Re: [Nuke-users] GPU FFT?

[Magno Borgo]

A non-nuke alternative would be to pre-render the FFT using a python  
package like scipy or similar, which already have fast FFT  
implementations, then load/read it inside NUKE. That might help on some  
pipelines, though the inverse edited FFT would still be needed to handled  
inside Nuke to be practical.





On Mon, 20 Mar 2017 11:09:30 -0400, jon parker   
wrote:



Thanks for the responses! I'm not a full-time comper so am unaware of
some of the tricks out there, like using ZDefocus.  And I should have
added that large kernels / images are involved so the standard
convolve node falls behind at the resolution we are working with.

Magno, I wasn't aware of your blinkscript.  If I have some time I'll
peek at the code.

Someone should put together a commercial implementation for Nuke some
day, sounds like there is demand out there for it.

-Jon

On Mon, Mar 20, 2017 at 7:18 AM, Magno Borgo  wrote:
I coded a naive Blinkscript DCT (and inverse) implementation a while  
back,

you can check the code and use as a start for a better implementation.

http://www.nukepedia.com/blink/other/dct-discrete-cosine-transform

Magno.





On Mon, 20 Mar 2017 03:39:40 -0400, Mads Lund   
wrote:


we did some CPU vs FFT tests aswell, but I can't remember the exact  
kernel
size where FFT started to be more efficient, but it was rather low. On  
the

flip side Convolve on the GPU seem to have some memory problems and (at
least for us) cause some random out of memory issues, even on beefy  
cards.


We found that using gaming techniques: up-scaling low frequency areas  
(then

convolving) and only full convolving of high frequency areas to be to be
quite efficient for the vast majority of our renders where the plate is
gigantic or a large kernel size is needed.
The result is perceptually indistinguishable.




man. 20. mar. 2017 kl. 06.14 skrev Deke Kincaid :


The convolve node was fixed up quite a while a go(4-5 years ago, it is
what the zdefocus is based on).  In our tests though we still find the  
FFT
nodes faster on the farm vs convolve in CPU mode.  If you have a gpu  
farm

then convolve is faster.


On Sun, Mar 19, 2017 at 6:23 PM Michael Habenicht 
wrote:


The latest version of the convolve node has gpu support.

Am 20.03.2017 um 01:20 schrieb jon parker:
> Greetings Nuke users,
>
> I'm just wondering if there are any faster, more robust FFT tools
> available for Nuke besides the (hidden) built-in nodes?
>
> The built-ins do the job, but they are pretty slow and definitely
> prone to crashing fairly often.
>
> Or, alternatively, something that does fast image convolution, some
> other way, under the hood could work too.
>
> Cheers,
> Jon
> ___
> Nuke-users mailing list
> Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


--
Best regards. Mads Hagbarth Lund




--
Using Opera's mail client: http://www.opera.com/mail/

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] GPU FFT?

[Magno Borgo]

I coded a naive Blinkscript DCT (and inverse) implementation a while  
back,  you can check the code and use as a start for a better  
implementation.


http://www.nukepedia.com/blink/other/dct-discrete-cosine-transform

Magno.





On Mon, 20 Mar 2017 03:39:40 -0400, Mads Lund  wrote:

we did some CPU vs FFT tests aswell, but I can't remember the exact  
kernel size where FFT started to be more efficient, but it was rather  
low. On the flip side Convolve >on the GPU seem to have some memory  
problems and (at least for us) cause some random out of memory issues,  
even on beefy cards.


We found that using gaming techniques: up-scaling low frequency areas  
(then convolving) and only full convolving of high frequency areas to be  
to be quite efficient for >the vast majority of our renders where the  
plate is gigantic or a large kernel size is needed.

The result is perceptually indistinguishable.




man. 20. mar. 2017 kl. 06.14 skrev Deke Kincaid :
The convolve node was fixed up quite a while a go(4-5 years ago, it is  
what the zdefocus is based on).  In our tests though we still find the  
FFT nodes faster on the >>farm vs convolve in CPU mode.  If you have a  
gpu farm then convolve is faster.



On Sun, Mar 19, 2017 at 6:23 PM Michael Habenicht   
wrote:

The latest version of the convolve node has gpu support.

Am 20.03.2017 um 01:20 schrieb jon parker:

Greetings Nuke users,

I'm just wondering if there are any faster, more robust FFT tools
available for Nuke besides the (hidden) built-in nodes?

The built-ins do the job, but they are pretty slow and definitely
prone to crashing fairly often.

Or, alternatively, something that does fast image convolution, some
other way, under the hood could work too.

Cheers,
Jon
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

--Best regards. Mads Hagbarth Lund




--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Starting Nuke 9 or 10 from a single configuration

[Magno Borgo]

Check for the version and load what you want.
I usually do that for compiled plugins.

Example
#init.py
#J_ops

if nuke.NUKE_VERSION_MAJOR == 8:
   nuke.pluginAddPath('/Users/magnoborgo/.nuke/J_Ops_v8/')
if nuke.NUKE_VERSION_MAJOR == 9:
nuke.pluginAddPath('/Users/magnoborgo/.nuke/J_Ops_v9/')



On Wed, 01 Mar 2017 10:14:38 -0500, John Coldrick  
 wrote:


This is a challenge we have and I'm getting slow reactions from Foundry  
so I thought I would ask the Braintrust.  :) > Essentially I'm just  
wondering if you can if/then your way out of this problem, can I start  
either 9 or 10 from a single init/>menu.py and end up with properly  
sourced plugins?  The biggest single challenge we have is we're trying  
to do with *without* >using $NUKE_PATH, which drives a different init.py.


Not really asking for a walkthrough, but just wondering if anyone  
happens to know if it's trivial/tough/impossible to do.  If >the  
recommended way really is not using a single configuration but wrapping  
in NUKE_PATH, that's fine.  It really seems to be an >issue of the  
startup order.


Appreciate any thoughts!  We're on Windows if it matters.

Thanks

J.C.





--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Offset project size scale for Roto work

[Magno Borgo]

It should be possible to scale it by just transforming the "root" layer,  
if that doesn't work, you could also try to bake transforms and then scale  
on a fresh node (you can try this to bake:  
https://github.com/magnoborgo/RotoPaintToSplineWarp2)



Magno.




On Fri, 16 Dec 2016 11:33:57 -0500, Michael Hodges  
 wrote:


I need to merge two rather complicated legacy scripts into one nuke  
project for reasons I’d rather not go into.


One script’s project settings was UHD (3840 x 2160) and the other was  
created in 1920 x 1280.


Of course the script that was created in 1920x1280 has all of it’s roto  
curves and scalings all misaligned within the new UHD Project.  Is there  
a good approach to offset roto curves and transformations values to  
match a new global scale in order to realign or am I totally screwed?  I  
was really hoping that an axis type node could scale it all up or down  
uniformly.



Thanks,

Michael___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Importing functions for menu.py

[Magno Borgo]

Also your quotes on the end of line are not the same  '‘


On Sat, 22 Oct 2016 09:50:15 -0400, Daniel Hartlehnert   
wrote:


Ok, just as i sent the follow-up question, your answer comes in, spooky  
:O



Am 22.10.2016 um 15:27 schrieb Justin GD :

You need to put it as a string when you have arguments in the function;  
So it doesn't execute when Nuke starts, but only when the user click on  
the >>menu.


nuke.menu("Nodes").addCommand("User/FiletypeConverter",  
'dh_pycode.fileconverter(my_args)', '‘)



No need to put it as a string when there is no args I believe.



Cheers,

J

2016-10-22 13:58 GMT+01:00 Bruno-Pierre Jobin :

Don't you have to pass dh_pycode.fileconverter as a string?

nuke.menu("Nodes").addCommand("User/FiletypeConverter",  
'dh_pycode.fileconverter', '‘)


--
Bruno-Pierre Jobin

On Oct 22, 2016, at 7:08 AM, Daniel Hartlehnert   
wrote:


nuke.menu("Nodes").addCommand("User/FiletypeConverter",  
dh_pycode.fileconverter, '‘)

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users






--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] nuke locking up on auto save

[Magno Borgo]

When the project file size is too huge I've seen this behavior. If that's  
the case, you'll need to reduce file size, pre-rendering part's of you  
comp. Huge amount of  roto shapes or paints can be the cause of trouble  
sometimes..


Magno.





On Tue, 20 Sep 2016 18:54:37 -0400, adam jones  wrote:


Hey hey

I am pretty sure this has cropped up previousl;y, just cant find the  
posts.


Ever time nuke goes to auto save every thing freezes up and get the  
spinning disk.


am on windows 10

dual xeons

96 gig of ram

nVidia 1080

nuke 10v4

any suggestions

regards
-adam




--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] off-topic- RV show window frame masks

[Magno Borgo]

Thanks Bruno.


On Thu, 18 Aug 2016 12:51:56 -0400, Bruno-Pierre Jobin   
wrote:



Matte Aspect = ctrl + m
Matte Opacity = shift + m

--
Bruno-Pierre Jobin


On Aug 18, 2016, at 12:10 PM, Magno Borgo  wrote:

Hi guys, not extremely Nuke related but anyway:

Is there a way to  make RV show window frame masks (lines/overlays)  
like in Nuke? (2.35:1, 1.85:1 , etc)



Magno.

--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] off-topic- RV show window frame masks

[Magno Borgo]

Hi guys, not extremely Nuke related but anyway:

Is there a way to  make RV show window frame masks (lines/overlays) like  
in Nuke? (2.35:1, 1.85:1 , etc)



Magno.

--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Nuke v10 rotopaint error

[Magno Borgo]

Ops, they did it again.

Python access to b-splines points is broken on Nuke v10, which basically  
leaves you unable to do any manipulation with Python on the rotoshapes.


Please weight in  (supp...@thefoundry.co.uk) on the logged error so we can  
have it fixed:


error: 215774
B-spline ShapeControlPoint no longer accessible via python


If you use any of my scripts (rotopaint to splinewarp, convert roto to  
trackers, silhouette exporter and etc), stay in v9 for now.


Magno.
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] file output from topmost filename

[Magno Borgo]

Hey Gary!


Use  join, lrange and split the filename using "_" to remove the _  
frame numbering


brief explanation:

file_name_

split _>>  file name  (creates the list without _ )

lrange 0 end-1  >> file name (select the elements from list)

join "_" >> file_name (then you join them again using the underscore  
string)








[join [lrange [split [lindex [split [lindex [split [knob [topnode].file]  
.] 0] /] end] _] 0 end-1] "_"]_conv.%04d.exr






On Thu, 26 May 2016 14:06:42 -0400, Gary Jaeger   
wrote:



Hello. I’m interested in this

http://www.nukepedia.com/tcl/write-nodes-filename-from-topmost-read

but the problem I have is the same as somebody in that thread. If we’re  
reading a sequence with frame numbers, the frame number is included in  
the file name:


ex:
fu/bar/FILE_NAME/FILE_NAME_.exr

I get

fu/bar/FILE_NAME/FILE_NAME_.mov

but I want

fu/FILE_NAME.mov

any tips on how to modify that TCL to dow what I want?
[file dirname [knob [topnode].file]]/[lindex [split [lindex [split  
[knob [topnode].file] .] 0] /] end]_conv.%04d.exr




Gary Jaeger / 650.728.7957 direct / 415.518.1419 mobile
http://corestudio.com





--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] extracting concatenated transform matrix

[Magno Borgo]

Hey Michael. I think I have what you want.


I uploaded a script a few months ago that concatenates several transforms  
into a single matrix.

http://www.nukepedia.com/python/misc/concatenate2dtransforms


Magno.



On Wed, 18 May 2016 17:22:37 -0400, Michael Garrett  
 wrote:


Has anyone come up with a nice tool for this yet? I notice that the  
matrix in a 2d node is not concatenating when stacking 2d transforms. So  
I would, roughly >speaking, do something like link the 2d node's matrix  
to an Axis node, then do the concatenation with multiple Axis nodes and  
plug the result into a Card3d.


That's just off the top of my head, I haven't tested it and got it  
working. I know that linking matrices via live python expressions is  
very inefficient as each cel needs to >index into the next value of the  
list of the same matrix, and also there have traditionally been  
evaluation problems on frame update, particularly at render time.


Thanks,
Michael

On 14 July 2015 at 16:54, Frank Rueter|OHUfx  wrote:

Cool, thanks Magno, will check it out


On 14/07/15 22:57, Magno Borgo wrote:

Hi Frank.

I have something alike that you could borrow some of the code to do  
it, its not a complete solution but will help as a start.


The script takes all the transform info from a rotoshape (including  
the rotoshape transforms itself, extra matrix and everything) and the  
layers >>>transforms containing it and bake the transforms onto each  
roto point.


https://github.com/magnoborgo/RotoPaintToSplineWarp2/blob/master/RotopaintToSplineWarp_v2.py

with some adaptation you could make it run into the selected transform  
nodes/cpins  and apply it to a layer in a rotonode.


Magno.






On Tue, 14 Jul 2015 03:25:45 -0400, Frank Rueter|OHUfx  
 wrote:



Hi people,

I was wondering if anybody has a clever tool for this:
I have heaps of shots that need hand tracking, so my usual trees ends  
up having a bunch of concatenating transforms/corner pins etc.
I then sometimes find myself wanting to link the resulting transform  
into a RotoPaint node, so would need to concatenate the nodes and  
extract >>>>the final matrix to then link into the RotoPaint node.


Seeing Transform node's matrices are accessible via python I thought  
I'd ask if anybody has written such a thing?



Cheers,
frank


P.S.: Anybody going to Siggraph?


-- >>>> >>>>vfx compositing | workflow customisation and consulting






--Using Opera's mail client: http://www.opera.com/mail/


-- >> >>vfx compositing | workflow customisation and consulting




___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users






--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] tcl - add version of nuke script

[Magno Borgo]

Assuming you use names like "yourfilename_v001.nk"

you can also use a code like this to strip the version of it:


[lindex [split [file rootname [file tail [value root.name]]] _] end]












On Thu, 28 Apr 2016 05:51:08 -0400, Frank Harrison  
 wrote:



There's some stuff on versioned .nk files here:

https://www.thefoundry.co.uk/products/nuke/developers/100/pythondevguide/asset.html?highlight=version


hth

On 28 April 2016 at 10:33, Darren Coombes  wrote:

More like the script version, as in "comp_v01.1001.dpx"


Check out some of my work...

www.vimeo.com/darrencoombes/reel2015>>

Mob:  +61 418 631 079Instagram: @7secondstoblack

Instagram: @durwood0781
Skype:  darren.coombes81
Twitter:  @durwood81

On 28 Apr 2016, at 7:28 pm, Frank Harrison   
wrote:


the `version` keyword should work

http://www.nukepedia.com/reference/Tcl/group__tcl__builtin.html#gaa49b0ca2e81bbbee6257040db631cf36

On 28 April 2016 at 07:10, Darren Coombes   
wrote:
how would I add just the version of the nuke script to the end of a  
write node?




Check out some of my work...

www.vimeo.com/darrencoombes/reel2015
Mob:  +61 418 631 079Instagram:  
@7secondstoblack

Instagram: @durwood0781
Skype:  darren.coombes81
Twitter:  @durwood81


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users




--Frank Harrison
Senior Nuke Software Engineer
The Foundry
Tel: +44 (0)20 7968 6828 - Fax: +44 (0)20 7930 8906Web:  
www.thefoundry.co.uk
Email: frank.harri...@thefoundry.co.uk  
___

Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users




--Frank Harrison
Senior Nuke Software Engineer
The Foundry
Tel: +44 (0)20 7968 6828 - Fax: +44 (0)20 7930 8906Web:  
www.thefoundry.co.uk

Email: frank.harri...@thefoundry.co.uk




--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] extracting concatenated transform matrix

[Magno Borgo]

Hi Frank.

I have something alike that you could borrow some of the code to do it,  
its not a complete solution but will help as a start.


The script takes all the transform info from a rotoshape (including the  
rotoshape transforms itself, extra matrix and everything) and the layers  
transforms containing it and bake the transforms onto each roto point.


https://github.com/magnoborgo/RotoPaintToSplineWarp2/blob/master/RotopaintToSplineWarp_v2.py

with some adaptation you could make it run into the selected transform  
nodes/cpins  and apply it to a layer in a rotonode.


Magno.






On Tue, 14 Jul 2015 03:25:45 -0400, Frank Rueter|OHUfx   
wrote:



Hi people,

I was wondering if anybody has a clever tool for this:
I have heaps of shots that need hand tracking, so my usual trees ends up  
having a bunch of concatenating transforms/corner pins etc.
I then sometimes find myself wanting to link the resulting transform  
into a RotoPaint node, so would need to concatenate the nodes and  
extract the final matrix to then >link into the RotoPaint node.


Seeing Transform node's matrices are accessible via python I thought I'd  
ask if anybody has written such a thing?



Cheers,
frank


P.S.: Anybody going to Siggraph?


-- > >vfx compositing | workflow customisation and consulting





--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] splinewarp

[Magno Borgo]

Yeah the "freezespline" still works, but there'san additionall step where  
you need to duplicate the shapes manually.

 I did a update on it to work on Nuke v8/v9

https://github.com/magnoborgo




On Mon, 22 Jun 2015 19:40:30 -0400, Pat Wong  wrote:


Magno

Damn thats not good. Hopefully i can see they migth bug fix that. Do you  
know if yor freeze splice waro trick still works on nuke 8 or nuke9, i  
always found that very >useful


Kind Regards


Patrick Wong
077961 35224
www.wahwahdigital.com


On 18 June 2015 at 19:58, Magno Borgo  wrote:
Hi Pat. Since Nuke 7 you can't link the source and dest by python  
anymore... you need to do it by hand (or go back to nuke 6)



I have some tickets that would love some more votes to help do it via  
python, hit support please.


Bug 35393 – Unable to create splinewarp curve via python  
_curveknob.Shape

Bug 34896 – a way to join splinewarp shapes via Python







On Thu, 18 Jun 2015 11:08:00 -0400, Pat Wong   
wrote:



Hi Magno.



I hope you are well. You helkped me a few years back and i urgently  
need it again for a 'morphing job'.




im tryign to ressurect this tool and get it working in nuke 8 but cant  
seem to get the part working whicj links the source to destination  
splice to allow for warping ,
Do yo know if nuke 8 9 allows for proper importing of roto splines  
yet..



thank



Kind Regards


Patrick Wong
077961 35224
www.wahwahdigital.com


On 16 August 2012 at 04:39, Magno Borgo  wrote:

Nice touch, thanks Pete.

And Pat, glad to help :)




Hi Pat and Magno. regarding the "#remove the line below" caveat, if
you end the script with:
#begin code

if __name__ == '__main__':
   freezeWarp2()

#end code

then you don't have to worry about removing that line.

Pete



On Thu, Aug 16, 2012 at 10:58 AM, Pat Wong   
wrote:


great i figured it out,

i didnt remove the additonal line at the bottom of your python  
script

when i was running inthe script editor;




#remove the line below if you are using the script with menu.py (in
menus/DAG/etc)
freezeWarp2()




Its working precisely how i need it to now...

Thanks again Magno you are a life saver!!!! I owe you one.




On 15 August 2012 23:39, Magno Borgo  wrote:
The name of the shapes should be exactly  "source" and "dest"  
without quotes

for the script to work.

Take a look on this one:



set cut_paste_input [stack 0]
version 6.3 v4
push $cut_paste_input
SplineWarp3 {
 boundary_bbox false
 output morph

 curves {AnimTree: "" {
 Version: 1.2
 Flag: 0
 RootNode: 1
 Node: {
  NodeName: "Root" {
   Flag: 512
   NodeType: 1
   Transform: 0 0 S 0 0 S 0 0 S 0 0 S 0 1 S 0 1 S 0 0 S 0 1024 S 0  
778

   NumOfAttributes: 10
   "vis" S 0 1 "opc" S 0 1 "mbo" S 0 1 "mb" S 0 1 "mbs" S 0 0.5  
"fo" S 0 1

"fx" S 0 0 "fy" S 0 0 "ff" S 0 1 "ft" S 0 0
  }
  NumOfChildren: 1
  Node: {
   NodeName: "source" {
Flag: 576
NodeType: 3
CurveGroup: "" {
 Flag: 0
 NumOfCubicCurves: 2
 CubicCurve: "" {
  Type: 0 Flag: 8192 Dim: 2
  NumOfPoints: 12
  0 S 100 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0  
13 0 14 0
15 0 16 0 17 0 18 0 19 0 20 0 21 0 22 0 23 0 24 0 25 0 26 0 27 0  
28 0 29 0
30 0 31 0 32 0 33 0 34 0 35 0 36 0 37 0 38 0 39 0 40 0 41 0 42 0  
43 0 44 0
45 0 46 0 47 0 48 0 49 0 50 0 51 0 52 0 53 0 54 0 55 0 56 0 57 0  
58 0 59 0
60 0 61 0 62 0 63 0 64 0 65 0 66 0 67 0 68 0 69 0 70 0 71 0 72 0  
73 0 74 0
75 0 76 0 77 0 78 0 79 0 80 0 81 0 82 0 83 0 84 0 85 0 86 0 87 0  
88 0 89 0
90 0 91 0 92 0 93 0 94 0 95 0 96 0 97 0 98 0 99 0 100 0 0 S 100 1  
0 2 0 3 0
4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18  
0 19 0 20
0 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32 0 33 0  
34 0 35 0
36 0 37 0 38 0 39 0 40 0 41 0 42 0 43 0 44 0 45 0 46 0 47 0 48 0  
49 0 50 0
51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 0 61 0 62 0 63 0  
64 0 65 0
66 0 67 0 68 0 69 0 70 0 71 0 72 0 73 0 74 0 75 0 76 0 77 0 78 0  
79 0 80 0
81 0 82 0 83 0 84 0 85 0 86 0 87 0 88 0 89 0 90 0 91 0 92 0 93 0  
94 0 95 0
96 0 97 0 98 0 99 0 100 0 0 0 0 S 100 1 435 2 435 3 435 4 435 5  
435 6 435 7
435 8 435 9 435 10 435 11 435 12 435 13 435 14 435 15 435 16 435  
17 435 18
435 19 435 20 435 21 435 22 435 23 435 24 435 25 435 26 435 27 435  
28 435 29
435 30 435 31 435 32 435 33 435 34 435 35 435 36 435 37 435 38 435  
39 435 40
435 41 435 42 435 43 435 44 435 45 435 46 435 47 435 48 435 49 435  
50 435 51
435 52 435 53 435 54 435 55 435 56 435 57 435 58 435 59 435 60 435  
61 435 62
435 63 435 64 435 65 435 66 435 67 435 68 435 69 435 70 435 71 435  
72 435 73
435 74 435 75 435 76 435 77 435 78 435 79 435 80 435 81 435 82 435  
83 435 84
435 85 435 86 435 87 435 88 435 89 435 90 435 91 435 92 435 93 435  
94 435 95
435 96 435 97 435 98 435 99 435 1

Re: [Nuke-users] splinewarp

[Magno Borgo]

Hi Pat. Since Nuke 7 you can't link the source and dest by python  
anymore... you need to do it by hand (or go back to nuke 6)



I have some tickets that would love some more votes to help do it via  
python, hit support please.


Bug 35393 – Unable to create splinewarp curve via python _curveknob.Shape
Bug 34896 – a way to join splinewarp shapes via Python







On Thu, 18 Jun 2015 11:08:00 -0400, Pat Wong  wrote:


Hi Magno.



I hope you are well. You helkped me a few years back and i urgently need  
it again for a 'morphing job'.




im tryign to ressurect this tool and get it working in nuke 8 but cant  
seem to get the part working whicj links the source to destination  
splice to allow for warping ,

Do yo know if nuke 8 9 allows for proper importing of roto splines yet..


thank



Kind Regards


Patrick Wong
077961 35224
www.wahwahdigital.com


On 16 August 2012 at 04:39, Magno Borgo  wrote:

Nice touch, thanks Pete.

And Pat, glad to help :)




Hi Pat and Magno. regarding the "#remove the line below" caveat, if
you end the script with:
#begin code

if __name__ == '__main__':
   freezeWarp2()

#end code

then you don't have to worry about removing that line.

Pete



On Thu, Aug 16, 2012 at 10:58 AM, Pat Wong  wrote:


great i figured it out,

i didnt remove the additonal line at the bottom of your python script
when i was running inthe script editor;




#remove the line below if you are using the script with menu.py (in
menus/DAG/etc)
freezeWarp2()




Its working precisely how i need it to now...

Thanks again Magno you are a life saver I owe you one.




On 15 August 2012 23:39, Magno Borgo  wrote:
The name of the shapes should be exactly  "source" and "dest"  
without quotes

for the script to work.

Take a look on this one:



set cut_paste_input [stack 0]
version 6.3 v4
push $cut_paste_input
SplineWarp3 {
 boundary_bbox false
 output morph

 curves {AnimTree: "" {
 Version: 1.2
 Flag: 0
 RootNode: 1
 Node: {
  NodeName: "Root" {
   Flag: 512
   NodeType: 1
   Transform: 0 0 S 0 0 S 0 0 S 0 0 S 0 1 S 0 1 S 0 0 S 0 1024 S 0  
778

   NumOfAttributes: 10
   "vis" S 0 1 "opc" S 0 1 "mbo" S 0 1 "mb" S 0 1 "mbs" S 0 0.5 "fo"  
S 0 1

"fx" S 0 0 "fy" S 0 0 "ff" S 0 1 "ft" S 0 0
  }
  NumOfChildren: 1
  Node: {
   NodeName: "source" {
Flag: 576
NodeType: 3
CurveGroup: "" {
 Flag: 0
 NumOfCubicCurves: 2
 CubicCurve: "" {
  Type: 0 Flag: 8192 Dim: 2
  NumOfPoints: 12
  0 S 100 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0 13  
0 14 0
15 0 16 0 17 0 18 0 19 0 20 0 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28  
0 29 0
30 0 31 0 32 0 33 0 34 0 35 0 36 0 37 0 38 0 39 0 40 0 41 0 42 0 43  
0 44 0
45 0 46 0 47 0 48 0 49 0 50 0 51 0 52 0 53 0 54 0 55 0 56 0 57 0 58  
0 59 0
60 0 61 0 62 0 63 0 64 0 65 0 66 0 67 0 68 0 69 0 70 0 71 0 72 0 73  
0 74 0
75 0 76 0 77 0 78 0 79 0 80 0 81 0 82 0 83 0 84 0 85 0 86 0 87 0 88  
0 89 0
90 0 91 0 92 0 93 0 94 0 95 0 96 0 97 0 98 0 99 0 100 0 0 S 100 1 0  
2 0 3 0
4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0  
19 0 20
0 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32 0 33 0  
34 0 35 0
36 0 37 0 38 0 39 0 40 0 41 0 42 0 43 0 44 0 45 0 46 0 47 0 48 0 49  
0 50 0
51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 0 61 0 62 0 63 0 64  
0 65 0
66 0 67 0 68 0 69 0 70 0 71 0 72 0 73 0 74 0 75 0 76 0 77 0 78 0 79  
0 80 0
81 0 82 0 83 0 84 0 85 0 86 0 87 0 88 0 89 0 90 0 91 0 92 0 93 0 94  
0 95 0
96 0 97 0 98 0 99 0 100 0 0 0 0 S 100 1 435 2 435 3 435 4 435 5 435  
6 435 7
435 8 435 9 435 10 435 11 435 12 435 13 435 14 435 15 435 16 435 17  
435 18
435 19 435 20 435 21 435 22 435 23 435 24 435 25 435 26 435 27 435  
28 435 29
435 30 435 31 435 32 435 33 435 34 435 35 435 36 435 37 435 38 435  
39 435 40
435 41 435 42 435 43 435 44 435 45 435 46 435 47 435 48 435 49 435  
50 435 51
435 52 435 53 435 54 435 55 435 56 435 57 435 58 435 59 435 60 435  
61 435 62
435 63 435 64 435 65 435 66 435 67 435 68 435 69 435 70 435 71 435  
72 435 73
435 74 435 75 435 76 435 77 435 78 435 79 435 80 435 81 435 82 435  
83 435 84
435 85 435 86 435 87 435 88 435 89 435 90 435 91 435 92 435 93 435  
94 435 95
435 96 435 97 435 98 435 99 435 100 435 0 S 100 1 870 2 870 3 870 4  
870 5
870 6 870 7 870 8 870 9 870 10 870 11 870 12 870 13 870 14 870 15  
870 16 870
17 870 18 870 19 870 20 870 21 870 22 870 23 870 24 870 25 870 26  
870 27 870
28 870 29 870 30 870 31 870 32 870 33 870 34 870 35 870 36 870 37  
870 38 870
39 870 40 870 41 870 42 870 43 870 44 870 45 870 46 870 47 870 48  
870 49 870
50 870 51 870 52 870 53 870 54 870 55 870 56 870 57 870 58 870 59  
870 60 870
61 870 62 870 63 870 64 870 65 870 66 870 67 870 68 870 69 870 70  
870 71 870
72 870 73 870 74 870 75 870 76 870 77 870 78 870 79 870 80 870 81  

Re: [Nuke-users] Rendering Apple Pro res 4444

[Magno Borgo]

Thank you J and Deke. It was in fact in Alexa colorspace, problem solved!

Magno.


On Fri, 22 May 2015 21:59:08 -0400, J Bills  wrote:

Hmmm... doesn't sound like your typical gamma 2.2 colorspace.  What  
colorspace is the extended range quicktime being read in with?  I would  
match that - probably >LogC.


On Fri, May 22, 2015 at 5:13 PM, Magno Borgo  wrote:
So I got this source file which is Quicktime Apple Pro res   
encoded, with a nice range of information, with some superwhites of  
values around 16 on some bright >>parts of the video.


Thought I can't render out a proper Apple Pro res  with my comp  
modifications, It's clipping whites at 1, what I'm missing?

Any help, nuke 8 or 9? Is it even possible?



--Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users







--
Using Opera's mail client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Rendering Apple Pro res 4444

[Magno Borgo]

So I got this source file which is Quicktime Apple Pro res  encoded,  
with a nice range of information, with some superwhites of values around  
16 on some bright parts of the video.


Thought I can't render out a proper Apple Pro res  with my comp  
modifications, It's clipping whites at 1, what I'm missing?

Any help, nuke 8 or 9? Is it even possible?



--
Using Opera's mail client: http://www.opera.com/mail/
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Removing light flicker

[Magno Borgo]

Awesome, thanks for the tips. Most methods got similar results so far, and  
Frank's suggestion is ,as someone else said, free.


Apparently what I have here is a problem on the electronic diafragm of the  
DSLR, not exactly light flickering (light source is stable)
and most methods get 80-90% of the flicker solved. Probably I will have to  
mix a lot of tricks if it doesn't get solved properly (freezing frames,  
local and global deflicker)


Magno.




Martin


On 27 Apr, 2014, at 7:52 AM, Frank Rueter|OHUfx  wrote:

I often find that the easiest deflicker setup works best and fastest  
for global flicker:
use a frame hold to and divide a blurred version of it by the moving  
footage that is blurred by the same amount. Than multiply the result  
back onto the original footage:


Group {
 inputs 0
 name Group1
 label "flickering footage"
 selected true
 xpos -329
 ypos -406
}
 CheckerBoard2 {
  inputs 0
  name CheckerBoard1
  xpos -166
  ypos -317
 }
 Transform {
  rotate {{frame*10}}
  scale 1.8
  center {1024 778}
  name Transform1
  xpos -166
  ypos -245
 }
 Multiply {
  value {{random(frame)}}
  name Multiply1
  xpos -166
  ypos -219
 }
 Shuffle {
  alpha white
  name Shuffle1
  xpos -166
  ypos -181
 }
 Output {
  name Output1
  xpos -166
  ypos -81
 }
end_group
Dot {
 name Dot2
 selected true
 xpos -295
 ypos -304
}
set N6b18140 [stack 0]
Dot {
 name Dot3
 selected true
 xpos -65
 ypos -304
}
Blur {
 size 2000
 name Blur1
 label "adjust to balance deflicker and ghosting"
 selected true
 xpos -99
 ypos -275
}
set N6bf6420 [stack 0]
Dot {
 name Dot1
 selected true
 xpos 80
 ypos -259
}
FrameHold {
 first_frame 1
 name FrameHold1
 label "frame at which to retain luminance"
 selected true
 xpos 46
 ypos -223
}
Dot {
 name Dot4
 selected true
 xpos 80
 ypos -173
}
push $N6bf6420
Merge2 {
 inputs 2
 operation divide
 name Merge1
 selected true
 xpos -99
 ypos -177
}
push $N6b18140
Merge2 {
 inputs 2
 operation multiply
 name Merge2
 selected true
 xpos -329
 ypos -177
}



On 27/04/14 08:42, Magno Borgo wrote:
Hi! Looking for tips and tricks on removing light flicker from stop  
motion/timelapse animation in post, any leads?

Nuke preferably, but also open to plugins and other options.

Thanks in advance.




--
	vfx compositing | workflow customisation and  
consulting ___

Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Removing light flicker

[Magno Borgo]

Hi! Looking for tips and tricks on removing light flicker from stop  
motion/timelapse animation in post, any leads?

Nuke preferably, but also open to plugins and other options.

Thanks in advance.


--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] copying translate values from transform node to spline warp point

[Magno Borgo]

I developed a script that converts rotopaint to splinewarp nodes, baking  
transforms, including an option to convert curves to "pin points"


https://github.com/magnoborgo/RotoPaintToSplineWarp2

You could just copy the transform animation to a rotopaint layer  
(transform animation), create the point where you need and convert to a  
pin, that should help you.





Hi all

I am trying to copy / link values from a transform node to individual  
points on a spline warp curve.


I have tried to do this in every way shape and form but on each occasion  
the point get moved to the lower left corner of the frame


they do in fact have the translation on them but all in the lower left  
corner, this occurs no mater where the points start out.


also the first position of the transform values are the same as the as  
the potion of the point in the spline curve.


just a final note each point gets a translation from a different  
transform node.


cheers
-adam


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Re: 8k limit on EXR output?

[Magno Borgo]

I've seen these when you do extremely huge transforms... but like you said, your image is already huge.Try to crop the comp reducing the bounding boxes.Maybe nothing to do with EXR, from the console:R  9| [17:49.30] Warning: Shuffle2: request() clipped from xywh=1914,0,65647,9953 to xywh=1914,0,8192,8192
R 10| [17:49.30] Warning: Clamp3: request() clipped from xywh=1917,-3,8192,8198 to xywh=1917,0,8192,8192
R 11| [17:49.30] Warning: Merge53: request() clipped from xywh=0,0,67558,11250 to xywh=0,0,8192,8192
R 12| [17:49.30] Warning: Merge23: request() clipped from xywh=0,-575,69000,11825 to xywh=0,0,69000,11250Sorry for my ignorance, but are there any likely suspects for this?
JulesOn Wed, Oct 9, 2013 at 5:56 PM, Jules Stevenson  wrote:
Hey List,We're trying to output a huge image (63k * 11k), which, quite rightly, is really hurting nuke. To simplify this we're trying to split the image up into chunks 8 horizontal chunks (8096 *11250px), pre-render them and re-stitch again. We're using exrs to store a load of additional layers we're using. All works fine except the exrs seem to only ever output to a maximum of 8k. The exr loads in with the correct bounding box, but anything above 8096 pixels is black.

I've never heard of any kind of limit on this, and I've been bashing my head against a wall for 2 days with this - has anyone experienced anything like this, or have any potential ideas?

Many thanks,Jules

-- Magno BorgoCertified Nuke Trainerwww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Nuke .fxs (Silhouette Shape format) exporter

[Magno Borgo]

Thank you guys.Give it a run, we've been testing it for a while and it has been very stable, if you run into any issues let me know.And yeah J, lots of hours on it. At first I tryed to run it just on the Silhouette side, but when I realized that I would need to implement a lot of the Nuke animation curve system in there, I decided to move it to Nuke. Lets say that now I can look at Nuke rotopaint code and see the curves, "the Matrix" style hehe.Magno.awesome - thanks magno!  that looks like it took a lot of work.On Sat, Aug 10, 2013 at 8:15 PM, Sri Kanth <srikanth@gmail.com> wrote:

Great..Thank you..!On Sunday, August 11, 2013, Diogo Girondi  wrote:

Nice one Magno!
On Sat, Aug 10, 2013 at 11:18 AM, David Yu <dave...@gmail.com> wrote:


Good stuff! 
Sent from Galaxy Note 2.
On Aug 10, 2013 10:08 PM, "Magno Borgo" <li...@borgo.tv> wrote:





Hello! For those knee deep into rotoscope:
I'm releasing an exporter script that will send shapes from Nuke to Silhouette, keeping tracking data.

More details at:
http://github.com/magnoborgo/NukeFXSExporter
http://www.youtube.com/watch?feature=player_embedded&v=Xn9oegH8umo


-- 
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/




http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/

http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
-- Sent from iPad 
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
-- Magno BorgoCertified Nuke Trainerwww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Nuke .fxs (Silhouette Shape format) exporter

[Magno Borgo]

Hello! For those knee deep into rotoscope:
I'm releasing an exporter script that will send shapes from Nuke to  
Silhouette, keeping tracking data.


More details at:
http://github.com/magnoborgo/NukeFXSExporter
http://www.youtube.com/watch?feature=player_embedded&v=Xn9oegH8umo


--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] sublime text

[Magno Borgo]

Good point Nathan.Magno.


`reload()` can definitely be useful, but keep in mind that it does not 
re-bind external references to a module. If moduleA imports moduleB, running 
`reload(moduleB)` will not update moduleA’s reference to it, so any calls 
moduleA makes to objects in moduleB will still be using old code. In order to 
update moduleA’s reference, you would need to reload it explicitly: 
`reload(moduleA.moduleB)`.
 
-Nathan


 
 

From: Magno Borgo 
Sent: Thursday, August 08, 2013 11:55 AM
To: Nuke user discussion 

Subject: Re: [Nuke-users] sublime text
 

Something that I use here when coding/testing my python scripts is the 
python reload() function. It will always import the latest modification saved to 
your script file.
No need to copy/pasteback and forth and if you want you can even assign 
this to a keyboard shortcut.
 
Something like this would work on the script editor:
 
nuke.pluginAddPath('/path/to/script/folder/')  # need to run this just 
1 time.
import yourscriptfile #need to run this just 1 time.
 
reload(yourscriptfile)
yourscriptfile.runthedesiredfunctionhere()
 
 
 
 
 
 

  Ah, many thanks! 
  Will look into this.
   
  // Fredrik
   
  
  On Fri, Jul 26, 2013 at 6:05 PM, Deke Kincaid <d...@thefoundry.co.uk> wrote:
  
Pretty much what Sam said.  Here’s the link. 
 
https://github.com/Nvizible/NukeExternalControl


 
-Deke KincaidCreative SpecialistThe FoundryMobile: 
(310) 883 4313Tel: (310) 399 4555 - Fax: (310) 
450 4516The Foundry Visionmongers Ltd.Registered in England 
and Wales No: 4642027


On Tue, Jul 23, 2013 at 6:14 AM, Sam Marrocco <smarro...@ringsidecreative.com> 
wrote:

  
  
  
  
  On 7/23/2013 4:40 AM, Fredrik Averpil wrote:
  

 
I was wondering, does Nuke have a command port, similar to what 
Maya has – enabling me to send code from my favourite editor to be 
executed inside of Nuke (without having to copy-paste the code back and 
forth)?
 I've been using a 
  project called nukeExternalControl (GitHub) to inter-app communicate 
  between python and nuke. It isn't fast but seems like it would probably do 
  the job for you provided that Nuke's python interface allows a "paste" 
  into it's editor. It definitely can run python commands internally on Nuke 
  from as external interpreter. Internally it seems to use 
  sockets.sam marrocco | chief 
  technical officer ringside.cutters.picnic.moonlink248 548 2500 w248 910 
  3344 cringsidecreative.com
   
  ___Nuke-users 
  mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
 ___Nuke-users 
mailing listNuke-users@support.thefoundry.co.uk, 
http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
   

-- 
Magno BorgoCertified Nuke 
Trainerwww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= 
-3


___Nuke-users mailing 
listNuke-users@support.thefoundry.co.uk, 
http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users-- Magno BorgoCertified Nuke Trainerwww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] sublime text

[Magno Borgo]

Something that I use here when coding/testing my python scripts is the python reload() function. It will always import the latest modification saved to your script file.No need to copy/pasteback and forth and if you want you can even assign this to a keyboard shortcut.Something like this would work on the script editor:nuke.pluginAddPath('/path/to/script/folder/')  # need to run this just 1 time.import yourscriptfile #need to run this just 1 time.reload(yourscriptfile)yourscriptfile.runthedesiredfunctionhere()Ah, many thanks!Will look into this.// FredrikOn Fri, Jul 26, 2013 at 6:05 PM, Deke Kincaid  wrote:

Pretty much what Sam said.  Here’s the link.https://github.com/Nvizible/NukeExternalControl



-
Deke Kincaid
Creative Specialist
The Foundry
Mobile: (310) 883 4313
Tel: (310) 399 4555 - Fax: (310) 450 4516

The Foundry Visionmongers Ltd.
Registered in England and Wales No: 4642027
On Tue, Jul 23, 2013 at 6:14 AM, Sam Marrocco  wrote:










On 7/23/2013 4:40 AM, Fredrik Averpil wrote:



I was wondering, does Nuke have a command port, similar to what Maya has – enabling me to send code from my favourite editor to be executed inside of Nuke (without having to copy-paste the code back and forth)?









I've been using a project called nukeExternalControl (GitHub) to inter-app communicate between python and nuke. It isn't fast but seems like it would probably do the job for you provided that Nuke's python interface allows a "paste" into it's editor. It definitely
 can run python commands internally on Nuke from as external interpreter. Internally it seems to use sockets.



sam marrocco | chief technical officer 
ringside.cutters.picnic.moonlink





248 548 2500 w
248 910 3344 c

ringsidecreative.com




___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
-- Magno BorgoCertified Nuke Trainerwww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Regrain blend on semi-transparent areas

[Magno Borgo]

Thank you Shailendra / Misho, nice to see other workflows.
Yeah thinking about it gamma correcting the grain matte will not mess with  
white and black points, i'll give it a try.















I'd just gamma up/down the matte in step 3 (you keep the 0 and 1)

m.

On 02.07.2013 15:40, Shailendra Pandey wrote:

Hi Magno,
 Thats a good question.
every time you composite a premultiplied image A over B
for pixels with alpha =1 you get A
 for pixels with alpha =0 you get B
 for pixels with 0average(based on the alpha) of the two input imagesNow if you remember  
averaging can be used to reduce grain e.g.

spatial(blurs) or temporal(frameblend)
 There are a few ways you can address the problem of soft grain on the
edges of your patch.
 1. composite your patch with soft alpha on the plate without grain2.  
accumulate all your mattes for grain in separate stream3. process your  
matte for grain e.g. compress range say using white

point in a grade(alpha) node and check black/white clamp
4. Use this matte to add grain on your comp.
 Hope that helps
 Cheers
SHAILENDRA
 On Tue, Jul 2, 2013 at 10:14 PM, Magno Borgo  wrote:


Looking for your experience/tips on the following situation:
 Removal area/clean plate, regrained to match the background plate.
It has an alpha channel, blurred to smooth transition.
But when its composited Over the original plate (heavy grain), on
the semi-transparent alpha region the grain is faint because of the
 blending.
 Creating an edge detection mask and grading just the grain of the
cleanplate on the transition area is working (making it stronger),
just wondering about other tips/techniques on this situation.
 --
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com [1]
www.borgo.tv [2]
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk,
http://forums.thefoundry.co.uk/ [3]
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
[4]

   Links:
--
[1] http://www.boundaryvfx.com
[2] http://www.borgo.tv
[3] http://forums.thefoundry.co.uk/
[4] http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
 ___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users





--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Regrain blend on semi-transparent areas

[Magno Borgo]

Looking for your experience/tips on the following situation:

Removal area/clean plate, regrained to match the background plate. It has  
an alpha channel, blurred to smooth transition.
But when its composited Over the original plate (heavy grain), on the  
semi-transparent alpha region the grain is faint because of the  blending.



Creating an edge detection mask and grading just the grain of the  
cleanplate on the transition area is working (making it stronger), just  
wondering about other tips/techniques on this situation.



--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] nuke render memory management

[Magno Borgo]

Try rendering on terminal with -c flag"-c 2048M" for example will limit memory usage to 2GBI'm running out of memory on the farm on one of my comps. I don't remember running into this kind of thing with a comp of this size on nuke 6.3x and I'm wondering if I have missed something. 

I've precomped all the branches that I can. I've also turned off all postage stamps. I've also tried turning down the cache memory usage to 10%, along with turning all memory settings to very low.
Renders consistently run out of ram on the farm. The farm nodes have 16GB RAM. On my comp station, this comp eats 22GB of RAM, until I clear the buffers.
It's a somewhat large comp. 781 nodes. 762 channels. About 75 read nodes (reads are CG passes)HD inputs are between 2MB/frame and 8MB/frameRendering zip1 16 half EXR
Environment:Windows 7Nuke 7.0v6same config on farm, except 16GB RAMLooking at my buffer report, I'm noticing a large amount of ChannleMerge nodes reporting memory usage of 300-400MB each. For example:
ChannelMerge
	total usage: 	339MB
	user data: 
	cacheRegion	1920x1080
	channels [... lots of channels listed here ...]
	usage	100%
	lossage	19
	weight	0x0another example:ShuffleCopy
	total usage: 	247MB	user data: 	cacheRegion	1928x1088	channels	extraTex_vrayUserColor_restaurant01.g,nigel.body,props.dump,lloyd.body,linus,extraTex_vrayUserColor_set03.r,reflect.red,reflect.green,reflect.blue,refract.red,refract.green,refract.blue,specular.red,specular.green,specular.blue,GI.red,GI.green,GI.blue,SSS.r,SSS.g,SSS.b,diffuse.r,diffuse.g,diffuse.b,lighting.red,lighting.green,lighting.blue,extraTex_vrayUserColor_interiors.g,extraTex_vrayUserColor_restaurant02.r,extraTex_vrayUserColor_set01.g,extraTex_vrayUserColor_set01.b,extraTex_vrayUserColor_set02.r,extraTex_vrayUserColor_set02.g,extraTex_vrayUserColor_set03.g,gush.a,normWorld.r,normWorld.g,normWorld.b,gg.body,gg.hair,bldg.green,bldg.blue,bldg.alpha,lloyd.stripe
	usage	100%





	weight	0x0Is this normal?Why are there lots of channels listed in the ChannelMerge node, when it should only be evaluating two input channels, and outputing one channel?
Why is this comp so heavy?What am I missing here? Thanks for your help,
Ryan

-- Magno BorgoCertified Nuke Trainerwww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] presenting the FreezeSplineWarp technique

[Magno Borgo]

Thanks everyone, glad that you liked.

Help me improve the scripts/technique by requesting priority to the  
following tickets with Foundry's support:


Bug 35393 – Unable to create splinewarp curve via python _curveknob.Shape
Bug 34896 – a way to join splinewarp shapes via Python

Magno.





Thank you for taking the time. Appreciated it.

Sent from my iPhone

On Jun 16, 2013, at 10:16 AM, "Magno Borgo"  wrote:

Hello everyone, doing a small plug for the video that I created  
describing what I call the "FreezeSplineWarp" technique:

http://youtu.be/rFP4jgfXpjM

With Corner Pins your are limited to 4 points stabilize/match-move, but  
with this technique you can use any number of points, getting more  
finer control. I show also the python scripts that I developed to  
automate the whole process on Nuke v7.


I hope that you like it.


--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] presenting the FreezeSplineWarp technique

[Magno Borgo]

Hello everyone, doing a small plug for the video that I created describing  
what I call the "FreezeSplineWarp" technique:

http://youtu.be/rFP4jgfXpjM

With Corner Pins your are limited to 4 points stabilize/match-move, but  
with this technique you can use any number of points, getting more finer  
control. I show also the python scripts that I developed to automate the  
whole process on Nuke v7.


I hope that you like it.


--
Magno Borgo
Certified Nuke Trainer
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] read svg (as roto shapes)

[Magno Borgo]

Its kind of convoluted but I have done it before:SVG > Illustrator > After Effects Masks > Export masks to Silhouette > Silhouette to Nuke.holaAnyone managed to read svg (or other standardised vector format) into rotoshapes?Or maybe there is some neat (cmdline) tool to convert svg to nuke roto
G-- ■ ■ ■ ■ ■ ■ ■ ■ ■ ■

-- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Auto unpack group shapes node

[Magno Borgo]

You'll need to automate it with python either on the export from Silhouete or within Nuke.Looks like you are using an older version of Silhouette (Bezier export?), newer versions can already export layers/shapes into single nodes and do what you want.Hii know  "Edit > Node > Group > Expand Group (Ctrl+alt+g)" this procedure but i want just click 1 button all shapes unpack like 50 character layer in Silhouette i am export for nuke shapes and import in nuke group node ,just click 1 custom button after 1 character all shapes in 1 backdrop all  50 character separate backdrop and shapes if any  possible
--  Thanks

-- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] spline warp correspondence points

[Magno Borgo]

You can do it with Python on the script editor.Edit the "setup" below  following the instructions, the hardest part is to figure out the point number.
#setup
curvename = "Ellipse1" #change to name of the curve
pointNumber = 3 #correspondence point number (starts on 0)
sourceordestination = 1 #sourcecurve is 0, destination curve is 1
time = 40 #time of the keyframe you want to remove
#endofsetup

rotoNode = nuke.selectedNode()
rotoCurve = rotoNode['curves']
cp = rotoCurve.getCorrespondencePoints(curvename)

animationcurve = cp.getAnimCurve(pointNumber, sourceordestination)
animationcurve.removeKey(time)
#also possible to remove all the keyframes with
#animationcurve.removeAllKeys()

rotoCurve.changed() #updates the UI





I'm afraid this isn't currently possible. It's logged as bug 18300.

Please contact supp...@thefoundry.co.uk to add your voice to the
bug!

Thanks,
Sam


On 09/05/2013 19:37, japeurotrash
  wrote:


  
  
  Does anyone know how to remove keyframes
from correspondence points?
I am able to remove them from shapes but not correspondence
points.
I tried thru the dope sheet, the un-keyfame button etc to the
point the timeline doesn't show any keyframes, but my
correspondence points are still moving when scrubbing thru the
timeline.
Any help would be greatly appreciated!

btw, using nuke 7.0.6 on linux
  
  
  
  ___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


-- 
Sam Smith
Nuke QA Engineer
The Foundry, 6th Floor, 48 Leicester Square, London, WC2H 7LT, UK
Tel: +44 (0)207 968 6828, Fax: +44 (0)207 930 8906
Web: www.thefoundry.co.uk
Email: sam.sm...@thefoundry.co.uk

The Foundry Visionmongers Ltd.
Registered in England and Wales No: 4642027 

  -- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] How do I add to Reformat node

[Magno Borgo]

Add this to your init.py:

nuke.addFormat ("3840 1080 1.0 Mynewformat") #width height pixelaspect name



I want to add a specific format to the Reformat node, so that I don't  
have to recreate it every time for a new project.

Is there a way to do this?

Thanks in advance,
Todd Groves


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Magno Borgo
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: AW: [Nuke-users] Alexa Artifacts

[Magno Borgo]

I've seen the exactly same artifacts when working on a film shot on Alexa. These are nasty specially when keying... same issue, shot directly in Proress .Magno.We've been having some problems with noise on some footages from Alexa, but nothing remotely near to that.diogo

On Wed, Mar 27, 2013 at 9:50 PM, Jonathan Egstad  wrote:

No idea, but it looks an awful lot like filtering from a slight resize operation.-jonathan

On Mar 27, 2013, at 5:29 PM, "Igor Majdandzic"  wrote:

do you mean in camera? because that was from the original qt footage

 -- 

igor majdandziccompositor |

i...@badgerfx.com

BadgerFX | www.badgerfx.com

 

Von: nuke-users-boun...@support.thefoundry.co.uk [mailto:nuke-users-boun...@support.thefoundry.co.uk] Im Auftrag von Jonathan Egstad

Gesendet: Donnerstag, 28. März 2013 01:10An: Nuke user discussionCc: Nuke user discussionBetreff: Re: [Nuke-users] Alexa Artifacts

 Looks like a very  slight resize was done.-jonathan

On Mar 27, 2013, at 4:56 PM, "Igor Majdandzic"  wrote:

Hey guys,we got footage from a shoot with Alexa being the camera. It was shot in ProRess 444. The problem is: The picture has some artifacts which confuse me the codec being 444. I attached some images which show some of the grain patterns. Is this normal?

 thx,

    Igor 

  

-- igor majdandzic

compositor |i...@badgerfx.com

BadgerFX | www.badgerfx.com

 Von: nuke-users-boun...@support.thefoundry.co.uk [mailto:nuke-users-boun...@support.thefoundry.co.uk] Im Auftrag von Deke Kincaid

Gesendet: Mittwoch, 27. März 2013 23:47An: Nuke user discussionBetreff: Re: [Nuke-users] FusionI/O and Nuke 

Hi MichaelI'm actually testing this right now as Fusionio just gave us a bunch of them.  Early tests reveal that with dpx it's awesome but with openexr zip compressed file it it is spending more time with compression, not sure if it is cpu bound or what(needs more study but its slower).  Openexr uncompressed files though are considerably superfast but of course the issue is that it is 18 meg a frame.  These are single layer rgba exr files.

-Deke KincaidCreative SpecialistThe FoundryMobile: (310) 883 4313Tel: (310) 399 4555 - Fax: (310) 450 4516

The Foundry Visionmongers Ltd.Registered in England and Wales No: 4642027 On Wed, Mar 27, 2013 at 3:26 PM, Michael Garrett  wrote:

I'm evaluating one of these at the moment and am interested to know if others have got it working with Nuke nicely, meaning, have you been able to really utilise the insane bandwidth of this card to massively accelerate any part of your day to day compositing?

 So far, I've found it has no benefit when localising all Reads in a somewhat heavy comp, or even playing back a sequence of exr's or deep files, compared to localised sequences on a 10K Raptor drive also in my workstation - hopefully I'm missing something big though, this is day one after all.  

 There may be real tangible benefits to putting the Nuke cache on it though - I'll see how it goes.

 I'm also guessing that as gpu processing becomes more prevalent in Nuke that we will see a real speed advantage handing data from a card like this straight to the gpu.

 Thanks,Michael___

Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/

http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users 





___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/

http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/

http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users___


Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
-- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Automatically control Text size

[Magno Borgo]

Perhaps work with a Reformat node after the text to make it fit somehow independently of the Font used (along with the expressions)Magno.Fixed width fonts are great and all, but they're not always what the client wants.Programs that have better Text tools do this. I'm just trying to figure out how to do this in Nuke.

Ron Ganbaremail: ron...@gmail.comtel: +44 (0)7968 007 309 [UK]     +972 (0)54 255 9765 [Israel]url: http://ronganbar.wordpress.com/


On 5 March 2013 12:42, Santiago Svirsky  wrote:







or you could use howard's _expression_ with a fixed-width font, which were invented for such cases.
;)




From: Ron Ganbar [ron...@gmail.com]
Sent: 05 March 2013 10:23
To: Nuke user discussion
Subject: Re: [Nuke-users] Automatically control Text size




Hi guys,-- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -3___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] mask

[Magno Borgo]

As a workaround you can draw (or read an image) make a Group and use it as  
Input Process.





You cannot, the options are hardcoded..

Contact supp...@thefoundry.co.uk and request these be customisable. It's  
a known issue, but the more people that request it, the quicker it'll be  
fixed \o/


On 22/02/13 03:25, Dan Rosen wrote:

Anyone know how to customize the new mask and title safe values in the
Viewer?

thx,
Dan


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users





--
Magno Borgo
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -3
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Tcl addition syntax

[Magno Borgo]

[expr [value which]+1]




Hi

TCL is certainly not my strong point but how do you add to a value?

ie. I have a switch node running from 0-2 which I want to add 1 to in  
the label.


So the label would say version:1 to 3 rather than 0 to 2

Something like
version: [value which +1 ]
but works.

Simple I'm sure but couldn't find anything on the web.

Thanks
Howard___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



--
Magno Borgo
www.boundaryvfx.com
www.borgo.tv
Brasil:Curitiba:GMT= -2 (dst)
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] EXR - Maya 2013 compression - Nuke performance.

[Magno Borgo]

Thanks, I'll check these out.Magnowoops, forgot to include the link to the tool:http://antimentalray.livejournal.comlooks like it has to be compiled though :/

-
Deke Kincaid
Creative Specialist
The Foundry
Mobile: (310) 883 4313
Tel: (310) 399 4555 - Fax: (310) 450 4516

The Foundry Visionmongers Ltd.
Registered in England and Wales No: 4642027
On Thu, Jan 31, 2013 at 11:51 AM, Deke Kincaid <d...@thefoundry.co.uk> wrote:

Mango: Last I knew you had to write an output shader for mental ray to calculate this and it is not done by default.  It is easiest to wrap this up into a post conversion script when you convert the file to scanline.  In Nuke you can hit “x” and type “autocrop” and it will analyze the sequence and create a bbox which you can then write out.


This guy wrote a little tool called exr4nuke which does this automatically for you:There is a plugin for MR in Maya called “Mental Core” which allows you to write out scanline exr’s, embedded camera metadata (not sure about adding the bbox/datawindow).


http://core-cg.com/http://core-cg.com/documentation/refpasses.html#compositing-render-passes


Also from doing some more research, it appears that Mental Ray defaults now to outputting scanline exr files with zips compression.  I’m not sure which version though as even the nvidia people from the thread below think it is possibly in Maya 2013 but not sure if it actually made it into it:


http://forum.nvidia-arc.com/showthread.php?9434-OpenExr-and-zip-scanline-compressions&highlight=nukehttp://forum.nvidia-arc.com/showthread.php?9434-OpenExr-and-zip-scanline-compressions&highlight=nuke


-
Deke Kincaid
Creative Specialist
The Foundry
Mobile: (310) 883 4313
Tel: (310) 399 4555 - Fax: (310) 450 4516

The Foundry Visionmongers Ltd.
Registered in England and Wales No: 4642027
On Thu, Jan 31, 2013 at 10:48 AM, Magno Borgo <li...@borgo.tv> wrote:





Let me ask something related to the thread. My EXRs  coming from maya2013 /mental ray renders do not have bounding box, am I missing some rendering option to get that?


Magno.


Well, as Deke mentioned, this option may just be applying ZIP compression 
to the individual tiles, as opposed to buffering the tile data and then writing 
a a scanline-based image at the end.
 
-Nathan


 -- Magno Borgowww.boundaryvfx.comwww.borgo.tv


Brasil:Curitiba:GMT= -2 (dst)___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

-- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -2 (dst)___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] EXR - Maya 2013 compression - Nuke performance.

[Magno Borgo]

Let me ask something related to the thread. My EXRs  coming from maya2013 /mental ray renders do not have bounding box, am I missing some rendering option to get that?Magno.


Well, as Deke mentioned, this option may just be applying ZIP compression 
to the individual tiles, as opposed to buffering the tile data and then writing 
a a scanline-based image at the end.
 
-Nathan


 -- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -2 (dst)___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] default for tracker label

[Magno Borgo]

Add this to you menu.py or init.pynuke.knobDefault("Tracker.label", '[knob transform]')Does anyone know how you would get the tracker node to display the value of the transform knob in the label so it's default?I usually do it by typing [knob transform] in the label section in the node tab. But it gets annoying having to do it every time.Thanks.
Darren Coombes - VFX CompositorCheck out some of my work…www.vimeo.com/53990514Mob:  +61 418 631 079Skype:  darrencoombesTwitter:  @durwood81

-- Magno Borgowww.boundaryvfx.comwww.borgo.tvBrasil:Curitiba:GMT= -2 (dst)___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Puppet Tool Deformation

[Magno Borgo]

You can achieve this with splinewap (there are pins as well), but its not "exactly" the same as the AE one, because on AE you have a triangulated mesh that you can deform/refine. Not impossible but probably not as easy to setup in Nuke.Hi Everyone,Is somebody know if they is a way to simulate the AE puppet tools inside Nuke ?The AE puppet tool is a kind of partial warp node which deforming part of an image according to the positions of pins that you place and move. Locators (pins) define what parts of the image should move, what parts should remain rigid, and what parts should be in front when parts overlap.

ThanksSpider Luddnel Spider Magne | Director - Lead Motion Compositor

555Lab – Alchemy between you and us 

104 rue de la Folie Méricourt 75011 Paris Office (+33)148 075 055

Mobile (+33)699 434 555 | 555lab.com



-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Fading Project 3d Shaders

[Magno Borgo]

I have several project3d shaders being merged with MergeMat nodes. Is there a way to fade/"mix" shaders? Like we can do with the merge node "mix" parameter?I can setup each project3d on its own scanline render and fade them, but perhaps there's an easier way?-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics+55.41.9982-5537+55.41.3013-4921Brasil:Curitiba:GMT= -2 (dst)___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Fading Project 3d Shaders

[Magno Borgo]

Nice :)That would be great. Perhaps with some fading animation setup, x frames before, y frames after the framehold.I could add this to projectionist as an option btw...On 4 dec. 2012, at 18:22, "Magno Borgo" <li...@borgo.tv> wrote:Thats it, thanks. Animated multiply node before the project 3d.
-- Julik Tarkhanov | HecticElectric | Keizersgracht 736 1017 EXAmsterdam | The Netherlands | tel. +31 20 330 8250cel. +31 61 145 06 36 | http://hecticelectric.nl


-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Fading Project 3d Shaders

[Magno Borgo]

Thats it, thanks. Animated multiply node before the project 3d.Does this do it? using a Multiple to fade out the shader before  merging it.On 5 December 2012 05:56, Magno Borgo <li...@borgo.tv> wrote:

I have several project3d shaders being merged with MergeMat nodes. Is there a way to fade/"mix" shaders? Like we can do with the merge node "mix" parameter?



I can setup each project3d on its own scanline render and fade them, but perhaps there's an easier way?



-- 
Magno Borgo

www.borgo.tv
3D | VFX | Motion Graphics
+55.41.9982-5537
+55.41.3013-4921
Brasil:Curitiba:GMT= -2 (dst)
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/


http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Fading Project 3d Shaders

[Magno Borgo]

I have several project3d shaders being merged with MergeMat nodes. Is  
there a way to fade/"mix" shaders? Like we can do with the merge node  
"mix" parameter?


I can setup each project3d on its own scanline render and fade them, but  
perhaps there's an easier way?




--
Magno Borgo

www.borgo.tv
3D | VFX | Motion Graphics
+55.41.9982-5537
+55.41.3013-4921
Brasil:Curitiba:GMT= -2 (dst)
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] nuke and mocha pro workflow problem

[Magno Borgo]

	I am not sure the best workflow to make the parameters between Nuke and Mocha Pro match so that exported shape data will paste into place in the rotopaint node in Nuke

Parameters that need to match:
clip length / frame rate / interlaced node / pulldown mode / PAR

In Nuke, original footage is 1440x1080_50i, but it has a Reformat Node applied (1920x1080) and a Field Select node applied - "even".

In Mocha Pro, same footage imported (1440x1080_50i HDV), but for the Pixel Aspect Ratio, do I go with HDV 1.333 OR do I make the PAR: HD to match the final plate in Nuke?
If you import the original footage, use the footage PAR (matching the 1440/1080=1,). After that apply the same reformat after your roto node.Use the same field setting on both aps or you'll have problems.
I have run tests, but so far, all the shape data comes in to the Left of where*it out to be.*

mocha pro frame offset is 1
Where do I set the frame offset in Nuke?  - to make sure they match?
Change frame offset in mocha, its easier, set it to the same start frame of your nuke comp
Nuke parameters:
clip length: 1-812 frames (in & out points @ 60 and 190)
frame rate: 24 fps
interlaced mode: have been choosing upper field/odd (will switch to even)
pulldown mode:* haven't done anything with this
PAR: HD

Mocha Pro parameters:
clip length: 1-812 (same in & out points)
frame rate: 24 fps
(bit depth is set to 24)
interlaced mode: even
pulldown mode - have not set
PAR: HDV 1.333
(I have also chosen PAR: DVDHD Pro (which as I understand it, is relatively the same as HDV
I have also set up the PAR to HD, but still the same problem)

Any advice or feedback would be greatly appreciated.

Thanks!
	-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Colourblind help

[Magno Borgo]

Maybe using the HueShift node you could rotate colors to something more suitable for you?
	Hello,

I'm a beginner in nuke and a paint and roto artist, I was just wondering if anyone has any tips or tricks in helping picking out colours. I not fully colourblind just a green and red deficiency. 

I do use the rgb colour numbers at the bottom of the screen and shuffle the colours around and flick through the rgb channels. However there's sometimes where where I can't see a certain shade so I would like to know if there is a way of displaying the colour numbers on screen, so that if there is a marker that's hide to find, it displays the higest number.

Thanks
	-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] "Accumulating" alpha

[Magno Borgo]

Nice, thank you guys. TimeEcho will do the trick.

Magno.



timeecho set to max

On Thu, Nov 22, 2012, at 10:53, Magno Borgo wrote:
Lets say I have an alpha channel of a moving object. How can I generate  
a

single image that represents this alpha through time?
This would generate the "trail" of the alpha. Something like frameBlend
but without averaging, I want to max it.



thanks in advance.
--
Magno Borgo

www.borgo.tv
3D | VFX | Motion Graphics
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users






--
Magno Borgo

www.borgo.tv
3D | VFX | Motion Graphics
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] "Accumulating" alpha

[Magno Borgo]

Lets say I have an alpha channel of a moving object. How can I generate a  
single image that represents this alpha through time?
This would generate the "trail" of the alpha. Something like frameBlend  
but without averaging, I want to max it.




thanks in advance.
--
Magno Borgo

www.borgo.tv
3D | VFX | Motion Graphics
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Keylight_ edge issue

[Magno Borgo]

Looks like a chorma subsampling issue.http://en.wikipedia.org/wiki/Chroma_subsamplingYou can use Keylight's "screen preblur" to improve it,or  try changing the Colorspace to YCbCr, blur the CbCr channels and return to the original colorspace  
	The edges in Keyframe is kind of broken (the footage is also not high res)
I played with screen gain but was not able to solve it. The composite is looking good; but edges of the premultiplied fg is ugly. Any ideas??

Thanks
Putopian vohra
	-- Magno Borgowww.borgo.tv3D | VFX | Motion Graphics___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Manipulating the pupil of an eye

[Magno Borgo]

I would do it with a spline-warp: track the pupil, use it to drive a spline and distort it at will with the destination curve.
	Hi, I'm working with footage of an eye and I'm trying to dilate the pupil ever so slightly. The challenge for me is getting the iris to look as it is contracting accordingly. By rendering out a normal shader from Maya (one channel from top to bottom, one from right to left and one with facing ratio) and using it with an IDistort (same method as this guy fakes refraction in Nuke: https://vimeo.com/30058027), I've managed to push the pixels around the pupil. It works to some extent, but doesn't seem like a good way of achieving this. 

How would everyone else go around doing this? Preferably in nuke only.
	-- Using Opera's revolutionary email client: http://www.opera.com/mail/___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] splinewarp

[Magno Borgo]

Nice touch, thanks Pete.

And Pat, glad to help :)



Hi Pat and Magno. regarding the "#remove the line below" caveat, if
you end the script with:
#begin code

if __name__ == '__main__':
freezeWarp2()

#end code

then you don't have to worry about removing that line.

Pete



On Thu, Aug 16, 2012 at 10:58 AM, Pat Wong  wrote:


great i figured it out,

i didnt remove the additonal line at the bottom of your python script
when i was running inthe script editor;




#remove the line below if you are using the script with menu.py (in
menus/DAG/etc)
freezeWarp2()




Its working precisely how i need it to now...

Thanks again Magno you are a life saver I owe you one.




On 15 August 2012 23:39, Magno Borgo  wrote:
> The name of the shapes should be exactly  "source" and "dest" without  
quotes

> for the script to work.
>
> Take a look on this one:
>
>
>
> set cut_paste_input [stack 0]
> version 6.3 v4
> push $cut_paste_input
> SplineWarp3 {
>  boundary_bbox false
>  output morph
>
>  curves {AnimTree: "" {
>  Version: 1.2
>  Flag: 0
>  RootNode: 1
>  Node: {
>   NodeName: "Root" {
>Flag: 512
>NodeType: 1
>Transform: 0 0 S 0 0 S 0 0 S 0 0 S 0 1 S 0 1 S 0 0 S 0 1024 S 0 778
>NumOfAttributes: 10
>"vis" S 0 1 "opc" S 0 1 "mbo" S 0 1 "mb" S 0 1 "mbs" S 0 0.5 "fo"  
S 0 1

> "fx" S 0 0 "fy" S 0 0 "ff" S 0 1 "ft" S 0 0
>   }
>   NumOfChildren: 1
>   Node: {
>NodeName: "source" {
> Flag: 576
> NodeType: 3
> CurveGroup: "" {
>  Flag: 0
>  NumOfCubicCurves: 2
>  CubicCurve: "" {
>   Type: 0 Flag: 8192 Dim: 2
>   NumOfPoints: 12
>   0 S 100 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0 13 0  
14 0
> 15 0 16 0 17 0 18 0 19 0 20 0 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28 0  
29 0
> 30 0 31 0 32 0 33 0 34 0 35 0 36 0 37 0 38 0 39 0 40 0 41 0 42 0 43 0  
44 0
> 45 0 46 0 47 0 48 0 49 0 50 0 51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0  
59 0
> 60 0 61 0 62 0 63 0 64 0 65 0 66 0 67 0 68 0 69 0 70 0 71 0 72 0 73 0  
74 0
> 75 0 76 0 77 0 78 0 79 0 80 0 81 0 82 0 83 0 84 0 85 0 86 0 87 0 88 0  
89 0
> 90 0 91 0 92 0 93 0 94 0 95 0 96 0 97 0 98 0 99 0 100 0 0 S 100 1 0 2  
0 3 0
> 4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0  
19 0 20
> 0 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32 0 33 0 34  
0 35 0
> 36 0 37 0 38 0 39 0 40 0 41 0 42 0 43 0 44 0 45 0 46 0 47 0 48 0 49 0  
50 0
> 51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 0 61 0 62 0 63 0 64 0  
65 0
> 66 0 67 0 68 0 69 0 70 0 71 0 72 0 73 0 74 0 75 0 76 0 77 0 78 0 79 0  
80 0
> 81 0 82 0 83 0 84 0 85 0 86 0 87 0 88 0 89 0 90 0 91 0 92 0 93 0 94 0  
95 0
> 96 0 97 0 98 0 99 0 100 0 0 0 0 S 100 1 435 2 435 3 435 4 435 5 435 6  
435 7
> 435 8 435 9 435 10 435 11 435 12 435 13 435 14 435 15 435 16 435 17  
435 18
> 435 19 435 20 435 21 435 22 435 23 435 24 435 25 435 26 435 27 435 28  
435 29
> 435 30 435 31 435 32 435 33 435 34 435 35 435 36 435 37 435 38 435 39  
435 40
> 435 41 435 42 435 43 435 44 435 45 435 46 435 47 435 48 435 49 435 50  
435 51
> 435 52 435 53 435 54 435 55 435 56 435 57 435 58 435 59 435 60 435 61  
435 62
> 435 63 435 64 435 65 435 66 435 67 435 68 435 69 435 70 435 71 435 72  
435 73
> 435 74 435 75 435 76 435 77 435 78 435 79 435 80 435 81 435 82 435 83  
435 84
> 435 85 435 86 435 87 435 88 435 89 435 90 435 91 435 92 435 93 435 94  
435 95
> 435 96 435 97 435 98 435 99 435 100 435 0 S 100 1 870 2 870 3 870 4  
870 5
> 870 6 870 7 870 8 870 9 870 10 870 11 870 12 870 13 870 14 870 15 870  
16 870
> 17 870 18 870 19 870 20 870 21 870 22 870 23 870 24 870 25 870 26 870  
27 870
> 28 870 29 870 30 870 31 870 32 870 33 870 34 870 35 870 36 870 37 870  
38 870
> 39 870 40 870 41 870 42 870 43 870 44 870 45 870 46 870 47 870 48 870  
49 870
> 50 870 51 870 52 870 53 870 54 870 55 870 56 870 57 870 58 870 59 870  
60 870
> 61 870 62 870 63 870 64 870 65 870 66 870 67 870 68 870 69 870 70 870  
71 870
> 72 870 73 870 74 870 75 870 76 870 77 870 78 870 79 870 80 870 81 870  
82 870
> 83 870 84 870 85 870 86 870 87 870 88 870 89 870 90 870 91 870 92 870  
93 870
> 94 870 95 870 96 870 97 870 98 870 99 870 100 870 0 0 0 S 100 1 0 2 0  
3 0 4
> 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19  
0 20 0
> 21 0 22 0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32 0 33 0 34 0  
35 0
> 36 0 37 0 38 0 39 0 40 0 41 0 42 0 43 0 44 0 45 0 46 0 47 0 48 0 49 0  
50 0
> 51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 0 61 0 62 0 63 0 64 0  
65 0
> 66 0 67 0 68 0 69 0 70 0 71 0 72 0 73 0 74 0 75 0 76 0 77 0 78 0 79 0  
80 0
> 81 0

Re: [Nuke-users] splinewarp

[Magno Borgo]

 0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32  
0 33 0 34 0 35 0 36 0 37 0 38 0 39 0 40 0 41 0 42 0 43 0 44 0 45 0 46 0 47  
0 48 0 49 0 50 0 51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 0 61 0 62  
0 63 0 64 0 65 0 66 0 67 0 68 0 69 0 70 0 71 0 72 0 73 0 74 0 75 0 76 0 77  
0 78 0 79 0 80 0 81 0 82 0 83 0 84 0 85 0 86 0 87 0 88 0 89 0 90 0 91 0 92  
0 93 0 94 0 95 0 96 0 97 0 98 0 99 0 100 0 0 S 100 1 0 2 0 3 0 4 0 5 0 6 0  
7 0 8 0 9 0 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 20 0 21 0 22  
0 23 0 24 0 25 0 26 0 27 0 28 0 29 0 30 0 31 0 32 0 33 0 34 0 35 0 36 0 37  
0 38 0 39 0 40 0 41 0 42 0 43 0 44 0 45 0 46 0 47 0 48 0 49 0 50 0 51 0 52  
0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 0 61 0 62 0 63 0 64 0 65 0 66 0 67  
0 68 0 69 0 70 0 71 0 72 0 73 0 74 0 75 0 76 0 77 0 78 0 79 0 80 0 81 0 82  
0 83 0 84 0 85 0 86 0 87 0 88 0 89 0 90 0 91 0 92 0 93 0 94 0 95 0 96 0 97  
0 98 0 99 0 100 0 0 0 0 L 0 4096 0.5 2 0 0 "" 0 100 "" 1 0 0 0 0 0 256 0  
"" 2 -3.34656 0 0 0 0 256 0 "" 3 -6.67242 0 0 0 0 256 0 "" 4 -9.95697 0 0  
0 0 256 0 "" 5 -13.1794 0 0 0 0 256 0 "" 6 -16.3192 0 0 0 0 256 0 "" 7  
-19.3556 0 0 0 0 256 0 "" 8 -22.2679 0 0 0 0 256 0 "" 9 -25.0354 0 0 0 0  
256 0 "" 10 -27.6375 0 0 0 0 256 0 "" 11 -30.0535 0 0 0 0 256 0 "" 12  
-32.2627 0 0 0 0 256 0 "" 13 -34.2444 0 0 0 0 256 0 "" 14 -35.978 0 0 0 0  
256 0 "" 15 -37.4428 0 0 0 0 256 0 "" 16 -38.618 0 0 0 0 256 0 "" 17  
-39.4832 0 0 0 0 256 0 "" 18 -40.0173 0 0 0 0 256 0 "" 19 -40.2 0 0 0 0  
256 0 "" 20 -36.5222 0 0 0 0 256 0 "" 21 -25.9117 0 0 0 0 256 0 "" 22  
-9.00305 0 0 0 0 256 0 "" 23 13.5692 0 0 0 0 256 0 "" 24 41.1706 0 0 0 0  
256 0 "" 25 73.1666 0 0 0 0 256 0 "" 26 108.923 0 0 0 0 256 0 "&qu

Re: [Nuke-users] splinewarp

[Magno Borgo]

Looks like you source/dest have a different number of keyframes. I forgot  
about that...


Use this script to bake the keyframes (it will create a key on every  
frame),


it converts a rotonode into a splinewarp.
http://www.nukepedia.com/python/misc/rotopainttosplinewarp/

After that run the previous script and it will work.

Magno.


Magno

thanks, Ive followed your instructions,

but i get this  error

# Result:
Traceback (most recent call last):
  File "", line 93, in 
  File "", line 84, in freezeWarp2
IndexError: list index out of range



any ideas?





--
******
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] splinewarp

[Magno Borgo]

Yeah indeed, I needed to do that in a project while ago a wrote a script.
I'll upload it to Nukepedia when I have time to polish it.

Usage:

Create  a splinewarp with 2 curves (ideally with same number of shape  
points, or copy from rotonode)


name the source shape "source"
name the destination shape "dest"

run the script

delete the destination shape. The source shape will have what you want.
any problems let me know.






oh dear not even a work around at all?



On 15 August 2012 16:46, Bill Gilman  wrote:
That's the main criticism so far of the spline warper.  You have to  
build the target shape from the source, you can't copy in a shape and  
link them up.  They know about it, I'm guessing they're working on it.


On Aug 15, 2012, at 8:37 AM, Pat Wong wrote:


Hi

Im trying to copy and paste the splines and animation from 2 seperate
rotospline shape nodes into a splinewarp node , one for the source and
the other for the source warp...


I can seem to get the spline over from the roto node into the spline
warp into the 'SOURCE', thats fine.

but when i try to copy the second roto shape across into the 'SOURCE
WARP' shape... it seems to overwrite the spline and animation in both
the 'SOURCE' and 'SOURCE WARPED'

Does anyone know if this is possible or not??


cheers

pat
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users



___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users




--
**
Magno Borgo

www.borgo.tv
www.boundaryvfx.com

splinewarp_source_dest.py
Description: Binary data
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Using tracker to stable then reapply move problems

[Magno Borgo]

Tracker nodes are correct, the problem is the bounding box.Set to rotoscope node "clip to" to none or bounding box.Hey allI am pretty sure I am just being really dumb here but...I have some rigg to remove so I am am useing the Tracker node (2 point track with (transition, rotation and scale) active on both track points, set my ref frame and set the transform to "stabilize" do the work I need to then copy the original tracker and set it transform to Match move and well it does not replace the move.once again I am sure I am being dumb but if any one has some pointers that would be great.project attachedcheers-adam -- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] PTGui distortion model

[Magno Borgo]

Yeah, thats a great solution. Check these for some nice tips going down  
that road:


http://www.nukepedia.com/written-tutorials/understanding-uvmaps-warping-with-stmap-pt-1/
http://www.nukepedia.com/written-tutorials/understanding-uvmaps-warping-with-stmap-pt-2/

I'm not sure if PTGui allows you to redistort an image based on your  
lens solution, but if it does, you can read in a 32-bit normalized UV  
ramp (0 to 1 in X and 0 to 1 in Y) that is the size of your undistorted  
plate and use the resulting distorted UV image to feed into nuke's lens  
distort node or an STMap.  I've used this technique coming from boujou  
and it's spot on.


John Mangia

Sent from my iPad

On Jun 13, 2012, at 11:20 AM, "Graham Houston"  
 wrote:


If I have distortion parameters from PTGui, (lens correction  
parameters: abc, image shift: de), how do I translate that into Nuke's  
LensDistortion node?
I see the 'Card Parameters' rollout, but the only values that affect  
the image output are the 'Radial Distortion' fields.

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users




--
**
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Clone of Roto Node?

[Magno Borgo]

Place the needed individual roto shape(s) into a single rotonode and use a comp operation to add it to the rest of the tree. If you need it elsewhere, you can pipe out the output to other places.Not convenient as have everything on the same rotonode but its a workaround.
	Hi,
I´d like to reuse animated shape in Roto node, but Nuke keeps telling me, that Roto cannot be cloned.

I know I can copy that shape between rotos, but then when I change the shape, I have to re-copy it to the other node.

How do you gou aboout it?

Thanks!
	-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Adding a tracker to a roto point

[Magno Borgo]

Try my python script:http://www.nukepedia.com/python-scripts/misc/bakerotoshapestotrackers/
	One can right-click a roto point and link to a tracker but can the reverse be done?  I have a rotoshape that is keyframed to a window.  Now I would like to "convert" one of the roto/bezier points to a tracker or get it's X/Y positional data.  Is this possible?

Right-clicking and "copy animation" doesn't work.  It creates a light blue box that is constantly 0.
	-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Flame Timewarp File to Nuke?

[Magno Borgo]

Take a look at http://framecurve.org/scripts/





   Hello,

   
   Does anyone know if there is a tool for Nuke to ingest a Flame

   Timewarp file?  Or something to convert the Timewarp data into
   Time+Ascii format, If one doesn't exist we will create something
   here to do the conversion for our needs, but if something is already
   out there...

   
   Thanks,


   -Schnee

  --
\/ davids / comp \/ 177
/\ tippettstudio /\ breakin'donz






--
******
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] greenscreen dark edges

[Magno Borgo]

If its really a despill problem, maybe try an Unspill? Play with the multiply node to adjust the balance of the rgb channels in the code below.set cut_paste_input [stack 0]
version 6.3 v1
push $cut_paste_input
Dot {
 name Dot11
 selected true
 xpos 483
 ypos -273
}
set N5666ee0 [stack 0]
Dot {
 name Dot9
 selected true
 xpos 309
 ypos -273
}
set N560c330 [stack 0]
Shuffle {
 red blue
 green blue
 alpha blue
 name BLUE
 tile_color 0x
 selected true
 xpos 275
 ypos -222
}
set N560f370 [stack 0]
Dot {
 name Dot19
 selected true
 xpos 69
 ypos -219
}
push $N560c330
Shuffle {
 red green
 blue green
 alpha green
 name GREEN
 tile_color 0xff00ff
 selected true
 xpos 383
 ypos -190
}
set N561b640 [stack 0]
Dot {
 name Dot15
 selected true
 xpos 364
 ypos -145
}
Dot {
 name Dot14
 selected true
 xpos 364
 ypos -27
}
push $N560c330
Shuffle {
 green red
 blue red
 alpha red
 name RED
 tile_color 0xffff
 selected true
 xpos 178
 ypos -193
}
set N5609d50 [stack 0]
Multiply {
 name Multiply3
 selected true
 xpos 178
 ypos -143
}
push $N560f370
Multiply {
 name Multiply2
 selected true
 xpos 275
 ypos -139
}
Merge2 {
 inputs 2
 operation average
 name Merge7
 selected true
 xpos 222
 ypos -88
}
Merge2 {
 inputs 2
 operation difference
 name SPILLMAP
 selected true
 xpos 222
 ypos -30
}
Multiply {
 value {0.2 0.7 0.3 1}
 name Multiply1
 label "CHANGE VALUES HERE\n\n"
 note_font_size 20
 selected true
 xpos 221
 ypos 20
}
Dot {
 name Dot20
 selected true
 xpos 255
 ypos 74
}
set N19645d40 [stack 0]
Dot {
 name Dot21
 selected true
 xpos 183
 ypos 198
}
Merge2 {
 inputs 2
 operation plus
 name Plus1
 selected true
 xpos 35
 ypos 195
}
push $N5609d50
Dot {
 name Dot18
 selected true
 xpos 135
 ypos -121
}
push $N19645d40
Merge2 {
 inputs 2
 operation plus
 name Plus
 selected true
 xpos 101
 ypos 115
}
push $N561b640
Dot {
 name Dot10
 selected true
 xpos 417
 ypos 121
}
push $N19645d40
Merge2 {
 inputs 2
 operation difference
 name Merge9
 selected true
 xpos 221
 ypos 118
}
push $N5666ee0
Dot {
 name Dot12
 selected true
 xpos 491
 ypos 246
}
ShuffleCopy {
 inputs 2
 green green
 name ShuffleCopy1
 selected true
 xpos 221
 ypos 243
}
ShuffleCopy {
 inputs 2
 red red
 name ShuffleCopy2
 selected true
 xpos 101
 ypos 243
}
ShuffleCopy {
 inputs 2
 blue blue
 name ShuffleCopy3
 selected true
 xpos 35
 ypos 309
}

	Hi guys.
I'm using keylight to pull a key. I'm also using Primatte for the core key. Both look ok when channelmerged together. The problem is I end up with dark edges, which I'm guessing are result of despill from keylight. I tried to use an erode but it's eating away some fine detail I want to keep. Anyone has any tips?
Thanks.
	-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Write node - TCL directory navigation question

[Magno Borgo]

I usually place the tcl inside the label of a backdropnode to test the results, in fact I have a toolset with several expressions saved on it:[join [lrange [split [file dirname [knob root.name]] "/"] 0 end-1] "/"]

[file dirname [value root.name]]

[lindex [split [value root.name] /] end]

[lindex [split [lindex [split [value root.name] /] end] .] 0]


[file tail [value root.name]]
[file rootname [file tail [value root.name]]]

[date %Hh%Mm%Ss]I like to append the date one on the filename of the quicktime previews, this way a new preview is always created, instead of overwriting and you can compare versions.Magno.Magno,That's a nice, compact way to disassemble and reassemble pathname components! I have used something similar with split before, but I didn't use join, which makes it a lot shorter than what I was doing!  ;^)One question I have is, how do you debug your evaluated _expression_ to make sure your pathname is what you want? What I've been doing is to attach a Text node after the Write node and view the result of the text message "[value Write1.file]" - or whatever the name of the Write node happens to be... That gives me an overlay of the printed path. When I look at the result of your _expression_, however, it is not evaluated. I get the same code that is typed in the file section - essentially what "[knob Write1.file]" would report. But when I use something like "[file dirname [file dirname [value root.name]]]" - I get the actual pathname that it's going to write to.  Not sure why that would be. Any idea...?Thanks,Rich
Rich BoboSenior VFX CompositorMobile:  (248) 840-2665Web:  http://richbobo.com/

On Apr 26, 2012, at 11:14 AM, Magno Borgo wrote:Try this one.[join [lrange [split [file dirname [knob root.name]] "/"] 0 end-1] "/"]Change "end-1" to end-2, end-3, and so on to go back on the directory structureHoward,Hmm... That just seems to append /../ to the path and doesn't actually perform the change directory command... Unless I'm missing something.Thanks,RichOn Apr 26, 2012, at 9:14 AM, Howard Jones wrote:[file dirname  [value root.name]]/../to be more precise HowardFrom: Howard Jones <mrhowardjo...@yahoo.com>To: Nuke user discussion <nuke-users@support.thefoundry.co.uk> Sent: Thursday, 26 April 2012, 14:11Subject: Re: [Nuke-users] Write node - TCL directory navigation question[file dirname  [value root.name]]/../ [file tail [value root.name]] should do it I think HowardFrom: Rich Bobo <richb...@mac.com>To: Nuke-Users Mailing List <nuke-users@support.thefoundry.co.uk> Sent: Thursday, 26 April 2012, 13:57Subject: [Nuke-users] Write node - TCL directory navigation questionHi,Does anyone know how to indicate a relative directory change in TCL? In a Write node, I'd like to designate the directory to write files to that is parallel to the one the script is in. I'm using this typical form to get the current script directory:[file dirname [file tail [value root.name]]] Then, I'd like to do the equivalent of  'cd ../different_dir'. So, I'd like to have a (hopefully short) chunk of code that would a) grab the script's current directory, b) navigate up one directory and c) indicate the parallel directory to write the files to. Should be easy - I just don't know if it's possible with TCL or if I have to combine some Python in there...Thanks for any help!RichRich BoboSenior VFX CompositorMobile:  (248) 840-2665Web:  http://richbobo.com/___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users--**Magno Borgowww.borgo.tvwww.boundaryvfx.com___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Write node - TCL directory navigation question

[Magno Borgo]

I usually place the tcl inside the label of a backdropnode to test the results, in fact I have a toolset with several expressions saved on it:[join [lrange [split [file dirname [knob root.name]] "/"] 0 end-1] "/"]

[file dirname [value root.name]]

[lindex [split [value root.name] /] end]

[lindex [split [lindex [split [value root.name] /] end] .] 0]


[file tail [value root.name]]
[file rootname [file tail [value root.name]]]

[date %Hh%Mm%Ss]I like to append the date one on the filename of the quicktime previews, this way a new preview is always created, instead of overwriting and you can compare versions.Magno.Magno,That's a nice, compact way to disassemble and reassemble pathname components! I have used something similar with split before, but I didn't use join, which makes it a lot shorter than what I was doing!  ;^)One question I have is, how do you debug your evaluated _expression_ to make sure your pathname is what you want? What I've been doing is to attach a Text node after the Write node and view the result of the text message "[value Write1.file]" - or whatever the name of the Write node happens to be... That gives me an overlay of the printed path. When I look at the result of your _expression_, however, it is not evaluated. I get the same code that is typed in the file section - essentially what "[knob Write1.file]" would report. But when I use something like "[file dirname [file dirname [value root.name]]]" - I get the actual pathname that it's going to write to.  Not sure why that would be. Any idea...?Thanks,Rich
Rich BoboSenior VFX CompositorMobile:  (248) 840-2665Web:  http://richbobo.com/

On Apr 26, 2012, at 11:14 AM, Magno Borgo wrote:Try this one.[join [lrange [split [file dirname [knob root.name]] "/"] 0 end-1] "/"]Change "end-1" to end-2, end-3, and so on to go back on the directory structureHoward,Hmm... That just seems to append /../ to the path and doesn't actually perform the change directory command... Unless I'm missing something.Thanks,RichOn Apr 26, 2012, at 9:14 AM, Howard Jones wrote:[file dirname  [value root.name]]/../to be more precise HowardFrom: Howard Jones <mrhowardjo...@yahoo.com>To: Nuke user discussion <nuke-users@support.thefoundry.co.uk> Sent: Thursday, 26 April 2012, 14:11Subject: Re: [Nuke-users] Write node - TCL directory navigation question[file dirname  [value root.name]]/../ [file tail [value root.name]] should do it I think HowardFrom: Rich Bobo <richb...@mac.com>To: Nuke-Users Mailing List <nuke-users@support.thefoundry.co.uk> Sent: Thursday, 26 April 2012, 13:57Subject: [Nuke-users] Write node - TCL directory navigation questionHi,Does anyone know how to indicate a relative directory change in TCL? In a Write node, I'd like to designate the directory to write files to that is parallel to the one the script is in. I'm using this typical form to get the current script directory:[file dirname [file tail [value root.name]]] Then, I'd like to do the equivalent of  'cd ../different_dir'. So, I'd like to have a (hopefully short) chunk of code that would a) grab the script's current directory, b) navigate up one directory and c) indicate the parallel directory to write the files to. Should be easy - I just don't know if it's possible with TCL or if I have to combine some Python in there...Thanks for any help!RichRich BoboSenior VFX CompositorMobile:  (248) 840-2665Web:  http://richbobo.com/___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users--**Magno Borgowww.borgo.tvwww.boundaryvfx.com___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Write node - TCL directory navigation question

[Magno Borgo]

Try this one. [join [lrange [split [file dirname [knob root.name]] "/"] 0 end-1] "/"]Change "end-1" to end-2, end-3, and so on to go back on the directory structureHoward,Hmm... That just seems to append /../ to the path and doesn't actually perform the change directory command... Unless I'm missing something.Thanks,RichOn Apr 26, 2012, at 9:14 AM, Howard Jones wrote:[file dirname  [value root.name]]/../to be more precise HowardFrom: Howard Jones  To: Nuke user discussion   Sent: Thursday, 26 April 2012,
 14:11 Subject: Re: [Nuke-users] Write node - TCL directory navigation question   [file dirname  [value root.name]]/../ [file tail [value root.name]] should do it I think HowardFrom: Rich Bobo  To: Nuke-Users Mailing List   Sent:
 Thursday, 26 April 2012, 13:57 Subject: [Nuke-users] Write node - TCL directory navigation question   Hi,Does anyone know how to indicate a relative directory change in TCL? In a Write node, I'd like to designate the directory to write files to that is parallel to the one the script is in. I'm using this typical form to get the current script directory:[file dirname [file tail [value root.name]]] Then, I'd like to do the equivalent of  'cd ../different_dir'. So, I'd like to have a (hopefully short) chunk of code that would a) grab the script's current directory, b) navigate up one directory and c) indicate the parallel directory to write the files to. Should be easy - I just don't know if it's possible
 with TCL or if I have to combine some Python in there...Thanks for any help!Rich
Rich BoboSenior VFX CompositorMobile:  (248)
 840-2665Web:  http://richbobo.com/

___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users  ___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users  ___Nuke-users mailing listNuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] RotoPaint knobDefault's for toolbar settings

[Magno Borgo]

type help('_curvelib') on script editor and run to find out the attributes.

ltt = lifetime
opc = opacity
and so on.



Hi Alex, you have to do it like this (for lifetime, for example). They
should really change the way this works, I have a bug id for it: bug
#10544


nuke.knobDefault("RotoPaint.toolbox", '''clone {
{ brush ltt 0}
{ clone ltt 0}
{ blur ltt 0}
{ sharpen ltt 0}
{ smear ltt 0}
{ eraser ltt 0}
{ reveal ltt 0}
{ dodge ltt 0}
{ burn ltt 0}
}''')

On Mon, Apr 9, 2012 at 3:45 PM, Alexander Berson   
wrote:

hello,

Is there a straight forward way of setting the tool bar knobDefaults on
create for the RotoPaint node?

ie

nuke.knobDefault('RotoPaint.toolbar_opacity', '0.05')
nuke.knobDefault('RotoPaint.toolbar_lifetime_type', 'all')

obviously don't work or I'm missing something.. also not sure what the
expression would be if I wanted it to apply these settings only under
certain paint tools inside of the roto paint node.

ie
clone and the eraser tool gets set to : opacity = .05
   lifetime =  
all
other tools get left at their default state of single frame and 1  
opacity.



thanks,


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users







--
**
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] how do i get nuke to say "Render Complete"

[Magno Borgo]

I missing?

** **

** **

Regarding the framepadding, maybe I'm misunderstanding you, but I  
don't

think that's quite what I'm after.

** **

In the python script I have to numerical values, frame number &  
version

number.

I use str(nuke.frame()) to get the frame number, so assuming I'm at  
frame

45 printing nuke.frame() = 45.

However, I want it to use, at least four digits, so 45 should be  
0045.***

*

The same applies for the version number, which I want to be three  
digits.



** **

I know I can write filname_.dpx in the writenode, but
again, that's not really what I'm after

** **

** **

** **

I have already looked at nuke.sample, but there doesn't seem to be a  
way
to fetch the already sampled pixels (shift + ctrl + mouse drag in  
viewer)



having to manually type in x & y coordinates isnt really worth it for  
me.



** **

Again, a huge thanks for you help, I really appreciate it.

** **

On Thu, Mar 29, 2012 at 4:35 PM, Diogo Girondi  


wrote:

To get just the filename you can do something like:

** **

Import os

print os.path.basename( nuke.Root().name() )

** **

Or

** **

nuke.Root().name().split('/')[-1]

** **

You can replace the / for os.sep()

** **

For the TypeError you can simply do int(version)+1 assuming the  
version

variable is assigned to a numerical character. 

** **

To use padded sequences just format your filename string correctly  
using

either  or %04d, I suggest you stick with %04d though. 

** **

To sample RGBA data look into nuke.sample on the docs. 

** **

** **

Cheers,

Diogo



On 29/03/2012, at 11:14, Jon Wesstr?m  wrote:

I'm still very new to python/nuke scripting, so these questions might  
be

dumb, but I'm getting tired of not being able to solve them.

** **

I'm working on a tool that makes it easy to make breakdowns of a  
script,
you basically select a node, press a button, and nuke renders out a  
frame.



** **

That part of the tool is done, but I'm getting stuck on all the stuff
that is generating the filename

** **

1: When pressing the renderbutton, the script spawns a writenode, puts
text in the file knob, renders the frame and then is deleted 

(
  
.removeAfterFrameRender(nuke.delete(nuke.toNode('FrameRender')))   
)



** **

That works fine, but I would also like to add an increment to the
filename, so I figured this would work:  
nuke.removeAfterFrameRender(version

= version + 1), but it gives me this error: 

TypeError: cannot concatenate 'str' and 'int' objects.

I have tried a bunch of different ways to solve it, but nothing is
working.

** **

2: Is there a way to force python/nuke to display number using a set
amount of increments?
Ex: I would prefer the frame do be written as 0015 in the filename,  
and

not 15.

** **

3: Is there a way to access the name of the script?

I know you can write nuke.root().knob('name').getValue(), but that
outputs not only the name, but also the entire filepath.


4: Not related to this, but is there any way to access the rgba data  
that

the viewer sampler samples?
Would like a stickynote that, when created writes down the currently
sampled colors in RGBA.

If you would like to take a look at the entire script:
http://slexy.org/view/s2gqT5q5Sq

Any help is greatly appreciated

/Jon Wesstr?m

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

** **

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users




___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


-- next part --
An HTML attachment was scrubbed...
URL:  
http://support.thefoundry.co.uk/cgi-bin/mailman/private/nuke-users/attachments/20120330/feac781c/attachment-0001.htm


--

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


End of Nuke-users Digest, Vol 49, Issue 29
**



___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users




--
**
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Gathering roto values to drive a tracker

[Magno Borgo]

Yes, via python.I've created a script recently that does that. :)http://www.nukepedia.com/gizmos/bakerotoshapestotrackers/
Hi there,Is there a way to link the translate and rotation values of a single point of a roto shape to a tracker on nuke?  I want to gather the rotation and translation of 2 points of a roto shape and use that information to drive a tracker.Any help will be appreciated Cheers,César Rodríguez B. Visual Effects Artist -- Quote of the week... 
"An elderly man was trying to find a place to sit and observe the Olympic Games, as he went to each section. All the other Greeks laughed as he tried to make his way through. Some ignored him. Upon entering the Spartan section all the Spartans stood and offered the elderly man their seats. Suddenly the entire stadium applauded. All the Greeks knew what was the right thing to do, but the Spartans were the only ones who did it."Xenophon  		 	   		  -- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] Python: Identifying shapes, stroke and layers

[Magno Borgo]

Hi! Python question for the experts:Is there an attribute that will tell me if the element inside a rotopaint node is a Shape, Stroke, or Layer?I'm using the number of attributes  to identify them, i.e Layers have (0-11) attributes , Shapes (42-44) and etc.
The problem is that I'm dealing with shapes imported  from a variety of softwares (Mocha, Silhouette)and sometimes the imported shapes have a weird number of attributes, making this method problematic.-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] OT: Free replacement to Shotgun

[Magno Borgo]

Try Shotrunner.com , not free but not as expensive as Shotugun.Hi all,I remember a while back there was a discussion here about production tracking software. A lot of people use Shotgun, and it is indeed a great tool. Unfortunately, I am stepping into two productions that can't even afford that, and I remember somebody in this forum linking to a free open source tool that can be used to track a production.

Anyone remembers what I'm talking about?Thanks all!Ron Ganbaremail: ron...@gmail.comtel: +44 (0)7968 007 309 [UK]

     +972 (0)54 255 9765 [Israel]url: http://ronganbar.wordpress.com/

-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] SplineWarp python question

[Magno Borgo]

Amazing, many thanks. I knew it was something simple that I was missing.



On 04/01/2012 15:27, Magno Borgo wrote:

Like this:
***

import nuke.rotopaint as rp
SplineWarpNode = nuke.createNode('SplineWarp3')
curvesKnob = SplineWarpNode['curves']
shape = rp.Shape(curvesKnob)

^
\- this is the problem



***
shape gives: TypeError: argument 1 must be (unspecified), not
_splinewarp.SplineKnob


Unfortunately the error message is most unhelpful here though.


I know that rotopaint node uses _rotopaint.RotoKnob and when i try
shape = rp.Shape(curvesKnob)on a roto node it works.

So which method should i use to create shapes inside a SplineKnob?


You are mixing the rotopaint and splinewarp Python bindings. While they  
reuse code, they have different runtime objects.
To make it work, import the splinewarp module instead of rotopaint, like  
so:

   import nuke.splinewarp as sw
then you can do:
   shape = sw.Shape(curvesKnob)

And for the rest of the code keep in mind that you shouldn't mix the  
splinewarp and rotopaint modules.



HTH,
 Wouter





--
******
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] SplineWarp python question

[Magno Borgo]

Like this:
***

import nuke.rotopaint as rp
SplineWarpNode = nuke.createNode('SplineWarp3')
curvesKnob = SplineWarpNode['curves']
shape = rp.Shape(curvesKnob)

***
shape gives: TypeError: argument 1 must be (unspecified), not  
_splinewarp.SplineKnob



I know that rotopaint node uses _rotopaint.RotoKnob and when i try
shape = rp.Shape(curvesKnob)on a roto node it works.

So which method should i use to create shapes inside a SplineKnob?

Thanks in advance,
Magno







Hi Magno,

could you send some of the Python you've been trying?

Thanks,
 Wouter

On 03/01/2012 16:45, Magno Borgo wrote:

Another question on the same subject. Is it possible to create Shapes
inside spline warp nodes via Python scripting? Because i'm trying to
apply some of the rotopaint python exemples on the node and its not  
working.

What I'm missing?


Magno.

Hi all,

The way source and destination curves work will change for the next
major version of Nuke, but in 6.3 the source and destination are
essentially the same curve, similar to Roto and it's feather curve.
(Hence the awkward naming..)

Regards,
Wouter



-- Sent from my HP Pre3

--------
On 15 Dec 2011 01:28, Magno Borgo  wrote:

Yeah the naming was not helping! Thanks Ivan, that was exactly what
i needed.

Magno.


Each control point has both the source and destination
attributes. The confusing part is that the naming of those
attributes makes more sense for a standard roto shape.

So, your source curve is:
controlPoint.center

and the dest curve is:
controlPoint.featherCenter

Also, if you're getting the position data out of each attribute,
keep in mind that the dest points are stored as an offset
relative to the src point, instead of an absolute position.

Have a look at the output of this:

node = nuke.selectedNode()

curves = node['curves']

sourcecurve = curves.toElement('Bezier1')

for p in sourcecurve:

print p.center.getPosition(nuke.frame()),
p.featherCenter.getPosition(nuke.frame())



Hope that helps.
Cheers,
Ivan




On Wed, Dec 14, 2011 at 4:21 PM, Magno Borgo mailto:mag...@pop.com.br>> wrote:

Hello!

I trying to figure out how to access the *control points *of
the***destination* curves of the SplineWarp node via python.

The points of the source curves are easy:

node = nuke.selectedNode()

curves = node['curves']

sourcecurve = curves.toElement('Ellipse1')


With sourcecurve[0], sourcecurve[1], etc i can access each
control point.



    Any help?


--
**
Magno Borgo

www.borgo.tv <http://www.borgo.tv>
www.boundaryvfx.com <http://www.boundaryvfx.com>

___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk
<mailto:Nuke-users@support.thefoundry.co.uk>,
http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users





--
******
Magno Borgo

www.borgo.tv
www.boundaryvfx.com




--
**
Magno Borgo

www.borgo.tv
www.boundaryvfx.com


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users






--
**
Magno Borgo

www.borgo.tv
www.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] SplineWarp python question

[Magno Borgo]

Another question on the same subject. Is it possible to create Shapes inside spline warp nodes via Python scripting? Because i'm trying to apply some of the rotopaint python exemples on the node and its not working.What I'm missing?Magno.Hi all,The way source and destination curves work will change for the next major version of Nuke, but in 6.3 the source and destination are essentially the same curve, similar to Roto and it's feather curve. (Hence the awkward naming..)Regards,     Wouter-- Sent from my HP Pre3On 15 Dec 2011 01:28, Magno Borgo  wrote: Yeah the naming was not helping! Thanks Ivan, that was exactly what i needed. Magno.Each control point has both the source and destination attributes. The confusing part is that the naming of those attributes makes more sense for a standard roto shape.So, your source curve is:controlPoint.center
and the dest curve is:controlPoint.featherCenterAlso, if you're getting the position data out of each attribute, keep in mind that the dest points are stored as an offset relative to the src point, instead of an absolute position.
Have a look at the output of this:




node = nuke.selectedNode()

curves = node['curves']
sourcecurve = curves.toElement('Bezier1')

for p in sourcecurve:
print p.center.getPosition(nuke.frame()), p.featherCenter.getPosition(nuke.frame())


Hope that helps.Cheers,IvanOn Wed, Dec 14, 2011 at 4:21 PM, Magno Borgo <mag...@pop.com.br> wrote:
Hello!I trying to figure out how to access the control points of the destination curves of the SplineWarp node via python.
The points of the source curves are easy:
node = nuke.selectedNode()curves = node['curves']

sourcecurve = curves.toElement('Ellipse1')
With sourcecurve[0], sourcecurve[1], etc i can access each control point.
Any help?
-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Separate Silhouette layers into separate Roto nodes

[Magno Borgo]

You can also now create custom "exporters" inside Silhouette with Python, there's an example on Silhouette Exemples/Actions folder that is half way what you need, you just need to adjust it to export individual nuke files from each of the Layers.Magno.as of silhouette 4.3, you can export the layers from silhouette into individual nodes like you are saying.I know you were looking for a Nuke side answer, but if you still have access to the silhouette source files, then there ya go!

otherwise there's nothing that exists (that I've come across) that will do the split on the Nuke side, but it's usually easy enough to cut & paste.  I know if you have billion layers it could conceivably get pretty hairy.

On Mon, Jan 2, 2012 at 12:32 AM, Jud Pratt  wrote:

Hi everyone!

Does anyone know if there is a script out in the wild that will let me import roto shapes out of Silhouette and then separate out the individual layers into their own roto nodes?

In other words... take what imports as a single roto node such as:

Layer A
        roto
        roto
        roto
Layer B
        roto
        roto
        roto
Layer C
        roto
        roto
        roto

and separate those out into 3 individual roto nodes instead of just one roto node with all 3 layers.

I'd like to be able to import the roto that was done without having to separate the pieces out by hand as there are sometimes a lot of layers to contend with.

If there's not a script already... a pointer in the right direction to help me roll my own would certainly be appreciated!

Thanks!
--Jud Pratt
j...@judpratt.com


___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/


http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users


-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] Import boujou target tracks

[Magno Borgo]

I don't use Boujou, but http://tracksperanto.guerilla-di.org/ can do the conversion for you if it doesn't import directly into Nuke.
Hi,

If we export the target tracks from Boujou 5.1 as a text file is it possible to import these to nuke as 2D tracks?

Thanks,

AH
-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Re: [Nuke-users] SplineWarp python question

[Magno Borgo]

Yeah the naming was not helping! Thanks Ivan, that was exactly what i needed. Magno.Each control point has both the source and destination attributes. The confusing part is that the naming of those attributes makes more sense for a standard roto shape.So, your source curve is:controlPoint.center
and the dest curve is:controlPoint.featherCenterAlso, if you're getting the position data out of each attribute, keep in mind that the dest points are stored as an offset relative to the src point, instead of an absolute position.
Have a look at the output of this:




node = nuke.selectedNode()

curves = node['curves']
sourcecurve = curves.toElement('Bezier1')

for p in sourcecurve:
print p.center.getPosition(nuke.frame()), p.featherCenter.getPosition(nuke.frame())


Hope that helps.Cheers,IvanOn Wed, Dec 14, 2011 at 4:21 PM, Magno Borgo <mag...@pop.com.br> wrote:
Hello!I trying to figure out how to access the control points of the destination curves of the SplineWarp node via python.
The points of the source curves are easy:
node = nuke.selectedNode()curves = node['curves']

sourcecurve = curves.toElement('Ellipse1')
With sourcecurve[0], sourcecurve[1], etc i can access each control point.
Any help?
-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com
___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

[Nuke-users] SplineWarp python question

[Magno Borgo]

Hello!I trying to figure out how to access the control points of the destination curves of the SplineWarp node via python.The points of the source curves are easy:
node = nuke.selectedNode()curves = node['curves']

sourcecurve = curves.toElement('Ellipse1')
With sourcecurve[0], sourcecurve[1], etc i can access each control point.Any help?-- **Magno Borgowww.borgo.tvwww.boundaryvfx.com___
Nuke-users mailing list
Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users