Jep Howard, that's what I got. I now sent it to support and they said it's a
bug, logged as Bug 8655.
And for everyone to try at home ;):
set cut_paste_input [stack 0]
version 8.0 v3
CheckerBoard2 {
inputs 0
format "1024 1024 0 0 1024 1024 1 square_1K"
name CheckerBoard4
selected true
xpos 10177
ypos -129
}
set N62265a0 [stack 0]
push $N62265a0
push $N62265a0
push $N62265a0
push $N62265a0
push $N62265a0
SphericalTransform {
inputs 6
input Cube
format "2048 1024 0 0 2048 1024 1 2k1k"
fix true
name SphericalTransform7
selected true
xpos 10177
ypos -47
}
set Ncb640c0 [stack 0]
Expression {
expr0 (0.5+x)/width
expr1 (0.5+y)/height
expr2 0
expr3 1
name Expression1
selected true
xpos 10098
ypos 4
}
SphericalTransform {
input "Lat Long map"
ry 180
format "2048 1024 0 0 2048 1024 1 2k1k"
fix true
name SphericalTransform21
selected true
xpos 10098
ypos 44
}
push $Ncb640c0
STMap {
inputs 2
uv rgb
name STMap3
selected true
xpos 10197
ypos 44
}
push $Ncb640c0
SphericalTransform {
input "Lat Long map"
ry 180
format "2048 1024 0 0 2048 1024 1 2k1k"
fix true
name SphericalTransform22
selected true
xpos 10318
ypos 50
}
Merge2 {
inputs 2
operation difference
name Merge6
selected true
xpos 10282
ypos 118
}
Howard Jones wrote on 06.08.2014 15:14:
> Make sure the centre of the uv map is 0.5 for each pixel not 0
>
> Ie (x+0.5)/width, (y+0.5)/height I think
> I have the right expression at work.
>
> Howard
>
>> On 6 Aug 2014, at 08:12 pm, "Patrick Heinen" <[email protected]>
>> wrote:
>>
>> Hey Ryan,
>>
>> the UV map I have is created in Nuke and even not filtering it doesn't get
>> rid
>> of that one line :(
>> And blurring only makes it worse in my case.
>> Appreciate your help though.
>> cheers,
>> Patrick
>>
>> Ryan O'Phelan wrote on 04.08.2014 21:27:
>>
>>> When you render uv maps, don't filter or antialias them, and you will not
>>> have edging issues. Adding bilateral or gaussian blur to everything but the
>>> edges helps too.
>>> Hope that helps.
>>>
>>> R
>>>
>>>> On Aug 3, 2014 3:12 PM, "Patrick Heinen" <[email protected]
>>>> <mailto:[email protected]> > wrote:
>>>>
>>>> I'm generating the STMap with (x+0.5)/width and (y+0.5/height). It doesn't
>>>> seem to be off by a pixel, as everything but that one pixel line is spot
>>>> on.
>>>> As for filtering, I have a case where it shouldn't filter at all as pixels
>>>> are getting mapped one to one. I attached an example script that shows the
>>>> issue.
>>>> Thanks for your help!
>>>>
>>>>
>>>>> On 03.08.2014, at 02:51, [email protected]
>>>>> <mailto:[email protected]> wrote:
>>>>>
>>>>>
>>>>> couldn't it be a filtering or multisample problem?
>>>>>
>>>>> I also had similar problems using STMap for tiling an image, and I ended
>>>>> up
>>>>> using UVTile
>>>>>
>>>>>
>>>>>> On Sun, Aug 3, 2014 at 12:26 AM, Deke Kincaid <[email protected]
>>>>>> <mailto:[email protected]> > wrote:
>>>>>>
>>>>>> How are you generating the STMap? It might be off by a pixel depending
>>>>>> how you generated it.
>>>>>>
>>>>>>
>>>>>> --
>>>>>>
>>>>>>
>>>>>> Deke Kincaid
>>>>>>
>>>>>> Creative Specialist
>>>>>>
>>>>>> The Foundry
>>>>>>
>>>>>> Skype: dekekincaid
>>>>>>
>>>>>> Tel: (310) 399 4555 <tel:%28310%29%20399%204555> - Mobile: (310) 883
>>>>>> 4313
>>>>>>
>>>>>> Web: www.thefoundry.co.uk <http://www.thefoundry.co.uk/>
>>>>>>
>>>>>> Email: [email protected] <mailto:[email protected]>
>>>>>>
>>>>>>
>>>>>>> On Fri, Aug 1, 2014 at 7:30 PM, Patrick Heinen
>>>>>>> <[email protected] <mailto:[email protected]>
>>>>>>> >
>>>>>>> wrote:
>>>>>>> Hey everyone,
>>>>>>>
>>>>>>> I'm using an STMap where the original left and right edge of the image
>>>>>>> touch each other in the image center. When the filtering is set to cubic
>>>>>>> on the STMap I get a one pixel line at that edge, that I don't get if I
>>>>>>> ran through the transforms without an STMap. Switching the filtering to
>>>>>>> Impulse, the line disappears, but obviously everything else looks
>>>>>>> crappy.
>>>>>>> I checked the values of the STMap and the source and destination pixels
>>>>>>> seem to be correct.
>>>>>>> Has anyone had this problem before? Or has someone maybe an idea, what
>>>>>>> could cause this?
>>>>>>> Thanks and enjoy your weekend!
>>>>>>>
>>>>>>> Patrick
>>>>>>> _______________________________________________
>>>>>>> Nuke-users mailing list
>>>>>>> [email protected]
>>>>>>> <mailto:[email protected]> ,
>>>>>>> http://forums.thefoundry.co.uk/ <http://forums.thefoundry.co.uk/>
>>>>>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>>>>>>> <http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Nuke-users mailing list
>>>>>> [email protected]
>>>>>> <mailto:[email protected]> ,
>>>>>> http://forums.thefoundry.co.uk/ <http://forums.thefoundry.co.uk/>
>>>>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>>>>>> <http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Nuke-users mailing list
>>>>> [email protected]
>>>>> <mailto:[email protected]> ,
>>>>> http://forums.thefoundry.co.uk/ <http://forums.thefoundry.co.uk/>
>>>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>>>>> <http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users>
>>>>
>>>>
>>>> _______________________________________________
>>>> Nuke-users mailing list
>>>> [email protected]
>>>> <mailto:[email protected]> ,
>>>> http://forums.thefoundry.co.uk/ <http://forums.thefoundry.co.uk/>
>>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>>>> <http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users>
>>>
>> _______________________________________________
>> Nuke-users mailing list
>> [email protected], http://forums.thefoundry.co.uk/
>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
> _______________________________________________
> Nuke-users mailing list
> [email protected], http://forums.thefoundry.co.uk/
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>
_______________________________________________
Nuke-users mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
