Technicolor Cinestyle released!

Holey smokes ...timing!!! I'm doing a camera test today...so I might have some test footage to post later tonight!
 
I'm having problems loading Cinestyle using EOS Utility 2.8.1 on my T2i - the "Register User Defined style" option is greyed out, and when I use the Picture Style Editor Cinestyle gives an error message "Cannot load image. Subsequent editing has been disabled for this file." I'm using Mac OSX 10.6.7 - anyone got any thoughts as to why this doesn't work?

edit: Never mind, didn't have it in manual mode. I are dumb!

edit 2: Well I can access Cinestyle in manual mode, but the changes don't appear to have loaded when I switch over to video mode. Is there a way on a T2i to enable video recording under manual mode by activating LiveView (like on a 5D)?
 
Last edited:
So basically, with the new picture style, the camera does not capture any sort of black spectrum off the histogram map. All picture style's I've encountered record 100% black when there's no light, but the new style only records at a 90%-ish black tone when there's no light. Not sure if this is a positive or negative thing, but it seems as though the new style is recording a contrast so flat that it's turning blacks into greys - basically, it's artificial to an extent. They need to scoot this range down a bit toward the blacks, causing the blacks to be back at %100 and the whites to be less exposed; I still think they style is too sensitive to whites.

Basically, I think it'd be interesting to see what effects would come out of a style that only recorded up to 90% black (as CineStyle does) and only up to 90% white. The thing is that with the natural post-process, sometimes slight additions of contrast get added outside our will (and especially when things are uploaded to youtube), and if the white and black spectrum was recorded at around 90%, then it'd give a bit of slip room when it comes to slight amounts of contrast being added. However, the negative to this is that with the blacks recording only at 90%, it seems as though that the %99-%90 black light spectrum would blend, actually causing less information to be formed. For instance, if you had a color chart that showed every color from %99 black to %90 black, the CineStyle would blend all of those colors while a style with slightly more contrast seemingly would record them all. When it comes to getting the most detail out of your shots, this may actually be a negative instead of a positive, though the difference is probably microscopic.

But with the whites? Personally, I'd want my whites to be like this. The reality is that with digital recording, you'll always have blown out highlights somewhere down the line, but with proper techniques (shallow depth of field, certain angles, ect ect) sometimes you can aesthetically hide this. But when the whites are still at %100, it creates what will always look like a burn-out - same with how %100 blacks tend to look inverted. If this style were to record white at %95, the majority of what people struggle with in post would at least look better - it wouldn't recover lost info, but it would make that blob of lost info look darker and smoother (or in the case of blacks, lighter and smoother). CineStyle does this with blacks, I wish it did it with whites!

Also, it's extremely cloudy outside today, so my tests with windows and stuff may be coming out a heck of a lot better than normal (making me love this new style too much), but so far, this new style seems incredibly smooth in the transitions between contrasts. But I'm trying my hardest not to put faith in it until it's sunnier outside.

Anyway, here's a picture using CineStyle - the left is without Light Optimizer, the right is with it on High. Does anyone know a way to get the lighting optimizer to stay on High all the time? (Since it variates when turned on?) With it on, the photo looks like what my eyes see in real life, and that's pretty cool!:

Dark.png
 
Last edited:
Snapper123 said:
So basically, with the new picture style, the camera does not capture any sort of black spectrum off the histogram map. All picture style's I've encountered record 100% black when there's no light, but the new style only records at a 90%-ish black tone when there's no light. Not sure if this is a positive or negative thing, but it seems as though the new style is recording a contrast so flat that it's turning blacks into greys - basically, it's artificial to an extent. They need to scoot this range down a bit toward the blacks, causing the blacks to be back at %100 and the whites to be less exposed; I still think they style is too sensitive to whites.

Basically, I think it'd be interesting to see what effects would come out of a style that only recorded up to 90% black (as CineStyle does) and only up to 90% white. The thing is that with the natural post-process, sometimes slight additions of contrast get added outside our will (and especially when things are uploaded to youtube), and if the white and black spectrum was recorded at around 90%, then it'd give a bit of slip room when it comes to slight amounts of contrast being added. However, the negative to this is that with the blacks recording only at 90%, it seems as though that the %99-%90 black light spectrum would blend, actually causing less information to be formed. For instance, if you had a color chart that showed every color from %99 black to %90 black, the CineStyle would blend all of those colors while a style with slightly more contrast seemingly would record them all. When it comes to getting the most detail out of your shots, this may actually be a negative instead of a positive, though the difference is probably microscopic.

But with the whites? Personally, I'd want my whites to be like this. The reality is that with digital recording, you'll always have blown out highlights somewhere down the line, but with proper techniques (shallow depth of field, certain angles, ect ect) sometimes you can aesthetically hide this. But when the whites are still at %100, it creates what will always look like a burn-out - same with how %100 blacks tend to look inverted. If this style were to record white at %95, the majority of what people struggle with in post would at least look better - it wouldn't recover lost info, but it would make that blob of lost info look darker and smoother (or in the case of blacks, lighter and smoother). CineStyle does this with blacks, I wish it did it with whites!

Also, it's extremely cloudy outside today, so my tests with windows and stuff may be coming out a heck of a lot better than normal (making me love this new style too much), but so far, this new style seems incredibly smooth in the transitions between contrasts. But I'm trying my hardest not to put faith in it until it's sunnier outside.
Click to expand...

That's pretty much what LOG (vs REC709) is supposed to do. The curve STARTS at +5 or 10 IRE from black and ends -5 or 10 IRE from white. Read this from David Mullen, ASC from this post:

Black seems to be between 5 IRE and 10 IRE with most of these Log formats. I don't see as much advantage at that end, once something is black, it's black, so recording a lifted black may slightly help with some color-correction of dark scenes, but it also tends to make you think you have more usable shadow detail than you actually have because once you set the blacks to 0 IRE, the shadows look darker.
Click to expand...

The LOG profile gives you a poor estimation of your shadows and highlights. The best way to get used to using this profile is to have a picture style that resembles your desired post-graded image loaded into your camera, and use that for exposure and lighting. Then switch to the CineStyle prior to pressing record. Of course if folks aren't planning on color grading all that much in post, then this profile probably isn't going to be much good for them.
 
What actually David is claiming is not totally correct... let me elaborate:

Let’s assume that we want to capture an 8 stops image from a sensor that has 12 bits internal processing that’s 4096 steps of light intensity to an 8bit compressed H264 file that’s 256 steps.

In “linear” the top brightest stop n. 8 will consume 128 steps, the stop n. 7 will consume the next 64 steps and the stop n. 6 will consume the next 32 steps. The top 3 brightest stops of an image when is being described in linear will result to “consume” the 87,5% of the available bits in a file, leaving the lower 5 stops of the mid-tones/shadows to be described by just 32 bits…!!!

On "Log" we allocate 10 stops in equivalent segments like 25 bits for each stop. That results to a secondary data compression of an applied curve, which needs to be decompressed with the use of an appropriate 3D LUT in order to counteract accurately also the saturation of the converted image. By doing that we effectively recreate information that wasn’t visible in the original file but was there in the original image the sensor was capturing.

So the difference between Linear and Log is night and day… given that you use the correct conversion 3D LUT.
 
Evangelos Achillopoulos said:
What actually David is claiming is not totally correct... let me elaborate:

Let’s assume that we want to capture an 8 stops image from a sensor that has 12 bits internal processing that’s 4096 steps of light intensity to an 8bit compressed H264 file that’s 256 steps.

In “linear” the top brightest stop n. 8 will consume 128 steps, the stop n. 7 will consume the next 64 steps and the stop n. 6 will consume the next 32 steps. The top 3 brightest stops of an image when is being described in linear will result to “consume” the 87,5% of the available bits in a file, leaving the lower 5 stops of the mid-tones/shadows to be described by just 32 bits…!!!

On "Log" we allocate 10 stops in equivalent segments like 25 bits for each stop. That results to a secondary data compression of an applied curve, which needs to be decompressed with the use of an appropriate 3D LUT in order to counteract accurately also the saturation of the converted image. By doing that we effectively recreate information that wasn’t visible in the original file but was there in the original image the sensor was capturing.

So the difference between Linear and Log is night and day… given that you use the correct conversion 3D LUT.
Click to expand...

I think he's speaking in laymans terms...but thanks for the technical breakdown. Very interesting.
 
So this with 5DtoRGB should yield the most dynamic range yet out of the camera.
I won't have my camera until next week, anyone care to test?
 
Yep thats correct... but to "decompress" the image you need to color correct it in Color i.e. with a Cineon 6500K 3D LUT in order to get the advantage of the internal floating point color processor of Color to get back that 10 stops... FCP is just 8 bits RGB... Color and DaVinci are >16bit RGB...

Have in mind though to see the difference you will also need 10bit monitoring... since the regular monitors are 8 bits... If you Filmout you will be able to see 11 to 12bits of image in the screen...
 
Last edited:
You must log in or register to reply here.
Back
Top