(link) a very useful real-world Red One MX test by John Brawley

[Erich Ocean]

http://johnbrawley.wordpress.com/2011/06/02/posted-available-light-tests-are-finally-online/

Red One MX fared well. Based on the comments (on the linked page) of the original showing, I feel pretty confident that the Epic would have also tied the Alexa on dynamic range. The 12-bit processing in the Red One really seems to be the limiting factor on the sensor. Resolution-wise, Red One MX was the clear winner.

The highlight handling on the Epic is in a whole 'nother league compared to the Red One. I can't wait to see the next round of tests with Alexa and Film vs. Epic.

 

[Joel J. Feigenbaum]

Great study. Thanks for posting the link.
It was easy to identify 16mm, Sony F3 and Canon material. Much tougher to I.D. Alexa, RED MX and 35mm on the Vimeo feed. Made myself smile when I flagged RED MX as the most pleasing look and guessed it was RED MX.
Can you imagine how Epic would have done?

 

[Eric Santiago]

3 hours sleep and I wake up to this link at ProVideoCoaliton
I didnt check the link above hoping more info on the project itself.
And of course my fave was the RED MX however others faired wel too

 

[Jesper Sichlau]

Good test, I guessed the Red One and 1D right.

 

[Michel Hafner]

Liked Alexa best followed by 35mm and Red MX. Red MX had clippy highlights and drowning shadow detail at times. Was that really optimally exposed and graded?

 

[Erich Ocean]

Liked Alexa best followed by 35mm and Red MX. Red MX had clippy highlights and drowning shadow detail at times. Was that really optimally exposed and graded?

That DP has done a ton of RED shooting, so I'd have to say "yes". I really do think the MX is ham stringed by the 12-bit AD on the Red One; Epic is clearly much improved in highlight handling (where the "bits" are at their most sensitive).

 

[Peter Strietmann]

I'm curious what film stocks were used for the 16mm and 35mm. Nicely done, looks like a lot of work to produce this. Thanks.

 

[Jason Richter]

while i appreciate the test and all the work that went into it i think that this piece is unfortunately poorly put together as a demonstration of the differences with each camera. first off it's over 18 minutes. Not sure why you needed to shoot each short 6 different times with each camera and then show each short to us in full. Plus, why did you keep which camera was used till the end? I would have actually like to know what i was looking at. Then to either have to write it down myself or remember which camera went with each letter and then go back and look at the pieces again, it could have been much more thought out. I hope you put together a real quick maybe 2 minute comparison and actually show what camera is associated to what shots during those shots.

J

 

[John Brawley]

Hi Peter.

5219 and 7219.

jb

 

[John Brawley]

As mentioned in the original post, this was a test designed to be seen in a cinema. Maybe you didn't read it to get the context of what we were attempting.

It was designed as a piece of drama first, not a set of test charts. We wanted each short film to have a chance to have it's own emotional impact. Individual shots cut against each other won't have the same dramatic impact. There are plenty of those kinds of tests out there already.

6 x 2 min shorts is actually 12 minutes, not 18 mins. The other 6 mins of material is what you asked for. A series of individual shots cut one after the other, plus some 200% resizes and some examples of the ungraded material to show how much we had to work the shots.

We chose to withhold the cameras because people bring their prejudice with them when they know what the material is that they are watching. This was a "blind" test to see what people liked first, not so they could confirm what they think they are already seeing.

jb

 

[Stefan Antonescu]

I don't get this. Everything looks so bad except Alexa...

 

[Solomon Nero]

I don't get this. Everything looks so bad except Alexa...

Well, i thought the RED, ALEXA and 35mm looked "better" then the others, but I think the really crushed blacks and colors are all because of the over-done color grading...when comparing samples from different cameras...its probably best, if you decide to grade, color grade...lightly...

 

[Erich Ocean]

Well, i thought the RED, ALEXA and 35mm looked "better" then the others, but I think the really crushed blacks and colors are all because of the over-done color grading...when comparing samples from different cameras...its probably best, if you decide to grade, color grade...lightly...

The grading was for cinema -- 35mm and DCI -- so naturally, a web video is going to look crushed in the blacks (and not have super accurate color). Nevertheless, I'm still glad the material was released -- you still get a relative sense for how the cameras fared next to each other. I would love to see the material released for additional screenings in the original formats.

 

[Julio Quintana]

The blind approach was really interesting and kept me engaged. I went into the test with the following assumptions: Alexa has better range than M-X, and the F3 has better range than 1D. I wrote down my guesses as I went along, and these assumptions led me to incorrectly switch Alexa with M-X, and F3 with 1D. I guess more expensive isn't always better.

I think the most shocking thing to find was just how bad the F3 looked. Thanks for doing this, Erich.

 

[John Brawley]

I think the most shocking thing to find was just how bad the F3 looked. Thanks for doing this, Erich.

I would caution against making too harsh a judgement of the F3. The version we had was an engineering sample provided by Sony before the camera was released. I'd suggest that the finished camera is a lot better. This test was shot in December 2010.

jb

 

[Erich Ocean]

Thanks for doing this, Erich.

Well, I'm happy to post the link, but John Brawley did all the work. (John, I wish I knew you were on RedUser, I'd have PM'd you and had you start the thread instead...)

 

[Zach Gray]

John,

Thank you so much for sharing this test. For me it was very useful. I let you evaluate if the movie would work dramatically or if camera artifacts would take you out. It's an excellent test for a storyteller, less so for an engineer. I also appreciated that it was in a pretty raw shooting locations. In the real world everything is a bit out of control. Thanks for the considerable time and effort.

 

[sander kamp]

Great test and thanks a lot for sharing this! Even with the limitations of Vimeo you still get an idea of how the cameras do.

 

[Antoine Fabi]

The Alexa has a rich, lovely image for sure.

Antoine

 

[Sanjin Jukic]

John,

thanks for a nice test.

It's a pity that we can't see it in cinema but from Vimeo downloaded file still it can be seen clearly who is who.

Film 35mm Kodak 5219 is still a king in terms of color rendering, balanced DR and smooth roll of latitude.

And of course no clipping with film material at all.

Then comes Alexa and after that RED1-MX.

The rest of the cams we should forget for a certain reason at the moment...

Samples 01

Samples 02

Samples 03