
>>> >>>> >> >> >>>>>> exact
>>> >>>> >> >> >>>>>> reason yet but this is a progress.
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> Thanks,
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> On Wed, Feb 20, 2013 at 3:49 PM, Aashish Chaudhary
>>> >>>> >> >> >>>>>> <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >>>>>>> On Wed, Feb 20, 2013 at 3:36 PM, Bill Lorensen
>>> >>>> >> >> >>>>>>> <bill.lorensen at gmail.com> wrote:
>>> >>>> >> >> >>>>>>>> Thanks for looking at this, but unfortunately, the
>>> >>>> >> >> >>>>>>>> mac and
>>> >>>> >> >> >>>>>>>> pc
>>> >>>> >> >> >>>>>>>> images
>>> >>>> >> >> >>>>>>>> do not look the same.
>>> >>>> >> >> >>>>>>>
>>> >>>> >> >> >>>>>>> Yeah.. I realized the same. It seems like windows is
>>> >>>> >> >> >>>>>>> capturing
>>> >>>> >> >> >>>>>>> the
>>> >>>> >> >> >>>>>>> scene after first render and mac / linux captured that
>>> >>>> >> >> >>>>>>> it
>>> >>>> >> >> >>>>>>> should
>>> >>>> >> >> >>>>>>> be
>>> >>>> >> >> >>>>>>> after second render. I guess
>>> >>>> >> >> >>>>>>> I can try to setup camera parameters before hand and
>>> >>>> >> >> >>>>>>> see if
>>> >>>> >> >> >>>>>>> that
>>> >>>> >> >> >>>>>>> prevents this behavior.
>>> >>>> >> >> >>>>>>>
>>> >>>> >> >> >>>>>>> Thanks
>>> >>>> >> >> >>>>>>>
>>> >>>> >> >> >>>>>>>>
>>> >>>> >> >> >>>>>>>> On Wed, Feb 20, 2013 at 2:27 PM, Aashish Chaudhary
>>> >>>> >> >> >>>>>>>> <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >>>>>>>>> Bill,
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>> I pushed a branch and hoping that it will fix the
>>> >>>> >> >> >>>>>>>>> issue
>>> >>>> >> >> >>>>>>>>> atleast
>>> >>>> >> >> >>>>>>>>> on mac
>>> >>>> >> >> >>>>>>>>> and linux. I may have to check for stereo support
>>> >>>> >> >> >>>>>>>>> but I
>>> >>>> >> >> >>>>>>>>> will
>>> >>>> >> >> >>>>>>>>> for
>>> >>>> >> >> >>>>>>>>> that.
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>> I will keep an eye on the dashboards, in the mean
>>> >>>> >> >> >>>>>>>>> time
>>> >>>> >> >> >>>>>>>>> please
>>> >>>> >> >> >>>>>>>>> review my branch.
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>> Thanks
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>> On Tue, Feb 19, 2013 at 11:24 AM, Aashish Chaudhary
>>> >>>> >> >> >>>>>>>>> <aashish.chaudhary at kitware.com> wrote:
>>> >>>> >> >> >>>>>>>>>> Bill,
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>> I will have a look at it today.
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>> Thanks,
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>> On Mon, Feb 18, 2013 at 11:55 AM, Bill Lorensen
>>> >>>> >> >> >>>>>>>>>> <bill.lorensen at gmail.com> wrote:
>>> >>>> >> >> >>>>>>>>>>> Aashish,
>>> >>>> >> >> >>>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>> TestSplitViewportStereoHorizontal is the only test
>>> >>>> >> >> >>>>>>>>>>> failing
>>> >>>> >> >> >>>>>>>>>>> on
>>> >>>> >> >> >>>>>>>>>>> the Mac
>>> >>>> >> >> >>>>>>>>>>> Gerrit build (and all OSX systems). It would be
>>> >>>> >> >> >>>>>>>>>>> great
>>> >>>> >> >> >>>>>>>>>>> to
>>> >>>> >> >> >>>>>>>>>>> have
>>> >>>> >> >> >>>>>>>>>>> it
>>> >>>> >> >> >>>>>>>>>>> fixed. David Gobbi fixed some failures of this
>>> >>>> >> >> >>>>>>>>>>> test in
>>> >>>> >> >> >>>>>>>>>>> this
>>> >>>> >> >> >>>>>>>>>>> gerrit
>>> >>>> >> >> >>>>>>>>>>> patch:  http://review.source.kitware.com/#/c/9747/
>>> >>>> >> >> >>>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>> If you try to fix it, please make sure you update
>>> >>>> >> >> >>>>>>>>>>> to
>>> >>>> >> >> >>>>>>>>>>> master
>>> >>>> >> >> >>>>>>>>>>> before
>>> >>>> >> >> >>>>>>>>>>> doing any work.
>>> >>>> >> >> >>>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>> Thanks,
>>> >>>> >> >> >>>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>> Bill
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>>
>>> >>>> >> >> >>>>>>>>>> --
>>> >>>> >> >> >>>>>>>>>> | Aashish Chaudhary
>>> >>>> >> >> >>>>>>>>>> | R&D Engineer
>>> >>>> >> >> >>>>>>>>>> | Kitware Inc.
>>> >>>> >> >> >>>>>>>>>> | www.kitware.com
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>>
>>> >>>> >> >> >>>>>>>>> --
>>> >>>> >> >> >>>>>>>>> | Aashish Chaudhary
>>> >>>> >> >> >>>>>>>>> | R&D Engineer
>>> >>>> >> >> >>>>>>>>> | Kitware Inc.
>>> >>>> >> >> >>>>>>>>> | www.kitware.com
>>> >>>> >> >> >>>>>>>>
>>> >>>> >> >> >>>>>>>>
>>> >>>> >> >> >>>>>>>>
>>> >>>> >> >> >>>>>>>> --
>>> >>>> >> >> >>>>>>>> Unpaid intern in BillsBasement at noware dot com
>>> >>>> >> >> >>>>>>>
>>> >>>> >> >> >>>>>>>
>>> >>>> >> >> >>>>>>>
>>> >>>> >> >> >>>>>>> --
>>> >>>> >> >> >>>>>>> | Aashish Chaudhary
>>> >>>> >> >> >>>>>>> | R&D Engineer
>>> >>>> >> >> >>>>>>> | Kitware Inc.
>>> >>>> >> >> >>>>>>> | www.kitware.com
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> --
>>> >>>> >> >> >>>>>> | Aashish Chaudhary
>>> >>>> >> >> >>>>>> | R&D Engineer
>>> >>>> >> >> >>>>>> | Kitware Inc.
>>> >>>> >> >> >>>>>> | www.kitware.com
>>> >>>> >> >> >>>>>> _______________________________________________
>>> >>>> >> >> >>>>>> Powered by www.kitware.com
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> Visit other Kitware open-source projects at
>>> >>>> >> >> >>>>>> http://www.kitware.com/opensource/opensource.html
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>>> Follow this link to subscribe/unsubscribe:
>>> >>>> >> >> >>>>>> http://www.vtk.org/mailman/listinfo/vtk-developers
>>> >>>> >> >> >>>>>>
>>> >>>> >> >> >>>>
>>> >>>> >> >> >>>>
>>> >>>> >> >> >>>>
>>> >>>> >> >> >>>> --
>>> >>>> >> >> >>>> | Aashish Chaudhary
>>> >>>> >> >> >>>> | R&D Engineer
>>> >>>> >> >> >>>> | Kitware Inc.
>>> >>>> >> >> >>>> | www.kitware.com
>>> >>>> >> >> >>>
>>> >>>> >> >> >>>
>>> >>>> >> >> >>>
>>> >>>> >> >> >>> --
>>> >>>> >> >> >>> | Aashish Chaudhary
>>> >>>> >> >> >>> | R&D Engineer
>>> >>>> >> >> >>> | Kitware Inc.
>>> >>>> >> >> >>> | www.kitware.com
>>> >>>> >> >> >>
>>> >>>> >> >> >>
>>> >>>> >> >> >>
>>> >>>> >> >> >> --
>>> >>>> >> >> >> | Aashish Chaudhary
>>> >>>> >> >> >> | R&D Engineer
>>> >>>> >> >> >> | Kitware Inc.
>>> >>>> >> >> >> | www.kitware.com
>>> >>>> >> >> >
>>> >>>> >> >> >
>>> >>>> >> >> >
>>> >>>> >> >> > --
>>> >>>> >> >> > Unpaid intern in BillsBasement at noware dot com
>>> >>>> >> >
>>> >>>> >> >
>>> >>>> >> >
>>> >>>> >> >
>>> >>>> >> > --
>>> >>>> >> > Unpaid intern in BillsBasement at noware dot com
>>> >>>> >>
>>> >>>> >>
>>> >>>> >>
>>> >>>> >> --
>>> >>>> >> | Aashish Chaudhary
>>> >>>> >> | R&D Engineer
>>> >>>> >> | Kitware Inc.
>>> >>>> >> | www.kitware.com
>>> >>>> >
>>> >>>> >
>>> >>>> >
>>> >>>> >
>>> >>>> > --
>>> >>>> > Unpaid intern in BillsBasement at noware dot com
>>> >>>>
>>> >>>>
>>> >>>>
>>> >>>> --
>>> >>>> | Aashish Chaudhary
>>> >>>> | R&D Engineer
>>> >>>> | Kitware Inc.
>>> >>>> | www.kitware.com
>>> >>>
>>> >>>
>>> >>>
>>> >>>
>>> >>> --
>>> >>> Unpaid intern in BillsBasement at noware dot com
>>> >>
>>> >>
>>> >>
>>> >>
>>> >> --
>>> >> Unpaid intern in BillsBasement at noware dot com
>>> >
>>> >
>>> >
>>> > --
>>> > | Aashish Chaudhary
>>> > | R&D Engineer
>>> > | Kitware Inc.
>>> > | www.kitware.com
>>>
>>>
>>>
>>> --
>>> | Aashish Chaudhary
>>> | R&D Engineer
>>> | Kitware Inc.
>>> | www.kitware.com
>>
>>
>>
>>
>> --
>> Unpaid intern in BillsBasement at noware dot com
>
>
>
>
> --
> Unpaid intern in BillsBasement at noware dot com

-- 
| Aashish Chaudhary
| R&D Engineer
| Kitware Inc.
| www.kitware.com