Shoot at Kerner Optical: RED 39, an F23, Greenscreen, and Tests Galore

[sbroock]

This past Friday I had the good fortune to bring RED #39 to the amazing facilities at Kerner Optical in San Francisco. Kerner is the legendary ILM model and stage complex that now caters to a wide range of major motion pictures and visual effects clients. Be sure to check out their website at www.kerner.com.

The visit was made possible by my friend Alex Lindsay, former ILM'r and founder of the Pixel Corps, and we were joined by Ted Schilowitz and Deanan DaSilva from RED. As a special treat, they brought "Capetown" -- one of the pre-production RED ONEs.

The goal of the day was a side by side comparison of RED #39 and a Sony F23 under a wide range of conditions. Extensive chart tests measuring color, contrast, resolution, etc. were recorded, and we will have a full write up of the results very shortly.

As there is a great deal of interest, we also shot green screen tests of actresses and, separately, blown smoke.

2K Quicktimes exported from RED Alert were immediately available for Projection and review in the Kerner auditorium.

We shot extensive additional green screen material later in the evening at a separate location and shot plates in the hills above Golden Gate Bridge on Saturday.

We will be posting extensive samples for everyone to play with.

In terms of the green screen, which many of you are curious about, our jaws absolutely dropped when we saw the results of our shoot. There is absolutely no question that this camera is the real deal for extracting the cleanest keys you have ever seen. To whet your appetite, I have attached one quick sample below (plate, alpha, comp). This is with ONE CLICK in Shake using Keylight. Excuse the JPGs for now, I'll link to full 4K Tiffs so you can try pulling your own keys.

It was a phenomenal day -- a huge shout out to Alex, RED, and the Kerner team.

You will love your cameras. More surprises to come. :)

-Scott

[jbeale]

Wonderful stuff, looking forward to your writeup!

[Alexander Nikishin]

Look forward to reading the full write-up.

[Sanjin Jukic]

Amazing!

[sbroock]

Another quick "one click" key out of Shake using Keylight.

Caveat -- JPGS so you will see artifacts if you try pulling keys on the images. We will be posting full plates.

[JD Holloway]

Oh man that chiffon soft focus shot/comp. rocks
Looking forward to detailed report.

[filip kovcin]

please send also sony's f23 files for comparision!!!

filip

[Justin Kirchhoff]

Looks awesome. I'm excited to see the results.

[Kevin Halverson]

Can't wait to see what is said about this test over on CML!

[Andrew Benz]

Can't wait to see what is said about this test over on CML!

I was thinking the same thing. I guess I need to sign up again but this time with a secondary e-mail account... I got worn out with back and forth and the tangents. I am still weeding out my primary e-mail account from the mostly spam bs... but the good post are worth it. :nerd:

I am really interested in the results and ensuing fallout... no matter where it lands as long as it is accurate. I see no reason to doubt that at all...

Thanks for the thread Scott.

[Kevin Halverson]

Hey Andrew,

I agree, there is a fair amount of crap flung around on CML, but there is more than enough good stuff there to be worth the weeding out process. It can seem more than a bit heavy handed over there, but if all you are going to do is lurk, you'll be just fine.

Just keep a healthy skepticism for the outcome of any tests and use them as a guide for your own testing not as a final arbitrator. Its not that I am assuming bias (which may or may not happen) but I have seen some pretty incompetent testing regiments with equally unreliable results.

None the less, the process is fun to watch. The drama can get a bit like daytime soaps, but its all good fun from the side lines!

[Andrew Benz]

Hey Andrew,

I agree, there is a fair amount of crap flung around on CML, but there is more than enough good stuff there to be worth the weeding out process. It can seem more than a bit heavy handed over there, but if all you are going to do is lurk, you'll be just fine.

Just keep a healthy skepticism for the outcome of any tests and use them as a guide for your own testing not as a final arbitrator. Its not that I am assuming bias (which may or may not happen) but I have seen some pretty incompetent testing regiments with equally unreliable results.

None the less, the process is fun to watch. The drama can get a bit like daytime soaps, but its all good fun from the side lines!

It is good fun and the good posts are well worth it. Right now, I am rebuilding my career and keeping my circle of questions to here and some pm's to great people... you will not see me on cml.:ninja: :help:

I fully agree with you as well on testing... this is another area where I will be asking some questions, so all is the guidance very appreciated.

I will certainly travel to get my hands on and test all the major items, its great to have people here that answer questions and move me further along the process in an intelligent manner.

Cheers-- Andrew

[Tony Lorentzen]

Very impressive stuff. Thanks for sharing, guys!

[David Cubbage]

That key with the scarf is amazing; thank you so much Scott for sharing! I look forward to reading the full report!

David

[Christopher Barry]

I hope we get to see some A/B comparisons of the footage between the two cameras.

Regarding the scarf key, it seems strange to me, the alpha file does not reveal the small green screen section between the arm and stomach. Unless the alpha frame is from a different part of the sequence?

[Christopher Barry]

After zooming in, it does not appear to be keyed, and also, that section I noted appears to be about the same colour and luminance of the beach background plate, so blends well.

[sbroock]

After zooming in, it does not appear to be keyed, and also, that section I noted appears to be about the same colour and luminance of the beach background plate, so blends well.

To be sure, you will find elements in this key that could be improved, but that would be "not seeing the forest for the trees." ;)

In a normal VFX pipeline, you would tend to isolate trouble areas with separate keys, combining alphas in a complete matte.

Here, the purpose of the post is not to show a refined key, but what was possible with a one click sample of the background.

I hope to have a nice 4K plate from the shoot for everyone to play with later today.

-Scott

[Chris Gearhart]

Fantastic! I've been wanting to compare Red with an f23 or the viper also, so I'd look forward to your report and A/B also. Thanks, Scott.

[Gavin Greenwalt]

I'm on the edge of my proverbial seat! When's the footage coming? :D

[Adrian T.]

Waiting... :whistling:

[Floris Liesker]

Someone I know since IBC compared a Red to a Viper.
Actually, his Red to his Viper.
Red did very well, he said. Details coming up.

[Brice Ansel]

Now the red codec is around I would love a short sequence r3d file of that shooting.
Cheers Brice

[Jim Arthurs]

BTW, since this thread is back, you gotta love the full page ad from BandPro in the Oct American Cinematographer for Kurner, where they're heaping praise on their new F23 cameras (note the plural here...)

How does that situation stand now after the RED tests?

[Michael Schrengohst]

And now that REDCINE is out - could we get a greenscreen clip?

[brandon herman]

we're these results ever posted? in a different thread?

thanks

[Jim Arthurs]

we're these results ever posted? in a different thread?

thanks

This might be related...

http://www.reduser.net/forum/showthread.php?p=121635#post121635

Pity we didn't get more info on this...

[sbroock]

The good news is that the material and feedback exists on the comparison.

The hold-up lies in the write-up which is being generated by Alex with analysis by Kerner. Unfortunately, Alex decided to have a kid a few weeks ago and is a bit behind.

The NYU event was an overview of the RED for the students and the plates we shot at Kerner.

Mark Pederson and I made a presentation to the Faculty of the NYU Film school several weeks go, and they were blown away.

We'll give much more notice for the next event at NYU which may be in Feb.

[Noah Posnick]

Glad there's going to be another event in February. I was signed up, but unfortunately missed it and would love to see the footage and hear what you guys gotta say.

[Esmaile Neissi]

Amazing tests thanks scoot

[Esmaile Neissi]

Amazing Tests guys

[BASSAM MSSALATIE]

wooow :innocent: impressive keyring .:angry02:

[steve palazzese]

.....no words.

waiting for the F23 comparison.

[Finner]

Lattitude wise my $ is on the f23. Thought the f23 is a lot more $ to get that dynamic range.