Is R3D 4:4:4 or 4:2:2?

[Adam Watson]

The title says it all. Maybe there is a concept of it that I don't understand but I can't seem to find any indication of whether Red RAW is 4:4:4 in nature or 4:2:2. Anyone know for sure?

I know you can output 4:4:4 in transcode but is that 4:4:4 level of color information actually there to begin with?

[Graeme Nattress]

Neither - it's RAW. 4:2:2 etc. refers to the situation where chroma bandwidth is reduced in a Y'CbCr encoded signal to achieve a smaller bandwidth.

Graeme

[Adam Watson]

so it can be transcoded into whatever, got it.

[Martin Stevens]

Are R3Ds 16 bit or 12 bit?

Is there a way to change the recording bit depth on the Epic?

Is the Epic-X a 16 bit system?

[Yaque Silva-Doyle]

Epic R3ds are 16 bit

[Bruce Allen]

The title says it all. Maybe there is a concept of it that I don't understand but I can't seem to find any indication of whether Red RAW is 4:4:4 in nature or 4:2:2. Anyone know for sure?

I know you can output 4:4:4 in transcode but is that 4:4:4 level of color information actually there to begin with?

First off, if you're talking 1080p... sure RED has 4:4:4 level of detail.

But at 4K or 5K:

The Bayer sensor reconstruction can get you more than 4:2:2 quality in some cases... but for all practical purposes, no: you have either a red or a green or a blue pixel, not a RGB pixel. It's a real stretch to see how you can possibly get high-quality 4:4:4 info out of this sucker.

Personally I have been doing tests and am moving my post pipeline to 4K... but at 4:2:2.

I really haven't seen the RED getting enough chroma detail beyond 4:2:2 for it to be worth the massive increase in disk space that 4:4:4 requires.

If you compare a 1.8GB folder of 4K DPX files... you can do OpenEXR at 4:2:2 with light B44 compression and get practically identical quality but at 577MB.

Unless you have a lot of stock investments in hard drive companies, 4:4:4 4K just doesn't make sense to me.

Bruce Allen
www.boacinema.com

[Graeme Nattress]

Bruce, please don't use imprecise terminology. 4:2:2 etc. don't apply to a measure of chroma resolution in any sense, but a process by which resolution is deliberately reduced by some circuitry or process for bandwidth reduction reasons.

For instance, you could record a VHS tape onto Digibeta at 4:2:2. At which point you definitely have a 4:2:2 signal, but that actually tells you nothing about the measured chroma resolution of the signal. Similarly a BetaSP dub to the same Digibeta would be 4:2:2 but the actual measured chroma resolution would be vastly superior. A direct camera recording to that Digibeta would also be 4:2:2 and depending on the camera could have a vastly superior measured chroma resolution.

The reason, of course, why such analogue era compression as 4:2:2 works so well is that we don't see much detail in chroma compared to luma.

The issue of how colour filter array sensors see colour resolution compared to luma resolution is a different issue. In such a system we're dealing with broad colour filters where any detail is visible in all three channels in most circumstances.

Graeme

[Adam Watson]

Okay, this makes sense. Thanks for the response.

[Craig Parkes]

Graeme, well your assessment about imprecise technical terminology is useful, I think Bruce's point is a very apt one when working in workflows above HD, which is this:

At what point does maximum chroma resolution at a 4K source become a waste of bandwidth for a given application.

By offering RAW Red gives us the choice about this, unlike in the video world where traditional workflows have had video engineers basically make this call for us by the limitations/expectations of industry standard hardware. What they said was 'good enough' became the status quo, or the ideal, when in reality it's all a compromise.

When we are actually working with files in a RAW originated - all digital workflow, we each individually become a lot more enabled, but also responsible, as we can shoot ourselves in the foot by not using suitable compression.

If at 4K there is generally a negligible difference delivering 4:2:2 Y'Cb'Cr compressed footage from a compressed Bayer pattern sensor that shoots a RAW master at 5-4k versus delivering 4:4:4 Y'Cb'Cr footage from the same, then that's something that is useful to know when working with 4K files.

As you state we don't see much detail in chroma compared to luma and as you say 4:2:2 compresssion works well.

Do you think, given the increase of digital workflows we can expect to work with increasingly either RGB compressed signals or we are likely to continue to use Y'Cb'Cr formats for the forseeable future, not just for legacy reasons but also for the compression efficiency they provide.

To put it in a form similar to the original question: What are the advantages to moving R3D RAW footage into a 4:4:4 Y'Cb'Cr format at a given resolution, over moving to a 4:2:2 Y'Cb'Cr format at a given resolution.

Also a question for Bruce - if you are working in a 4K colour space with a 4:2:2 resolution, within your workflow is there a difference if you downrez that to 1920x1080 in a 4:4:4 Y'Cb'Cr supported format over downrez to a 4:2:2 Y'Cb'Cr format, or do you find the tools you have throw the additional chroma resolution that could be obtained from the higher resolution source footage away so once in 4:2:2 format you are effectively stuck there (without further intermediary transforms, such as to full bit depth RGB file format, and then going back to 4:4:4 Y'Cb'Cr?.

Also I wonder if somewhat counter intuitively, you could effectively result in more chroma detail artifacting if the down conversion to 4:4:4 from a 4:2:2 source over going to a 4:2:2 if the down conversion isn't a nice round number - i.e if it's the down conversion maths is messy, would you sometimes be better off with slightly less detailed chroma regions to 'blur' artifacting from an imprecise downscale?

I presume high end compositing software would get this sort of thing right (internally moving everything into RGB and then back out to your new format at the maximum bit depth possible), but I'd be curious if After Effects or general purpose NLE's/finishing software gets this right

[Graeme Nattress]

Craig, that's another question altogether! It's a question that applies to all footage. What I'd suggest is to keep the quality as high as possible as long as possible, only reducing it significantly for any distribution codec.

In terms of compression the reduction from 4:4:4 to 4:2:2 reduces the pre-compression bandwidth by 1/3, and usually in a very visually transparent way, making good use of the foibles of the human visual system. However, it is my opinion that instead of such a simple decimation, you'd actually get a better image by keeping the chroma resolution full, and using a slightly more lossy compression on the chroma to make-up for that 1/3 head-start that 4:2:2 gets over 4:4:4.

Graeme