Applying a Glow to Object ID

[Corbae]

RikiB,

I'm kinda new to Fusion myself, but I think placing the glow right after the loader of your object should do the trick.

[Jayk2k]

Apply your glow node, and then open the fusion tab (radioactive tab, common controls). Then check the "use object" or "use material" radio button, and it will open 2 sliders. Just set the appropriate slider (o or m) to the desired channel. If you rendered rpf from max, the "correct edges" can help by using the pixel coverage channel.

[PolyMangler]

ok so DF is like the best, but Im kinda coming off of Combustion and there is this 1 feature I loved in Combustion but doesnt seem to exist in DF. How do I apply a glow on the object ID level or a render node. For example I just want this 1 thing in my scene to glow, not the whole scene! I know this must be possible with digital fusion so please I beg you to enlighten me!

Thanks! you guys ROCK!

Those methods will work but the problem with them is that they will limit the glow to the borders of the object.

The method i use is put a glow right after your loader.

1) select the glow
2) in your image viewer right-click and move down to glow mask--> Bitmap , this will put a
mask on your glow
3) Click on the mask tab of your glow tool, you'll see a mask called Bitmap, in the panel you will see a title called drop tool here, click and drag your loader w/ the obj info on it and drop it on that bar.

4) next check on use object and/or use matiral and use the pick to pick your surface (make sure your viewing the loader)

5) viola you are done the glow will now be allowed outside the border.

p.s. If you're compositing 3d elements alot in DF it's a good rule of thumb to render double-size with NO aliasing and the resize the loader by half, this helps smooth your z-buffer and object buffers....plus renders faster and from what we've found looks better

[vonbon]

:arteest: im working on a lil project in DFX+ that i want to add 3d elements to overtop video. Using a Sony Cam(720x480, 1:2 aspect) widescreen option on the camera. 1 person told me that LightWave will create fields automatically, but I also saw on DF's site where they were saying somthing bout (Though most 3D software allows you to render fielded frames, many animators prefer to produce a sequence of half height images - one for each field. This is often produces faster renders, but the images must be interlaced into a fuill frame before they can be output to tape).

and now you confused me more by saying to render Double Size with no AA

Does that stuff matter only if your going back to tape? I plan to finalize it digitally, DVD or VCD.
If I do it the way "Eyeonline" is saying will it automatically stretch them to show on screen correctly. just want to know the "EasiestBestmostFlexible" way to composite full frame 3D elements.

[PolyMangler]

Though most 3D software allows you to render fielded frames, many animators prefer to produce a sequence of half height images - one for each field. This is often produces faster renders, but the images must be interlaced into a fuill frame before they can be output to tape).[/u][/i] .


The whole rendering at half and stretching thing I'm not familier w/ we output everything to quicktimes and give it to the avid guys at full frame.

But the part about rendering double size I can clarify. The issue starts of w/ the fact that LW nore Maya output a proper coverage buffer (DF explains in depth what it does). In short the coverage buffer tells DF how the seperate 3d elemants were aliased together in the render so that when you composite different elements in front of or behind them they look nice.

When the coverage buffer is absent you have a harsh, jagged line that shows up when one composited 3d object over the other, Also if you use something like 3d fog or depthBlur it will show up around the edges of all your objects. To test this render out a ntsc frame of your 3d object w/ a z-buffer and add a fog or depth blur.

To work around this problem we started rendering all of our sequences 2x(720x486) with no aliasing. The reason for no aliasing is that when you resize in dfx it will alias the image for you and save you buukoo render times. What also happens is that z-buffer also gets aliased along w/ the rest of the image smoothing out that harsh jagged line that i mentioned before.

As a bonus, and this is a maya thing, we found that along w/ all of z-buffer stuff we found that our textures looked sooo much better letting DF do the aliasing.

I'm sure this method has it's downfalls and it's not a perfect fix for the coverage buffer problem but it's loads better then just living w/ the harsh lines that pop up.

I hope this helped clear things up and if not just post back,

Cheers,
Bruce

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.