FS-100 Picture Profiles

Rick Burnett said:
I don't think so.

(1) The graph doesn't even show ITU709, look at the legend.

(2) You are looking at the output, that is just showing you how the information is being expanded or compressed into a smaller space. If you look at the input side of the graph, they all have the same input length values. Just because you squeeze down the top and bottom doesn't mean your DR from your input isn't kept, it's just over a smaller range of values in the stored value. Basically, all you are doing is changing how far apart the bars are on the display, but not how many bars. Unless of course your curve completely crushes out the bottom, or blows out the top, which this is not showing.

The scope isn't going to lie. I believe from my eyes that both Cinetone 1, Standard and ITU are able to obtain the same dynamic range. I think Cinetone 2 crushes the blacks too much to the point shadow detail is NOT kept and you do lose DR.

I also think that the FS100 was overexposed in that shot. The FS100, like the AF100, does better exposing for the highlights, which does not look like the case.
Click to expand...

You're right...I was looking at the standard...duh. It just seems the itu709 gives us the most info...but maybe it's because the curve is simply pushed higher...essentially increasing the rgb output? So for grading....which setting is best? Hmmm
 
Haha, well when I originally looked at this document, I was like, what?!? Where is the curve for itu709. I'll admit that you've asked the $10Million dollar question. Not having time to do a lot of testing I shot with Cinetone2 two weeks ago and the editor loved the footage. Now, I am switching to Cinetone1 as I want a tiny bit more DR, but really it comes down to knowing what you want to do with the footage. If I am going to be crushing the blacks anyways, then I'd be more inclined to crush them in camera and not waste data throughput on something I don't need. Given the cameras is 8-bit, I always shoot with the 'as close as I can' in camera mentality.

That is a good question about the output if it uses higher values in the 8-bit range. Given the compression of the dynamic range from the sensor to the 8-bit format, it might not give you more dynamic range, overall, but it just might give you a slight bit more depth in the values of the highlights (since they are not compressed down). This would only really matter if something you are doing will be expanding out the highlights on a curve in my opinion, and never something I do for narrative work. Normally, I might curve the highlights even more.
 
What bothers me is that I feel I have to use a higher fstop for ITU709 and lower fstop for cinematone. That's what's making me think that cinematone is the flattest curve and 709 simply pushes the mid part higher. Hmm I'm going to go check which one grades better so I can sleep tonight....lol
 
Texas Jones said:
What bothers me is that I feel I have to use a higher fstop for ITU709 and lower fstop for cinematone. That's what's making me think that cinematone is the flattest curve and 709 simply pushes the mid part higher. Hmm I'm going to go check which one grades better so I can sleep tonight....lol
Click to expand...

The way I see it, the camera struggles with highlights, but handles shadows and low-light superbly. Therefore, in post, it's easier to boost the darks than it is to drop the highs. If you're using a higher f-stop with 709, you're protecting your highlights more. Seems like the right way to go to me.
 
Texas Jones said:
Sure thing.
Click to expand...

Thanks, very interesting

Having had a good read I guess that I will be steering clear of too much knee to save highlights

I think that DR is a trade off with 'contrast/colour fidelity'

As the top ends are flattened the tonal and brigtness variations are reduced, at some point they will come so close together as to not be resolvable in the small color/compression space

result? banding and blocking

can a grade pull back info that is not resolved.. no

Just a guess

S
 
Last edited:
The way I see it, the camera struggles with highlights, but handles shadows and low-light superbly. Therefore, in post, it's easier to boost the darks than it is to drop the highs. If you're using a higher f-stop with 709, you're protecting your highlights more. Seems like the right way to go to me.
Click to expand...

That's what I'm leaning toward.

Re-watching the video, the Standard and ITU709 are exactly the same on the scopes. Cinematone just isn't right to me. Middle grey is at 40% which means the midtones are pushed down stretching the area from grey to white creating high contrast and steep curve, while the shadows and midtones seem to muddy together a bit more because this area of the curve is being squeezed causind less contrast and flatter curve. What I want is a somewhat straight curve, middle black at 50%, and as much highlight and shadow detail as possible.

Standard.jpg

ITU709.jpg


What I was guessing before was that Standard/ITU709 was pushing the midtones up, in actuality Cinematone is pushing the mids down.

I'm thinking of sticking with the Standard for awhile. Grey is at 50% and it retains a little more highlight info than ITU709. So here's my settings and I'm sticking to it!! (At least until I'm convinced otherwise...which will probably be tomorrow, haha)

Black Level: +8
Gamma: Standard
Black Gamma: Middle, +7
Knee: Manual, 85%, 0 slope
Color Mode: Cinematone 1
Color Level: +4
Color Phase: 0
Color Depth: All 0
 
Last edited:
What are people thinking in regards to color mode?

What kind of strengths have people noticed in Cinematone 1 vs 2? Why are you picking 1 over 2?
 
my settings are very similar to yours Richard and I have been avoiding cinematone 1 and 2
only thing I don't understand is the Color Level +4? what is the reason for that?

Cinematone 1 seems to be a little warmer and more saturated than Cinematone 2
 
speedracerlo said:
my settings are very similar to yours Richard and I have been avoiding cinematone 1 and 2
only thing I don't understand is the Color Level +4? what is the reason for that?

Cinematone 1 seems to be a little warmer and more saturated than Cinematone 2
Click to expand...

Great question. I have no idea. Lol

No seriously I did that to match my usual ideal end-result color so I wouldnt have to push it too far in post. My thinking is that in a 4:2:0 color space you should get it close in camera and make minor changes in post if anything. But now that you mention it...I will probably just set it at zero to keep bleeding minimal.

(see told you I would change my mind, haha)
 
Black Level

Black Level

I just connected my new Fs100, looking at a 12 step black wedge, on a scope(waveform monitor). It looks like the pedestal settings(when black sits right on RGB16) for Black Level correspond to -15 for STANDARD and ITU709 modes, and -8 for the Cinegamma modes. Raising the Black Level above these values result in muddier and noisier blacks.
 
Last edited:
Hey Cuervo, what does RGB16 refer to? Also, by pedestal, do you mean the BLACK LEVEL setting in the picture profile menu? Setting it to -15 doesn't crush the blacks?
 
cuervo said:
I just connected my new Fs100, looking at a 12 step black wedge, on a scope(histogram). It looks like the pedestal settings(when black sits right on RGB16) for Black Level correspond to -15 for STANDARD and ITU709 modes, and -8 for the Cinegamma modes. Raising the Black Level above these values result in muddier and noisier blacks.
Click to expand...

those noisy blacks can be crushed in post to look just like Black -15 or whatever setting you have
it would be harder to recover any lost information in dark shadow areas if you shoot with crushed blacks like that
 
speedracerlo said:
those noisy blacks can be crushed in post to look just like Black -15 or whatever setting you have
it would be harder to recover any lost information in dark shadow areas if you shoot with crushed blacks like that
Click to expand...

RGB 16 corresponds to zero IRE, or full black in the NTSC ITU709 color space.
Relative to the comment that one risks crushing the blacks with this setting, I would, in general, agree. However, I would add that the DP has to trade crushing the blacks with noisy and muddy blacks. I would rather mildly crush the blacks, assuming a WFM isn't available for run and gun. In a studio environment, the use of a WFM is pretty critical to nailing the right black level. I am suggesting that these settings I provided, get you pretty close if you don't have a WFM handy. I've seen a lot of suggestions that Black Level should be set above zero. These measurements seem to confirm that that just isn't the right way to go. If one feels uncomfortable with setting black up against the wall, then use a setting between zero and these settings to allow some headroom...or footroom, if you prefer.
 
Cuervo, if you had a color grading stage in your workflow, do you still think keeping your blacks low is ideal? Or would you raise them? If so, any thoughts as to where?
 
Neex said:
Cuervo, if you had a color grading stage in your workflow, do you still think keeping your blacks low is ideal? Or would you raise them? If so, any thoughts as to where?
Click to expand...

I prefer that the captured image is as close to the final coloring as possible. Of course, this isn't always possible. Once the shadows have been crushed there is no recovering the lost information from the image. Even in a 10 bit color space, raising the black level in post will produce noise. So, capturing with as low a black level as possible, without losing shadow detail is the goal. As I said, in an ideal world, I would always set up my shots with a waveform monitor. Since this isn't practical, knowing the bottom limit and raising the black level is a question of experience with different lighting conditions and the amount of shadow in the scene. I can't stand low contrast, milky looking footage with poorly defined blacks.

In my workflow, I convert all footage to 10 bit when captured to my workstation. Then, I can apply an inverted S-curve to correct the hi's and lo's. The more correction needed, the greater the chance of banding(stairstepping) and noise, not to mention the increased rendering time needed for CC. CC is destructive to compressed 8-bit formats, and, should be avoided whenever possible.
 
Last edited:
cuervo said:
In my workflow, I convert all footage to 10 bit when captured to my workstation. Then, I can apply an inverted S-curve to correct the hi's and lo's. The more correction needed, the greater the chance of banding(stairstepping) and noise, not to mention the increased rendering time needed for CC. CC is destructive to compressed 8-bit formats, and, should be avoided whenever possible.
Click to expand...

Why convert to 10bit? It doesn't add info into the image, just makes the file bigger.
 
ProRes (a standard format used in all flavors of Final Cut even the much maligned version) is a 10-but codec partially for this very reason. It gives you more latitude in your color corrections to adjust beyond the original data which is your starting point.
 
Texas Jones said:
Why convert to 10bit? It doesn't add info into the image, just makes the file bigger.
Click to expand...

It is not necessary in some apps to convert first to 10-bit on input, but, that depends on the internal representation of the video information. That said, what IS important is that you work in 10-bit and you output in 10-bit if you plan to do anything else with it. 10-bit through an entire process will reduce quantization errors in color correction and other image manipulation. ALWAYS work higher than your target bit depth. This is true for video AND audio. I just find it easier to convert EVERYTHING I do to ProRes 422 and work from there.
 
You must log in or register to reply here.
Back
Top