Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

Holiday Update + IPP2 Preview Build

rand thompson said:
Graeme,

I knew couldn't be that easy. Well, back to the drawing board. I came here to reduser.net to learn, and I'm learning. Next, learning to make color space conversion matrices ha ha.


Trying to understand all of this.


http://www.ryanjuckett.com/programming/rgb-color-space-conversion/
Click to expand...
Have you read Charles Poynton's book? I highly r commend it.

To make a 3x3 color space conversion matrix you need the x, y for Red, green, blue and white (white point). It's linear algebra. I would also take a look at Bruce Lindbloom's website. Graeme's paper provides the RGBW x, y values need to build the matrix. You will need to roll your own gamut mapping code to deal with out of gamut colors. This is part of what IPP2 does for you.
 
Stacey Spears said:
Have you read Charles Poynton's book? I highly r commend it.

To make a 3x3 color space conversion matrix you need the x, y for Red, green, blue and white (white point). It's linear algebra. I would also take a look at Bruce Lindbloom's website. Graeme's paper provides the RGBW x, y values need to build the matrix. You will need to roll your own gamut mapping code to deal with out of gamut colors. This is part of what IPP2 does for you.
Click to expand...

Stacey,

Been creating Google accounts all day for relative's new Christmas phones and Tablets. I haven't read Charles Poynton's book, but will since you've suggested it. I will also check out the website you suggested also. I want to at least try to understand better what IPP2 does even if I never actually manage to build a matrix of my own.

Thanks again Stacey
 
Stacey,

do you mean Charles A. Poynton's

"Digital Video and HD, Second Edition: Algorithms and Interfaces" or "A Technical Introduction to Digital Video 1st Edition" or maybe both or something else?
 
If I intend to use a 3D Lut in red cine X. Do I just select redlogfilm like I would with dragon?
 
Corey Benoit said:
If I intend to use a 3D Lut in red cine X. Do I just select redlogfilm like I would with dragon?
Click to expand...

The IPP2 LUT is intended for use on top of REDWideGamutRGB color and Log3G10 gamma. Redlogfilm is now obsolete moving forward.

If you're using a third-party LUT, well, you're going to have to research what input it needs.
 
Aaron Lochert said:
The IPP2 LUT is intended for use on top of REDWideGamutRGB color and Log3G10 gamma. Redlogfilm is now obsolete moving forward.

If you're using a third-party LUT, well, you're going to have to research what input it needs.
Click to expand...

The luts I am using with dragon I just use redlogfilm. I don't think anyone has made specific luts for Helium yet that I know of. Are the skin tone banding issues going to be fixed in the widegamut/log310 or is that fixed in the ipp2? I hope it's fixed on the log end. As I intend to use luts for film grading and I don't want to have skin tone issues.
 
Corey Benoit said:
The luts I am using with dragon I just use redlogfilm. I don't think anyone has made specific luts for Helium yet that I know of. Are the skin tone banding issues going to be fixed in the widegamut/log310 or is that fixed in the ipp2? I hope it's fixed on the log end. As I intend to use luts for film grading and I don't want to have skin tone issues.
Click to expand...


Corey,

I believe I asked a similar question myself and if this isn't the same question you are asking I apologize. Here is what I asked and the answer I received

rand thompson said:
I was just playing around with the new IPP2 preview again using the 8K helium Footage provide by Mark Toia. I noticed that there was some sort of green banding "Like" areas on the caucasian skin of the female subject after applying the Ipp2 Preview Lut (that was so generously provided by Phil Holland) to the RWG/LOG3g10 settings in Redcine-X pro "One Off" preview. What causes this and can it be fixed with the "Final" implementation of the Helium color Pipeline? Forgive me if this issue has already been addressed.



I used a 16-bit Tiff. file to grade from Redcine-X pro "One off" because the .R3d file could only be imported into Davinci Resolve 12.5.4 without the IPP2 preview lut so only as Redwidegamutrgb-Log3g10.

http://www.phfx.com/temp/reduser/phfx_RED_RWGRGBwithLG310_IPP2PreviewLUT.zip

Redcine-X Pro "One Off" Preview - Redwidegamut-Log3G10 plus IPP2 Preview Lut provided by Phil.
31717041411_d86dfe680a_h.jpg
[/url]B001_C096_0902AP by rand thompson, on Flickr[/IMG]

https://www.dropbox.com/s/cb17nkdb0x7r8n8/B001_C096_0902AP.0000127F.R3D?dl=0


Redcine-X Pro "One Off" Preview - Redwidegamut-Log3G10 plus IPP2 Preview Lut provided by Phil and a grade away from the green tinge.


31799086646_dd6ff57f7e_h.jpg
[/url]Untitled00086400j by rand thompson, on Flickr[/IMG]
You can still see some green areas still present after the grade.
Click to expand...



Phil Holland said:
To fix this properly you need to update to the most recent firmware to achieve the proper calibration profile for Helium that "solves" the artifact.

IPP2 is a new image processing pipeline and not really related to the green-ish tint across mid-IREs in a given exposure. The firmware is the solution. For the time being you will need to grade the artifact out if it's effected your captured footage.
Click to expand...


And Thanks to Abe Espinoza for making me realize that you actually could work with the IPP2 lut with a .R3D file inside Davinci Resolve.


Abe Espinoza said:
No, not yet but you can use the IPP2 LUT within resolve. Set your footage to REDWIDEGAMUTRGB/Log3g10 and apply the IPP2Preview LUT as an output lut. I'm personally enjoying the look of this new improved color pipeline with Scarlet Dragon footage.
Click to expand...
 
Last edited:
Is it my imagination that the new updated firmware is more noisy???

Could just be me but just shot some lower light stuff at 2500 and it's totally un-usable.
Now, I fairly new to Red and of course the Epic-w is very new, .. but I was surprised at what I got.

I'll try and post some stills later.

Curtis
 
rand thompson said:
Stacey,

do you mean Charles A. Poynton's

"Digital Video and HD, Second Edition: Algorithms and Interfaces" or "A Technical Introduction to Digital Video 1st Edition" or maybe both or something else?
Click to expand...
Digital Video and HD, Second Edition.
 
Curtis boggs said:
Is it my imagination that the new updated firmware is more noisy???

Could just be me but just shot some lower light stuff at 2500 and it's totally un-usable.
Now, I fairly new to Red and of course the Epic-w is very new, .. but I was surprised at what I got.

I'll try and post some stills later.

Curtis
Click to expand...

Make sure you re-blackshade after every firmware update.
 
Stacey Spears said:
For those that have not noticed it yet, color is not the only thing improved in IPP2. Here is an animated GIF showing the difference in detail. GIFs have limited color depth, so ignore that, but just look at the detail difference. Sorry I can't match them perfectly in terms of levels due to differences in IPP1 vs. IPP2, but take a look at the multi bursts and wedge. Notice that one is sharper and has less color artifacts in the black and white wedge tip. This is IPP2.
Click to expand...

The improvement is significant. Must be improvements to the actual debayer.
 
This is great!! Just wondering if there's any mx sensor examples using ipp2?
 
Damian Luchow said:
This is great!! Just wondering if there's any mx sensor examples using ipp2?
Click to expand...

Phil Holland posted a few in his series of tests. It does improve things, though the differences aren't as obvious as with the newer sensors.
 
M and MX works well with IPP2. The benefits are similar to that seen on the more modern sensors.

Graeme
 
Great, was trying to find Phil's mx tests.. couldn't find it. Anyone have the link?
 
Thanks Phil!! Yeah, the ipp2 definitely has a bit more punch to it I'd say.
 
Just a quick comparison of DC2/RG4 vs RWG/Log3G10 + Ipp2 preview (on Scarlet-W)
VEKTFxv.gif
 
You must log in or register to reply here.
Back
Top