Jussi Rovanperä

I've been using Fusion to study how color transformations work, that's a good tool for a non-programmer like me.

The primaries for RGB color gamuts are XYZ coordinates.

The matrix transform from one set to another is done in XYZ color space, so those transform matrixes cannot be used in RGB.

I'm rarely using the rgb mixer, except for some channel repair stuff and B&W conversions. I understand how the RGB mixer controls transform the rgb color space, but it is still very unintuitive way to do things, so I prefer the more direct way of using the primary controls and the different curves.    

I'm sure there are some RGB mixer ninjas lurking here, so please chime in!

If you put 2 corrections parallel in Resolve, and change node sizing of the corrections, you can see the parallel operation break

I'm not aware of other software that have parallel operations like Resolve, I think that mixing of corrections is a unique feature.

But with many layer-based grading apps, keyers can key from the previous layer, source or a designated layer, and cc layers can also grab the original source again on top of graded layers, so you can do similar routings that you could do with the layer mixer and nodes.

Resolve seems to be using optimized media instead of the original, so another thing to take into account.

Fusion Connect works, but there are caveats:

no handles

transcoded clip has timeline resolution

transcoded clip has grade applied (unless the grade is turned off)

trancoded clip has 00:00:00:00 timecode

So at the moment the main benefit of Fusion Connect is really the versioning, or quick fixes.

   DNXHR MXF's work well.

Sorry for stating misinformation, actually you can bring a background plate as matte clip and layer with the original in the node editor, so there's no necessarily need for 2 video tracks. Is this new on Resolve 14, or have I missed this on 12.5?

And sorry for hijacking a Baselight thread, I just wanted to correct what I wrote earlier...

Now you need to stack video tracks with alphas, instead of bringing the clips to the node tree and comping there.

It's funny that right now Resolve has node based color, but layer based comp...

Yes, the input size is copied with a still, if "copy grade: preserve input sizing" is disabled.

There's edit sizing that translates from edit to color and back. To me that's a more logical way to resize than input sizing. Maybe they kept input sizing for backwards compability.

Maybe look at some VJ apps, I'm sure there are some simple ones out there that will do what you want.

Scratch Play can playback the PIZ / DWAA and DWAB files @25fps

EDIT: Sorry, that was a bit too optimistic, Scratch Play seems to cache single clip playback.

With Scratch I get realtime (25fps) for PIZ (cpu at 80%) , 22fps for DWAA and 20fps for DWAB (cpu at 100%).

With Resolve the cpu doesn't go to 100%, but stays at 80%.

Resolve 14 b7:

PIZ 19.5fps

DWAA 45  15fps

DWAB 45 13.5fps

I tried different compression options between 0-45, but the files Resolve wrote are all similar in size, so something is wrong I guess...

10 hours ago, Thomas Singh said:

Similar frame rates suprises me. It could mean that it is a bottle neck somewhere that restricts the full potential. It could also mean that the DWAA/DWAB compression algorithm is not well enough integrated in Resolve. In Nuke the difference is huge. 

Thanks Thomas, that's good to know. I gotta check with the Resolve 14 beta, if there is a difference to 12.5.

1 hour ago, Emily Haine said:

Hi Jussi. DWAB is more efficient than DWAA because the algorithm is optimized for increased performance in most of the common post production tools. You should be able to notice the difference.

I get similar frame rate for both in Resolve. Piz compression gives me similar frame rate as well.

Hi, has anyone used DWAA/DWAB compressed EXR intead of uncompressed frames or prores/dnxhr as an intermediate codec? Could be a good solution for a frame-based workflow with UHD/4K material.

However With UHD footage and timeline I get 17 fps playback in Resolve 12.5.

With my current computer (4ghz quad i7) the processor usage is 80%ish, GPU is at 40%, Drive speed not a problem, 100MB/s off an SSD. I wonder what is the bottleneck with playback, probably the CPU?

Typically the debayering is done by the SDK provided by the camera manufacturer, so it is kind of "black box" process, and the raw settings are identical in different apps.

If you're interested in different demosaicing methods, you could check the free RawTherapee app for still photos, as it has about 10 different demosaicing methods, and a ton of other esoteric processing options as well.

Resolve has now almost 50 ResolveFX nodes that can be dropped in the node tree without using color nodes, so I think that might be a hint where things are going with the Fusion integration.

1 hour ago, Nicolas Hanson said:

I have a complicated shot where I would like to reduce a natural bright light from a window behind the subjects back. When the subject move closer to the camera (away from the window), the 'halo' is reduced and it's easier to get a cleaner separation. I have tried the curves in a number of combinations and the soft clip controls without getting to any satisfiying result, but you might have some other inputs?

You're in Resolve? Are you using the "Highlights"?

12 hours ago, Shane Taylor said:

Can you offer any insight as to the brightness ranges I landed on in the final.

I think the contrast is fine, but the white balance is magentaish, like Abby stated. 

I'd pull down the yellow-green saturation first, as that is the dominant color now and not really wanted in the final image, and then cool the white balance and adjust the exposure/contrast. I'd avoid desaturating the whole image.  

I remember that Steve Scott mentioned in fxguide podcast that they use a customized version of Lustre. But that was couple of years ago.