Differences in image rendering SXT>Amira>Mini

General discussion
Post Reply
ilyamarcus
Posts: 2
Joined: Mon Mar 20, 2017 3:56 pm

Hey,

I have done a quick test over the weekend running many different cameras Arri SXT/Amira/Mini + Sony F55/F5/Fs7 + Varicam35 + Red Dragon.
I had some mistakes that I did during the testing and I would definitely test cameras differently next time.

Having that said, I have noticed a difference between the Arri cameras both in sensitivity and color rendering.
At T8 I had to tweak the SXT by 1/3 of a stop down and the Mini by 1/2 stop to get to the same exposure, and on T11 I left all cameras at the same exposure so I could see a visual representation of any difference in the sensors.
All files are read in Davinci Resolve in an ACEScc project with an Alexa IDT viewed on a calibrated Rec709 Monitor, all shots Are done with a Zeiss Master Prime 35mm.

Would you say this is normal? Are all Arri cameras supposed to give you the same result given that they are in the same environment?
You do not have the required permissions to view the files attached to this post.
Brant
Posts: 9
Joined: Sun Nov 13, 2016 6:59 pm

Your test is really interesting. I too have found aesthetic differences, though more anecdotal (not tested) between the Alexa Mini and Amira. Below is a comparison of my own, totally different shoots but with the same lens set. So it is not too meaningful, but I did see differences I did not expect. Interestingly, Arri made no comment on that post. I wonder if they will comment on yours!

https://www.arri.com/forum/viewtopic.php?f=105&t=329

Can you provide higher quality frame grabs from your test? I was able, after making many adjustments, to get the Mini and SXT to match nearly identically. But I am surprised these cameras were so different looking out of the gate, in terms of sensitivity and and color. The Mini was much cooler in highlights, and registered the gray card quite a bit higher in IRE.

Is this something that is unavoidable in camera design or implementation? Could it be addressed in future SUPs? If both the Mini and SXT were recording in RAW, would they then be identical??
ilyamarcus
Posts: 2
Joined: Mon Mar 20, 2017 3:56 pm

I hope I will be posting soon a long blog post/video about some of the stuff I saw in my testing, till now I can say I was very interested in how the Varicam35 works and must say its quite amazing. So as soon as ill post that, I will attach HQ picture.

I definitely didn't expect to see that the Amira is 1/2 stop darker then the Mini and 1/3 then the SXT. Also it seams that the SXT is much cooler/greenish and that the Mini is quite warm.
I had the understanding that all Arri cameras should match, especially if I take RAW out of the equation and use base settings on all of them. Would be amazing to get some clarification about this from Arri.
Jan Heugel
Posts: 516
Joined: Wed Aug 13, 2014 3:15 pm
Location: Munich, Germany
Contact:

Hello everyone,

all ARRI cameras are being calibrated up to our high standards when they undergo burn-in and functionality tests after assembly.
Slight differences may occur and can easily be corrected.

If not in post but directly on set matching the cameras closer can be achieved by performing an "auto white" on a grey card for all cameras. In case you are on a multi-cam show, differences can be addressed via live painting.

Cheers,
Jan
Jan Heugel
Application Engineer
Post Reply