Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Black Shading Tests

Just looking at the manual under black calibration and noticed this:

2. Set the exposure to the exposure you want to use for your shoot.

Hmmm, shouldn't that specify shutter angle instead or am I just a crazy person? If the manual is correct, wouldn't that mean re-calibrating dozens of times during a shooting day?
 
Using the wrong black shade is not a solution, its just achieving an effect you desire without understanding why. Not a smart person's method ;)
Graeme lays it out perfectly :)
 
Graeme, are these crushed blacks reversible? Ie fixable in RAW post?

Or are they permanently captured more "crushed" than if recorded with proper Blackshade?

With an extremely wrong black shade it is probably possible to permanently kill some data. I've not seen this happen, but that doesn't mean it couldn't. In my example above where black got put at 90 rather than 100, it would have to be put near 0 to start killing actual data.

Because the black shade is applied while recording, what you've done is alter the data point at where actual black is stored in the R3D, and that could potentially be undone, but as I note, that's not how the system is designed to work.

Graeme
 
D.E.B. is coming. Also, I mean this is really more of a personal note, shooting at 4K, 5K, 5.5K, and 6K for a 4K finish when projected or on a UHD 4K screen this whole "red sparkies" phenomenon was blown WAY out of proportion.

I'm actually blown away by what D.E.B. actually does. And while I enjoy reading the chaos of "will it happen in camera" or "now we have to process our footage in Redcine-X Pro". Just like a color science update, which this technically is, it will get released in the SDK and we'll see what Red can do about getting it deployed in camera. For 1080p HD-SDI or HDMI recorders this really shouldn't be an issue. I can see it "might be" something to look into when the 4K Broadcast Module lands later this year as some will be shooting 4K for 4K and I have no idea what the red specks look like in the uncompressed stream.

I'm astounded lately by the pixel peeping. Other cameras have some comical things that are never brought up and I'm not going to bash any manufacturer, but there's such hyper focus and insanity sometimes I'm wondering what people are doing exactly.

I'm also astounded that Red found a way to solve that red speck thing and did something that makes the footage overall better, but you know. I can see what's actually good about that.

Phil, I added an edit apparently after you had quoted me in which I said

"For the record, I'm down with DEB. But I know there are some people on here who need something clean out of camera and if bumping up to 1/100 th gives them a feed they can use until something better comes along, where's the harm?"


So yes, I'm in complete agreement with you on DEB. That will suit my workflow just fine. It's just that I empathize with those who make their living in a way that requires something akin to a live feed.
 
Its worth pointing out also just how wrong a blackshade can be - a 1 second blackshade on 1/48th footage for example creates a dramatically lifted exposure. Going that direction the result is clear to see - going the other way is less apparent, as burying the black is harder to quantify, but the effect is just as real.
 
Phil, I added an edit apparently after you had quoted me in which I said

"For the record, I'm down with DEB. But I know there are some people on here who need something clean out of camera and if bumping up to 1/100 th gives them a feed they can use until something better comes along, where's the harm?"


So yes, I'm in complete agreement with you on DEB. That will suit my workflow just fine. It's just that I empathize with those who make their living in a way that requires something akin to a live feed.

I think the key points here are:

1. The heavily oversampled 1080 HDSDI output is pretty damn clean - the red noise barely rears its head in the first place.
2. Mis-adjusting with black shade only does the same thing as intelligently using shadow, contrast and curves in-camera, but potentially more destructively to recorded R3Ds and with less finesse.
 
I think the key points here are:

1. The heavily oversampled 1080 HDSDI output is pretty damn clean - the red noise barely rears its head in the first place.
2. Mis-adjusting with black shade only does the same thing as intelligently using shadow, contrast and curves in-camera, but potentially more destructively to recorded R3Ds and with less finesse.

Thanks John, hope those who this affects are reading this.
 
With an extremely wrong black shade it is probably possible to permanently kill some data. I've not seen this happen, but that doesn't mean it couldn't. In my example above where black got put at 90 rather than 100, it would have to be put near 0 to start killing actual data.

Because the black shade is applied while recording, what you've done is alter the data point at where actual black is stored in the R3D, and that could potentially be undone, but as I note, that's not how the system is designed to work.

Graeme

Great. Thanks for succinct answer. Sounds like proper blackshading is best way to move forward.

thx again!

best
 
DEB and full debayer will be your friend for any red speckle noise that might be leftover AFTER you grade.

When it's implemented in resolve for example and you are grading from r3d you might only DEB pushed shots where there is leftover noise after grading. I don't think it's going to have to be an "always on" type of thing, just like you would not "neat video" an entire timeline (unless you are noise)

For the hdsdi out so much information is being tossed to get to that 1080, that I suspect it will be pretty clean there too, or as Graeme says you can always add a bit of curve or contrast to get the blacks down a bit.

But we've already seen that great ISO 400 limbo example straight out of the camera for all the speculators out there. :)


battistella
 
i test again and must say i stay with Black Shading 1/50 = Shooting 1/50

at my Dragon, it´s better than the 1/200 Shading (with 1/50 shooting)
 
Its worth pointing out also just how wrong a blackshade can be - a 1 second blackshade on 1/48th footage for example creates a dramatically lifted exposure. Going that direction the result is clear to see - going the other way is less apparent, as burying the black is harder to quantify, but the effect is just as real.

nice BS in 1fps :) I love this look creamy :)
I often use it for video clips, but the best is still to Dominik: (hehe

https://vimeo.com/62930319

It is certain John, to be definitely wrong with a BS shutter speed different..
Each of us knows.
But visually I like best ', but I do not think being a person "less intelligent" because they do not use methods such as textbook. I just did my tests and I set the camera as I like.
I believe that everyone can achieve the same results in a different way.
 
Fair enough Luigi - I think when you arrive at a decision like that with careful testing and know how to repeat it, you can exempt yourself from my 'not smart' statement :) :)

I'm more concerned that a rather black-art concept might be aped by users better served using more straightforward controls designed for the purpose.
 
Fair enough Luigi - I think when you arrive at a decision like that with careful testing and know how to repeat it, you can exempt yourself from my 'not smart' statement :) :)

I'm more concerned that a rather black-art concept might be aped by users better served using more straightforward controls designed for the purpose.

John certainly, the basic concept is very important :)
 
DEB = Dragon Enhanced Blacks

Not released yet but part of upcoming RCX (and R3D SDK for other apps) to solve the red noise issue with Dragon new OLPF aka 'OLPFv2'

recon thread on new FW / DEB in RCX
 
Back
Top