fusion additive merge more

What is fusion additive merge mode? It is default, but when I have a colored background, in transparent pixels of the foreground object, where alpha is 0, it adds a color instead of 0 alpha value.

Please help! Stuck with alpha-channel

Why if I render premult alpha background color affects the comp?

Don’t get me wrong, I’m trying to solve this issue for a week and still can’t find the answer. I understand how it adds in "over" operation (alpha of foreground object is inverted, multiplied with background and then foreground object is added).

For example the background color where the alpha equals 0 is blue, and the alpha there is 0. Why it’s not 0 in the end? Shouldn’t it be scaled down to 0? If we mult any color by 0, it’s 0. Why does the opposite happens?

Maybe the reason is fusion additive merge mode? It says it "simply adds pixels, assuming boths images are premultuiplied".

Please help!

Jesse & Joy – “Chocolate” (Music Video and Making Of Video)

Carlos Lopez Estrada, a director living in LA, created this fantastic music video that we at Pro Juice came across recently. There is also a great behind the scenes look at the making of this music video which is a must see. From the VIMEO site…

“During the first month of the year 2010, a group of 9 brave soldiers worked for 30 days straight, without seeing the light of day inside a secret dungeon located in Los Angeles, California. Their task was to gather 90 handmade cookies, sugar, milk, flour, chocolate, eggs and dough; and utilize them to create a 4 minute stop motion video for a foreign artist whose lyrics still remain unknown. They disappeared for 6 weeks and just recently emerged out of the darkness with a small floppy disk containing this: The Official music video for Chocolate!!!”

http://www.projuice.org/musicvideo_chocolate.html

using 3d objects in nuke as holdouts

hi!

maybe this is a stupid question, but i can’t for the life of me figure out how to do it 🙂

let’s say i bring in an object/objects into nuke from a 3d-app and then create additional fx-elements in nuke…think water-splashes on cards for example.
or even more simple…different scanlinerender-outputs in nuke and i want to use the 2 as holdouts for each other.

is there a way to use one as the holdout for the other so that it’s "matted out" basically!?

i tried using zmerge but that doesn’t do the same thing. also i’ve gone through various layermodes in combination with zmerge…but that all lead me nowhere it seems 🙂

thanks in advance for any help!

cheers
chris

Kitteh Kitteh

kittehkitteh
Tokyo Plastic has released the first of a series of short films based around what goes on in the dreams of a cute little kitty. Website series to come at KittehKitteh.com

1. The Tail Gunner, 2. Catzilla

Martin Vallin Photography

martinVallin
Martin Vallin injects energy into his still life photography.

maya cam to nuke problem

Hi, I moved around a camera in maya to match an image sequence for projection. When i export the GEO to nuke it lines up with my maya file, but when i grab teh camera it comes in at 0,0,0 no longer lined up with my geo.

Im sure im not ticking a checkbox or something easy, anyone know what it is?

proper custom viewer lut workflow

Ok, we’ve got a project where we’re working with S-Log plates and Linear 3D files.
I’m trying to sort out the pipeline for Nuke, but i’m having a bit of trouble.
In shake we’ve got it set up like this:
-S-Log to Lin node
-Lin to S-Log node
-Cinespace viewerlut to view the S-Log files, to see what the grade will do to them.

What i’ve got so far in Nuke is:
-S-Log curve for read nodes
-.3dl file from cinespace/cinecube
-expression nodes (with same expressions as the shake Slog-lin nodes)
-vectorfield node set to viewer_input (with the .3dl file, and color space in to S-Log and colorspace out to sRGB)

When I view an S-Log file (with the colorspace set to S-Log) with the vectorfield node, it looks really similar to what i see in my shake viewer with the cinespace lookup (but still the whites are a tiny bit brighter).
But it doesnt make sense to put the vectorfield input to S-Log does it? If i set it to sRGB, the whites get a bit too bright compared to what I see in the Shake viewer.

Ugh…it’s getting a bit confusing. Or is it that it’s never going to look completely look the same as in shake?
Can someone recommend me the correct way to set this up?

Thanks,

Tom

Tokyoplastic: Kitty Kitty

“12 years in the making and pretty much as good as Avatar though considerably shorter.” Tokyoplastic has shunned commercials/ outside world/ friends & family and been eating baked beans on toast for the past six months … resulting in their new Kitty Kitty series.

Nuke Stereo Master Class Plus 6.1 Preview

Quote:

The Foundry held its second Master Class in London on Friday, playing to packed house at the Vue theatre in Leicester Square. The event was standing-room only, having given away all the seats in only 24 hours when it was announced earlier this year.

The event focused on stereoscopic production and post…a timely program after the overwhelming success of Avatar at the box office. The Foundry actually went out and shot some stereoscopic scenes so that the entire process could be covered for their event. Throughout the day, a variety of speakers examined various issues regarding stereoscopic projects. fxguidetv will have coverage from the event in the next several weeks but in the meantime, click through for more about the event and a few details about the the upcoming Nuke 6.1 release.


Continued:
http://www.fxguide.com/qt/2294