Am I Rendering the RED file correctly?

[Francis]

Please excuse my noobness everyone, I'm still very new.

I did some filming over the past 3 months and have been unsure if I'm going about the process correctly.

The initial footage is kind of grainy/noisy in the dark regions without any adjustment.

ISO320, K4500-5000, T2.8? (maximum aperture) on REDone@24.

What I did in attempt to fix that is to push contrast and brightness in red alert to move the 3 colored curves to the right while keeping their peaks below the top border. Is that right?

I'm also not sure if I should be using AppleProRes or Avid or AJA Kona codecs for better quality.

And if anyone knows a quick explanation of DPX; all I know is that it processes each frame into an individual file or something, and that programs and just load the entire series for editing.

Thanks!

[Dan Hudgins]

If you want to grade your footage you can output 4K TIF and use my "freeish" programs to de-focus the blue and run an adaptive sharp soft filter on the footage, then down sample using my good downsample that blends 100% of the source pixel data.

Depending on the end use size, not using the sharpen in REDCINE can produce a better result, if you are going to DVD from 4K you need to blur the image rather than sharpen it, if you sharpen then use a resize that skips pixels you get MUCH more noise in the image.

To export for use in my programs set all controls in REDCINE at neutral and export as 48bpp 16bit TIF using REDLOG rather than Linear.

Another "free" program called Irfanview can be used to batch process and has some image controls and a good resample. Xnview may also be of use.

[Francis]

I'm sorry Dan, but you totally lost me with your programs. When you talk about not sharpening, are you referring to me boosting contrast and brightness?

[Dan Hudgins]

I'm sorry Dan, but you totally lost me with your programs. When you talk about not sharpening, are you referring to me boosting contrast and brightness?

There are settings in the RED programs for OLPF comp. and sharpen, those can be adjusted.

How sharp you make the images depends on the end use, the softer the images in both resolution and contrast the less the noise and compression artifacts show up to the viewer.

Adaptive sharpen and soften lets you sharpen edges and soften areas of low detail to reduce the noise and compression/wavelet artifacts that show up in areas of low detail more.

If you are going to a size less than 4K how the image is re-sized affects the noise in the result, there are two ways to resize,

1) Blur then resize using a method that skips pixels
2) resample using a method that uses all the source pixels

Many programs use method 1 since it can be faster, but is can also show more noise and aliasing.

So the noise you see on the screen can be caused by the way the image is resized to fit on the screen, or resized to fit into the film recorder raster.

You cannot get the "same" results for noise levels if you look at a 4K image on a 4K projector or look at a SD downsize on an SD projector since the smaller image will have the average of many pixels if the downsize is done right, from a given distance the SD image may look better, it all depends on what methods are used and how far you are away from the projected images (angle of view in your visual field).

My programs can color correct at true RGB 4K resolution, so may not have some issues that compressed editing systems have with regard to image artifacts, or you may see more noise because compressed systems blur parts of the image, and how you view the images on the screen varies also.

You cannot judge what a movie will look like on a 35mm print from what you see on a monitor since the film printing blurs the compression artifacts a little as well as the noise. Resampling to 1280 wide may get you close to what you will see if you view from 2x to 3x the screen width.

[Francis]

Thanks a bunch Dan. You cleared up a bunch for me.