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

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...

Cool - where can I find this LUT? Is it in the latest version of resolve, or do we need to download it from somehwere?

thx
 
Phil Holland posted a download link containing the LUT. It can be found on post #156 of this thread. It is not in the latest version of Resolve. It's the LUT Helium equipped cameras output which I am using within Resolve.
 
Last edited:
Nick Morrison said:
Forgive me if this has been mentioned before, but is IPP2 available within Resolve too?

And how does it effect Dragon footage? (whether it's Epic Dragon or Weapon Dragon).

thx
Click to expand...


Nick, I didn't know you could select the Ipp2 preview Lut as an output Lut until Abe said you could. But below is what Weapon Dragon footage looks like from redcine-X pro build 42.41605 from Red.com and redcine-X pro build 42.41985 "One Off" preview.

Kelvin: 5000
Tint:-16.000

DragonColor2-Redgamma4 no grading applied

redcine-X pro build 42.41605 from Red.com

30996280334_7882ebc0d2_h.jpg
[/url]W001_C005_1014V3 by rand thompson, on Flickr[/IMG]


31464132880_42b19bfd7b_b.jpg
[/url]Capture by rand thompson, on Flickr[/IMG]


Kelvin: 5000
Tint:-16.000

Redwidegamutrgb-Log3g10 + IPP2 Preview lut no grading applied

redcine-X pro build 42.41985 "One Off" preview


30996281204_7ff29a34eb_h.jpg
[/url]W001_C005_1014V3 by rand thompson, on Flickr[/IMG]


31464133100_b6d8fefc0c_b.jpg
[/url]girl in chair redwidegamutrgb-log3g10-ipp2 lut RCX-one offj by rand thompson, on Flickr[/IMG]
 
Awesome, thanks Rand. Nice comparison. The RGB parades are interesting. Looks like Green and Blue channels have been pulled back, while RED channel hasn't been touched.
 
Nick,

If you look closer, It seems that the reds have been dialed back some but how much of a difference it will make .......but what do I know about color transformation. What I know about the mathematical transformation of color could fit on the head of a needle.


DragonColor2-Redgamma4 no grading applied-Red channel

31721168901_020a96cc3c_o.jpg
[/url]Capture crop by rand thompson, on Flickr[/IMG]


Redwidegamutrgb-Log3g10 + IPP2 Preview lut no grading applied-Red channel


31721168831_20a6da2b6a_o.jpg
[/url]girl in chair redwidegamutrgb-log3g10-ipp2 lut RCX-one offj crop by rand thompson, on Flickr[/IMG]
 
Last edited:
Michael Tiemann said:
Ugh. I don't care about external recording. I have a SHOGUN and 2:1 magnification of an HD signal is not the same as a 1:1 magnification of 8K. I just want to MONITOR it so that I can confirm CRITICAL FOCUS all day long. If IPP2 somehow works around whatever stopped RED from sending a crop of the 1920x1080 pixels in the center of the sensor, that would make my day.
Click to expand...

My suggestion would be to use the edge focus mode during takes. It is a far better tool for judging focus during takes and it doesn't disrupt anything in the imaging pipeline like magnification apparently does.

Just make sure you have tools turned off for the client monitor so they don't freak out when their image has turned mostly black with some white and yellow.

(Ironically, on one of my first RED music video jobs, I forgot to turn the tools off and the director ended up liking some of the "looks" so we captured them with an external recorder hahaha.)
 
Michael Tiemann said:
Ugh. I don't care about external recording. I have a SHOGUN and 2:1 magnification of an HD signal is not the same as a 1:1 magnification of 8K. I just want to MONITOR it so that I can confirm CRITICAL FOCUS all day long. If IPP2 somehow works around whatever stopped RED from sending a crop of the 1920x1080 pixels in the center of the sensor, that would make my day.
Click to expand...

Also, there was a minor bug a while back that might help explain why RED cameras can't magnify during recording. When you magnified, the resolution would change in the top menu bar to 2K. So if the magnification is just a punch in of the sensor, then it would make sense why you can't magnify during record. It would be like shooting HDR, but with two separate sensor sizes.


If RED sensors has dual pixel auto focus like the C300II, then they could probably magnify while recording pretty easily.
 
Nick Morrison said:
Awesome, thanks Rand. Nice comparison. The RGB parades are interesting. Looks like Green and Blue channels have been pulled back, while RED channel hasn't been touched.
Click to expand...

Looking at RGB parades is not really telling you what's going on. In some places the changes from IPP2 are really subtle, but the over-all effect is not. If you put up the two images and flip between them you'll get a much better idea on the changes, which really can't be described in tems of levels in channels.

Graeme
 
thought I'd create a .gif..for Christmas ha ha drumroll cymbal

from Redwidegamutrgb-Log3g10 + IPP2 Preview lut(Darker) to DragonColor2-Redgamma4(Brighter)


31040334953_32d06081e0_o.gif
[/url]output_neGJZ12 by rand thompson, on Flickr[/IMG]
 
Last edited:
I truly appreciate the efforts, but comparing DragonColor2-Redgamma4 with Log3g10 + IPP2 Preview LUT is really comparing apples to oranges- one uses a video gamma based image and the second is a log type. You might as well compare ACES workflow with a LUT based one. Those two also use different underlying tech...
Also, is there a reason why you chose different brightness for two resulting images? Skintone is the most important as well as the most telling part of any image. Making them equal in brightness would probably tell use a bit more about the new color reproduction.
 
Graeme Nattress said:
Looking at RGB parades is not really telling you what's going on. In some places the changes from IPP2 are really subtle, but the over-all effect is not. If you put up the two images and flip between them you'll get a much better idea on the changes, which really can't be described in tems of levels in channels.

Graeme
Click to expand...

Gotcha. Thanks for the update. Looking forward to seeing how IPP2 works w/ Dragon footage!
 
jake blackstone said:
I truly appreciate the efforts, but comparing DragonColor2-Redgamma4 with Log3g10 + IPP2 Preview LUT is really comparing apples to oranges- one uses a video gamma based image and the second is a log type. You might as well compare ACES workflow with a LUT based one. Those two also use different underlying tech...
Also, is there a reason why you chose different brightness for two resulting images? Skintone is the most important as well as the most telling part of any image. Making them equal in brightness would probably tell use a bit more about the new color reproduction.
Click to expand...

In both cases we're comparing an output image targeted at a standard HD display so there's nothing wrong with the comparison at all.

Graeme
 
jake blackstone said:
I truly appreciate the efforts, but comparing DragonColor2-Redgamma4 with Log3g10 + IPP2 Preview LUT is really comparing apples to oranges- one uses a video gamma based image and the second is a log type. You might as well compare ACES workflow with a LUT based one. Those two also use different underlying tech...
Also, is there a reason why you chose different brightness for two resulting images? Skintone is the most important as well as the most telling part of any image. Making them equal in brightness would probably tell use a bit more about the new color reproduction.
Click to expand...

I did not change the brightness, its just the difference between the .tiff output image from Redcine-X pro "One Off" preview(redwidegamutrgb-Log3g10+ Ipp2 Preview Lut) and Redcine-X pro from RED.com(Dragoncolor2-Redgamma4)

I just used the "Darker" and "Brighter" tags so viewers could tell which was which, which came first and which came second.

So the Difference you are seeing "Is" the difference between the two.

Didn't mean to cause any confusion.
 
Last edited:
Graeme,

I was looking back through all of the S-Curves I had built to bring Redwidegamutrgb-Log3g10 into the Rec709 colorspace and found this one.

31740965641_5327c8fbd1_o.jpg
[/url]graph of curve by rand thompson, on Flickr[/IMG]



31740965721_725996a7a8_o.jpg
[/url]old curve by rand thompson, on Flickr[/IMG]


Here is what it looks like plus a secondary for skin and contrast and without turning her Beige top to white.



31711746252_be15836c97_h.jpg
[/url]Untitled00086527 by rand thompson, on Flickr[/IMG]

With beige to white top conversion.

31722377962_72ea86ce49_h.jpg
[/url]Untitled00086527tw by rand thompson, on Flickr[/IMG]


31753177531_0dd3aa0f69_b.jpg
[/url]top white by rand thompson, on Flickr[/IMG]
 
Last edited:
Graeme Nattress said:
In both cases we're comparing an output image targeted at a standard HD display so there's nothing wrong with the comparison at all.

Graeme
Click to expand...

What is a standard HD display- sRGB, Rec-709, then what gamma, P3 at what what white point? And even then, how do you explain the brightness difference?
In Baselight I can use ACES workflow with Photometric DRT vs standard Truelight workflow using the appropriate standard Photometric LUT and the brightness result will be identical. I would very much prefer an identical LUT workflow- RedLogFilm vs Log310 with the appropriate LUTs. Could it be, that Red never bothered to release an official RLF LUT? Yes, yes, I know, RLF is just a standard Cineon curve...
Anyway, only then you could truly see and appreciate the improvements. Otherwise, I'm sorry, but to me the test is still apples to oranges...
 
rand thompson said:
Graeme,

I was looking back through all of the S-Curves I had built to bring Redwidegamutrgb-Log3g10 into the Rec709 colorspace and found this one.
Click to expand...

But you're not doing a colour space transform, so your result is not REC709 colours. To get out of RWG to another colour space you need to do that convserion with a matrix, and also deal correctly with out of gamut colours.

Graeme
 
You must log in or register to reply here.
Back
Top