Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Alexa, Red MX & Sony F3 Camera Test Comparison

Camera tests

Camera tests

Although we are not told how the test was done, my first impression is that Alexa and F3 are aliasing like crazy, on the F3 it's clearly visible in the test charts, in the skin and on the brick building..
 
They are all beautiful cameras in their own way.

Alexa definitely has a nice organic quality to it though it does seem a tad soft at times.
Very nice overall tonality in that camera.

The F3 has a decent image though the aliasing on the res chart is a little surprising. Interested how the dynamic range might look with S-log gamma.

Looking at the skin tone window shots on the RED I have to wonder how blown out that histogram was. Definitely a shot where exposing for the windows and bringing the shadows up in grading would have worked. Of course RED's resolution was no comparison.

Wasn't the MX sensor supposed to eventually operate at a faster (5ms?) refresh to minimize rolling shutter?

Interesting test though.

Brian Timmons
britim-media
 
Can't wait for someone to do these test that actually knows what they are doing...

Jim
 
Jannard said:
Can't wait for someone to do these test that actually knows what they are doing...

Jim
Click to expand...

Here here!

#1. Each of those DR tests was exposed differently. Totally differently.
#2. Do we know what we're looking at with the Red? OK, the other two cameras record video files but was the Red footage output from the RAW at log? Hang on, no, it's REC709. So, once again, we see what amounts to the Red's video monitoring path compared to the recorded output of the other cameras.

The F3 certainly has Sony Color™ and bad aliasing. Gotta say, if it cost $3k it would be worth a look.
The Alexa wins the rolling shutter battle (assuming the pans were equal speed) and looks like it has a ProMist on the lens - ie: nice, pleasing skin tones but a touch soft.
The Red's rez stands out like dog's balls. Anything else we can't be too sure of...

Love to test Alexa to MX with a ProMist...
 
Shane Betts said:
Here here!

#1. Each of those DR tests was exposed differently. Totally differently.
#2. Do we know what we're looking at with the Red? OK, the other two cameras record video files but was the Red footage output from the RAW at log? Hang on, no, it's REC709. So, once again, we see what amounts to the Red's video monitoring path compared to the recorded output of the other cameras. .
Click to expand...

I think this indicates a fundamental problem with some of these comparison.
Each camera can have pretty different approaches that have an affect on the end result.

I think many people still don't quite grasp RAW.
If you don't expose RED without a decent understanding of how RAW works and just what you can do with it in post, you will not nearly reach what the camera is capable of.

Kind of like this recent test shot:
http://reduser.net/forum/showpost.php?p=741362&postcount=1


Brian Timmons
britim-media
 
So here are just some of the issues I have with this test.

1. Resolution. Resolution matters when presenting large, not small. If you shoot any of these cameras for a feature, we need to see what they look like enlarged to the same screen size. That may be a 60' screen or a 10' home theater. Not a 5" Vimeo presentation. At least scale up the smaller clip to matching size and publish a 100% section.

2. Color time the clips and grade to match. It is a bit unfair to take whatever output you got and post that. What did they use for RED? REDspace, REDcolor, REDcolor2, REC 709? My bet is they did not use REDcolor2 and REDlogFlim and grade from there. RED footage is designed to be graded. That's what REDCODE RAW is all about. If you choose to publish something like this, let us know what you did.

RED is moving. Stay current. I have heard that it is difficult to stay current with RED color space and gamma. Not as difficult as buying new cameras every time something changes. Inside the time that the RED ONE was released until now, Arri has released the D20, D21 and Alexa. If you owned all of them, you spent over $500K. In that same timeframe we got you to spend $6K on a sensor upgrade (on top of your $17,500) and learn a couple of new color spaces and gamma options. Seems like a fair deal to me but only if you pay attention. The price you pay with RED is upgrading your firmware, software... which is N/C. All you have to do is learn the best workflow before posting test results that people believe as gospel.

Someone will do these tests properly and post the results in a way that showcases these cameras in the proper light. The Alexa does have a bit more DR than the RED ONE MX. The RED ONEs (all of them) have a ton more resolution, less aliasing and moire then the others. OK... then comes EPIC. Can't wait.

BTW... who does a resolution test on an Ang. 24-290? Really?

Jim
 
Last edited:
Jannard said:
BTW... who does a resolution test on an Ang. 24-290? Really?

Jim
Click to expand...

You're going to love their next test.

Lensbaby3G.jpg


:smile5:
 
Definitely need 1:1 crops. The aliasing caused by the compression also renders the Vimeo version of the test useless for comparing resolution.

It would have also been nice to have a machine that moves each camera the exact same way for the skew test, but I think that may have been out of the scope for this kind of run and gun test. :}
 
I'll accept a little slop like failing to grade the charts. But when you shoot one dynamic range test on a cloudy snowy day and the other two on sunny clear spring days... I have no idea what you're hoping to demonstrate.
 
Jannard said:
RED is moving. Stay current. I have heard that it is difficult to stay current with RED color space and gamma. Not as difficult as buying new cameras every time something changes. Inside the time that the RED ONE was released until now, Arri has released the D20, D21 and Alexa. If you owned all of them, you spent over $500K. In that same timeframe we got you to spend $6K on a sensor upgrade (on top of your $17,500) and learn a couple of new color spaces and gamma options. Seems like a fair deal to me but only if you pay attention. The price you pay with RED is upgrading your firmware, software... which is N/C. All you have to do is learn the best workflow before posting test results that people believe as gospel.

OK... then comes EPIC. Can't wait.


Jim
Click to expand...

Power
 
Jannard said:
RED is moving. Stay current. I have heard that it is difficult to stay current with RED color space and gamma. Not as difficult as buying new cameras every time something changes. Inside the time that the RED ONE was released until now, Arri has released the D20, D21 and Alexa. If you owned all of them, you spent over $500K. In that same timeframe we got you to spend $6K on a sensor upgrade (on top of your $17,500) and learn a couple of new color spaces and gamma options. Seems like a fair deal to me but only if you pay attention. The price you pay with RED is upgrading your firmware, software... which is N/C. All you have to do is learn the best workflow before posting test results that people believe as gospel.
Click to expand...

Word Up. :coolgleamA:
 
Even if one of these tests ... ONE, actually graded the RAW footage and compared that to the video from the other cameras. Just one!

Next test I wanna see is Alexa, vs MX vs F3 vs the video split from an Arri 435 ...
 
IMHO... the tests that people will want to see in the coming months will be EPIC vs. the Sony F65.

Jim
 
Jannard said:
IMHO... the tests that people will want to see in the coming months will be EPIC vs. the Sony F65.

Jim
Click to expand...

Hey, that will be great!!
 
Jannard said:
... At least scale up the smaller clip to matching size and publish a 100% section.

Jim
Click to expand...

Well, actually, given that we're comparing a 4k camera to a couple of HD cameras, wouldn't be fairer to compare a 100% section of the Red to a 300% section of the other two?
 
Alexa vs. RED ONE resolution. Notice moire. EPIC is obviously a more pronounced difference.

Jim

1300801576.jpg


Interesting marketing. Wonder if anyone really buys this stuff...

1300845197.png
 
Last edited:
You must log in or register to reply here.
Back
Top