"True Colour" Scene File

[Paolo Ciccone]

Originally Posted by Luis Otero

Paolo,
Thanks a lot for this great effort from your part, specially because you have been willing to share all the information with this cybercommunity.

Just a small contribution to a community that helped me so much :)

Originally Posted by Luis Otero

In terms of white balance: you use the settings and then WB, or WB and then load the settings? I remember that that was an issue brought by Tim with his efforts. If I do not recall incorrectly, I think we needed to load the setting, then do the WB. I assume is the same you have found...

That is correct. The problem is that the HD100 stores the values of the A and B WB presets together with your configuration. If you, for example, had A WB at 4300K and B WB at 3200K when you saved a configuratio to the SD card, they will be restored when you reload that configuration. I didn't realized it the first time I calibrated the camera. BTW, WB is also affected by the ND filter so be sure to reset it every time *anything* changes in your setup.

Originally Posted by Luis Otero

In regards to the sharpening, what is the setting you particularly recommend?

Just a step above the "OFF" position, it's labelled "MIN"

--
Paolo
http://www.paolociccone.com

[Paolo Ciccone]

Originally Posted by Marc Colemont

Paolo,

FYI, The DVRack software guys confirmed there is a color space problem in the HDV mode for the HD-100. It uses 609 currently. It will be fixed in the next patch.

Marc, good to know and a little disappointing. Did they say if there's an ETA for the fix?

--
Paolo

[Marc Colemont]

Originally Posted by Paolo Ciccone

Marc, good to know and a little disappointing. Did they say if there's an ETA for the fix?

--
Paolo

Nope, They said to wait for the fix. At the time of the implementation of the HDV, they didn't had a HD100 to testout. They used another camera which apperently did not have the right HDV specifications and used another color space.
They are going to try if they can make an 'auto' color space setting, otherwise it will probably a menu setting they will add.
I hope all your work was not for nothing now.

[Paolo Ciccone]

Originally Posted by Marc Colemont

Nope, They said to wait for the fix. At the time of the implementation of the HDV, they didn't had a HD100 to testout. They used another camera which apperently did not have the right HDV specifications and used another color space.
They are going to try if they can make an 'auto' color space setting, otherwise it will probably a menu setting they will add.
I hope all your work was not for nothing now.

I don't think so. I'm rechecking the settings with the FCP VScope. I'm trying to find a way of feeding the camera signal to FCP with the minimum amount of "jumps" :)
I see that Serious Magic is Folsom. I could drive theretoday and bring them my HD100 for testing. Do you know the name of your contact in Serious Magic?

--
Paolo

[Paolo Ciccone]

Originally Posted by Marc Colemont

I hope all your work was not for nothing now.

Marc, thank ou for your concern. The Gamma, black and master black settings are not affected by this. I switched the camera to DV output for FireWire and cheched the color matrix and it's fine. There is a good article about the different color spaces and compensation for checking 709 on NTSC monitors at http://www.dsclabs.com/product_reviews.htm Check the "Image optimization" document.

My configuration is still the closest to 1:1 representation for color, if anybody can get any closer I'll be happy to change my parameters. As it stands now I believe that this setup is superior in accuracy and color rendering than the stock scenes.

For the first time, since I bought this camera, I'm getting the vibrant colors and image quality that I was expecting. I believe that, if you shoot the same scene with, say, Cinelike24pHD and my config, you'll be able to verify the difference by naked eye.

Ciao!

--
Paolo

[Marc Colemont]

Yes I tried your settings, the colors make indeed a difference :-)

My contact at Serious Magic is Benoit Ambry about the subject.

Serious Magic is Folsom... I didn't knew that. Great guys at Folsom.
I worked with them at Barco Belgium two years ago.

[DJ Lewis]

Paolo, in an earlier post someone said that they would use your color settings, but use tim's black stretch/knee tweaks. would this through off the colors drasticly? And are your settings acceptable for film out?

[Paolo Ciccone]

Originally Posted by DJ Lewis

Paolo, in an earlier post someone said that they would use your color settings, but use tim's black stretch/knee tweaks. would this through off the colors drasticly? And are your settings acceptable for film out?

No, mixing my color matrix settings with Tim's gamma and black stretch should not change the color result. I don't want to start arguing about this but I believe my settings, as published, will give you more latitude. But make your own judgement. Take a look at the results from the WFM and vectorscope that I publicshed. Your goal is to make the white chip register at 100 and the black chip at 0. If you look at the result of my gamma/black settings you see that it's pretty damn close to that.
Shoot the same scene with both settings, bring them into your NLE and check the result on a good monitor. Use what you like the most. Using the SD card it's a snap to switch between configurations. Keep in mind that I configured the camera for a "neutral" look. My target is to use all the limited bits available to capture as much as possible, leaving the job of altering and clipping those bit in post. Your approach could be different.

Depending on your type of shooting this could be right or no. You mentioned filmout and so I guess you're gonna shoot in a controlled environment where you can check the light levels. In this case my setings will give you the wide latitude , I believe close to the maximum available with this camera.
The one setting that you might want to change is to switch the detail to "MIN".

--
Paolo
http://www.paolociccone.com

[Steven Thomas]

Originally Posted by Paolo Ciccone

The one setting that you might want to change is to switch the detail to "MIN".

Paolo
http://www.paolociccone.com

Paolo,
Have you done any tests with the detail setting?

I'm interested in determining the effects of this control.

To me it appears to actually go beyond turning off any artifical enhancement. The image appears to be blurred

[Stephen L. Noe]

Hi,

DSC also makes a skin tone chart. Just wondering if you got the set of DSC charts or the single chart. The reason I ask is because the "true color" setting seems to make skin tone too rich and the suttle nature of gradients surrounding facial features seems to be missing. Browns and pinks are too saturated.

What are you findings where skin tone is concerned?

BTW: Don't get me wrong. Your work is excellent. I was wondering if you see the same thing I do where skin tone is concerned using the "true color" scene file.

[Tim Dashwood]

Originally Posted by Paolo Ciccone

I don't want to start arguing about this but I believe my settings, as published, will give you more latitude. But make your own judgement. Take a look at the results from the WFM and vectorscope that I publicshed. Your goal is to make the white chip register at 100 and the black chip at 0. If you look at the result of my gamma/black settings you see that it's pretty damn close to that.
Shoot the same scene with both settings, bring them into your NLE and check the result on a good monitor. Use what you like the most. Using the SD card it's a snap to switch between configurations. Keep in mind that I configured the camera for a "neutral" look. My target is to use all the limited bits available to capture as much as possible, leaving the job of altering and clipping those bit in post. Your approach could be different.

Depending on your type of shooting this could be right or no. You mentioned filmout and so I guess you're gonna shoot in a controlled environment where you can check the light levels. In this case my setings will give you the wide latitude , I believe close to the maximum available with this camera.

Paulo,

Some people are getting confused.
What you have done is bring the camera into a ITU709 (or 601 depending on how DVRAck was setup) color space spec. Bringing the white chip to 100IRE is the OPPOSITE and contradictory of extending latitude (AKA dynamic range.)
This is in no way close to the maximum available latitude with this camera. Your knee is set at 100%... essentially turned off.
Your own histograms prove that knee at 80% and black stretch 3 render the most latitude.
The chips on the DSC chart are not representative of the brightest white in the world or the blackest black, they are just a spec, like colour bars, that allow all of us around the world to be on the same page.

I suggest reading this article by Michael Bergeron. http://www.24pdigitalcinema.com/cinegamma.pdf

It refers to the Panasonic Varicam, but the concepts are exactly the same.

[Jim Giberti]

Originally Posted by Paolo Ciccone

Im quite convinced that the disabling the sharpening blurs the images a bit while keeping the sharpening at minimum seems to be the right setting in order to get a nice sharp image that is still soft enough to not look like video.
I also checked the effect of incremental sharpening with a WFM and you can clearly see the addition of video noise.

--
Paolo
http://www.paolociccone.com

Pretty much where I'm at with sharpening as well.

[Paolo Ciccone]

Hey Tim.

I'm a student of the School of Hard Knocks and you just gave me astrong lesson, thank you! :)
I'm not a cinematographer but I've been a professional bit pusher (aka software engineer) for 20 years and it was hard to figure out that a digital limiter, my own interpretation of the knee circuit, was going to give me more in the end product instead of less. I kinda immagined the knee as something similar to the audio auto level.
I couldn't find clear information about the subject, the most detailed description I found was in a portion of Scott Billps book: "In an attempt to soften the abrupt clip levels of video the knee compression circuit was devised". Of course the word "compression" didn't settle well with me. Later in the page he adds: "while this often creates better looking video, it isn't recommended for video destined for film due to the amount of information that is discarded in the equation" (boldface mine).

The article that you linked partially confirms this and the "squished" result that I'm getting from the WFM, when loading some other configurations, was interpreted by me as loss of data and not as "we are below the threeshold, we are safe".

Sometimes you gotta make a fool of yourself in front of thousands of people in order to gain some knowledge. Fair trade. :)

Anyway, do you know if the cinelike gamma in our camera is equivalent to the cinegamma circuit mentioned in the article?

Thanks again.

--
PaOlo
http://www.paolociccone.com

[Stephen L. Noe]

Paolo,

The "true color' scene file pulls off some great overall saturation. I had it loaded in the camera while I was driving and shot some out the window stuff.
Lakeshore Drive Chicago IL (175MB) Info about the camera setup is this. I used the "true color" scene file WB 6500. ND2 with Aperature set to around F2.8 and shutter 1/500. The scene file pulls off some great color in this scenario.

[Luis Otero]

Paolo,

I have been playing utilizing your parameters, and using Tim's Black Stretch and Knee parameters. Let me tell you that such combination has given me by far the best out of the HD100.

Just my 2 cents...

Luis