Welcome to our community

Be a part of something great, join today!

Register 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 :)

Little GEMINI Lowlight / Grading test.....

Tobi Rupp

Tobi Rupp

Well-known member
Joined
Mar 8, 2016
Messages
150
Reaction score
9
Points
18
Location
Germany
Hi everybody,

had the chance to play around with the new Gemini Sensor, took it outside and just filmed
and played around in Davinci Resolve 15 to get a feeling how it could perform....




The ability to use slow-motion in 5K FF without cropping the sensor is great, I love the flicker
effect with all these street lights, but would be very cool for us European guys to get straight
100fps in 5K FF....

All shots done with STD OLPF in Lowlight mode... mostly 1600 - 3200 ISO

Lenses Sigma ART Primes and one wide shot with the 12mm LAOWA ( a little underexposed by the way)

Best

Tobi
 
Last edited:
Birk Kromann said:
Looks great! Finally a RED sensor that can compete with the Alexa!
Click to expand...

Give it a speedmount and you get Alexa LF performance in a compact RED housing.

Format Resolution Aspect Ratio Max FPS Dimensions Image Circle Megapixels S35 Crop FF35 Crop
4.5K WS 4608x1944 2.37:1 134.13 27.65x11.66mm 30.01mm 8.96 0.90x 1.30x
4.5K HD 4320x2376 1.82:1 109.92 25.92x14.26mm 29.58mm 10.26 0.96x 1.39x

Even the virtual pixel size is bigger than the alexa 8.25 vs. 8.485(6x1.4142) micron with a speedmount.
Shooting 4.5k gives less rolling shutter.
Set the compression ratio to 2:1 to get lossless compressed RAW.

There is a lot to like about this camera.
 
looking good as always Tobi! Late Night , beautiful women, Red Gemini, what else could you ask for?
 
Josh Tyron said:
Awesome work Tobi! Im just waiting for min.
Click to expand...

Yeah,thanks guys, its funny whats possible to achieve with our tech these days. Not every shot is perfect nailed, but it was just me without any crew, just handheld the camera with mostly ugly poor street lights....
 
Pretty good! Nice lenses, too. This sensor is quite something.

I think people would be nuts to use any other digital camera for commercial work, except if the budget is very, very tight. You can take this thing anywhere, any time, and make it work. 8K is overkill for resolution. Raven is fine, as 4.5K is plenty, but it's not as flexible as Gemini.

Having said that, there is nothing wrong with the DSMC1 cameras, either. Older Alexas are now cheap, so they're fine, too, but way behind technically.
 
Wow. Looks real nice. Great work!
 
Hi everybody,

ok lets get a bit more technical and boring.

I did a little comparison of the actual sensors I just had at my office. (sorry for that background mess) ;-)

The "test" was not planed to share and I just did it very quick and roughly, but maybe interesting to see as well for you guys.

I shot the scene more or less well lit (at 800 ISO) , without any clipping the sensor at the blacks or highlights. Developed the same r3d files of each sensor straight in order
with the same ISO settings in Davinci Resolve 15 with the regular IPP2 workflow. (also for the Dragon/Scarlet-W with the same IPP2 ISO settings, not legacy)

The second run in that "test" I doesn't changed anything, but closed the lens 2 stops down and pushed it in the r3d development.

For me interesting to see how well the MONSTRO performing, even without dual ISO and low light mode. And how the Gemini sees way more
in the blacks....

Generally the MONSTRO tent to be a little green in the blacks and GEMINI a bit more to magenta, easy to match in the grading but
good to know.

I enabled the download function at Vimeo for the original file, so you could check it without additional Vimeo compression.

But keep in mind, its just a 4K H264 file.

Best



Tobi

 
Last edited:
Always nice to see a test, thanks for that. To bad you didn't test them with equivalent F-numbers(same angle of view) but with F numbers.
For that test it would be nice to have a fast zoom lens at hand that covers Monstro, like the Zeiss Cinema Zoom 28 - 80 mm T2.9-T22(you still would have to use the F-numbers"apature" to perform the test).
 
Awesome, love these test, do you have an HELIUM to compare? It should be interesting to compare DSMC2 sensor and view shadows and highlights recovery.
 
Maximillian,

I notice a while back when Luigi graciously posted all of that sample footage of comparisons between Monstro and Gemini, that when I applied the exact same grade to each, the gemini always had a little more magenta left behind than did Monstro. Tobi's observations appear to confirm that, at least to me.
 
BTW,


Here is a Black N' White of both the "Red Girl" and "Blue Chick" in your video.


image.jpg



image.jpg
 
rand thompson said:
Maximillian,

I notice a while back when Luigi graciously posted all of that sample footage of comparisons between Monstro and Gemini, that when I applied the exact same grade to each, the gemini always had a little more magenta left behind than did Monstro. Tobi's observations appear to confirm that, at least to me.
Click to expand...

Exactly what I am seeing, in my "test" I balanced via auto white balance and the white checker field in Resolve, what isn't always 100% on point, by the way... but all shots are more or less right balanced.
 
Thanks Tobi! I will get around to it a little later on. And Yeah, been there...done that Ha ha .
 
You must log in or register to reply here.
Back
Top