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 :)

DSMC v5.2.28 Release Build

Anyone noticing a cleaner 2000ISO? This is pretty clean giving that its all natural light and the worst light source you can use.
2000isonodebnoadd.jpg
 
Anyone noticing a cleaner 2000ISO? This is pretty clean giving that its all natural light and the worst light source you can use.
View attachment 56049

The new firmware uses RG4 as default which hides noise due to it's more aggressive curve on the low end.

Develop old firmware image and new firmware footage with RLF to gauge noise.
 
The new firmware uses RG4 as default which hides noise due to it's more aggressive curve on the low end.

Develop old firmware image and new firmware footage with RLF to gauge noise.

Sounds like good advice.

OTOH, people don't usually really complain about noise (all cameras have that) but the degree of visibility... :)
 
Not sure if it's just me but I'm getting what seems to be a cleaner image/lower noise floor on my Scarlet X with the latest firmware. I haven't done a black shade yet and in RedGamma 4, and it looks great out of the gate.

Anyone else noticing the same?
 
Not sure if it's just me but I'm getting what seems to be a cleaner image/lower noise floor on my Scarlet X with the latest firmware. I haven't done a black shade yet and in RedGamma 4, and it looks great out of the gate.

Anyone else noticing the same?

Here's your answer:

The new firmware uses RG4 as default which hides noise due to it's more aggressive curve on the low end.

Develop old firmware image and new firmware footage with RLF to gauge noise.

RLF = Red Log Film
 
The new firmware uses RG4 as default which hides noise due to it's more aggressive curve on the low end.

Develop old firmware image and new firmware footage with RLF to gauge noise.

True RG4 does crush it more then RG3 and REDlogfilm. But If RG4 been crushing the red noise this good in the blacks before in the previous builds, then less people would of complained about the noise in the red channel. I had the past few beta firmwares on my dragon that had RG4 and it still couldn't give me clean blacks like this release just did in RG4 at 2000iso. I usually shoot REDlogfilm with one of my custom curves, but wanted to test the new firmware at it's defaults and it looks better. I still get a red image when camera is warming up to temp, so it's not gone, but it's reduced when it comes time to shoot.

I did look REDlogfilm before I posted it, there's a lot less red noise in the image then the previous builds. There's the obvious noise in the blacks, but that's normal for any log at high ISO's. It looks a lot more of a even rgb noise then it did before where it was a very strong red noise in the image. I'm happy, I usually don't shoot past 800 so this can get fun to see what's possible with the camera.
 
I suggest leaving D.E.B. where it belongs, which is in post. When shooting very red environments (such as sapele hardwood floors that read much more red than they do brown), the chatoyance (http://en.wikipedia.org/wiki/Chatoyancy) of the wood fools the camera into thinking its seeing a lot of red noise, when in fact it's a property of the material. D.E.B. softens that up way too much. D.E.B. should be an option when it's needed, and the DRAGON sensor should be aloud to speak for itself when called for.

sapele_title.jpg
 
Michael, do you have proof it does what you say or are you just theorizing? If you have proof, please send to Graeme to investigate.

Here's a one second R3D trim of 6K FF video: https://www.dropbox.com/sh/38i9jgej0o7b6w3/AABrw5dACP3Prt_bwaEUPDx7a?dl=0

Look at the structure and the contrast of the floorboards, and notice that when you turn on D.E.B., the contrast goes down and the softness comes up. Make no mistake--there's plenty of noise in these dark floors. But the organic noise makes the floor look deeply textured with nice, contrasty wood grain. D.E.B. tames the noise, but at the expense of showing the contrast and texture of the wood. At least that's what I see. And that's why I'm happy that D.E.B. is a feature in post, not baked into the firmware.
 
I suggest leaving D.E.B. where it belongs, which is in post. When shooting very red environments (such as sapele hardwood floors that read much more red than they do brown), the chatoyance (http://en.wikipedia.org/wiki/Chatoyancy) of the wood fools the camera into thinking its seeing a lot of red noise, when in fact it's a property of the material. D.E.B. softens that up way too much. D.E.B. should be an option when it's needed, and the DRAGON sensor should be aloud to speak for itself when called for.

sapele_title.jpg

Hi Michael,

I read, with interest your observations, regarding the red/brown variation and the chatoyance you describe. I have found that there is quite a variance in the hues and, what I fancifully like to call the Trompe L'oleil of the "waterfall" that the African Bubinga pollelle veneers offer. With certain, select veneers, I find that the hues change dramatically, as the angle of incident light refracts throughout the "waterfall," etc.. This does" trick the eye," in terms of hue and pattern (movement) variance to a very large degree, camera sensor, etc. aside. Your point, however, as regards the DEB effect, using your example is well taken. There must be an update afoot, though, as even the photo of a RED stoplight, +/- DEB is quite revealing...

Thank you very much

Fury
 
Cory,

Are you seeing noise in the 1920x1080 Prores files you are delivering? Can you post some samples?

Battistella
 
http://www.reduser.net/forum/showth...-5-2-21-beta&p=1393257&viewfull=1#post1393257

Still pushing for DEB to be an in camera option for those of us needing quick Prores delivery using external recorders.

If you are recording with prores external recorder, it's important to expose correctly.
If the left goal post is filled up, crush the blacks, and hopefully you won't get noise that way.
Also, RG4 and RC4 on camera would be good.

Edit: you can also crush the blacks in camera, by going to Looks, and lowering the Shadow.
 
Camera Clock Resetting

Camera Clock Resetting

Ever since upgrading to this firmware, the camera's internal clock is resetting to 0 hours, Jan 1, 2010 after each power cycle. Scarlet X. This is with external AC power. This is new behavior happening since the upgrade from 3.x.
 
Back
Top