What magic in this sensor gives you this incredible image?

Amadou

Amadou

Carbonite Member
This was filmed by amateurs with a RED MYSTERIUM. Look at how cinema the image looks. Right below you have the FS7 image which looks much more video. This has nothing to do with RAW or bitrate as the FS7 has a great bitrate and looks vitually the same as the FS5 which looks like video. There is something else at work here and I can't figure out what it is. The Red Image just looks more even in the costrast and the edges are less sharp. Also, I don't really see the difference between the FS5 and FS7 image. So where does the magic come from?

RED ONE
https://www.youtube.com/watch?v=koQO45GtV3o&t=108s

RED Scarlet X Mysterium
https://www.youtube.com/watch?v=UWx0UUs7H1w

FS7
https://www.youtube.com/watch?v=pq8ttzVV56k&t=16s

FS5 (Look at the high contraty)
https://www.youtube.com/watch?v=_TrlDpDrYAA
 
I'm trying to figure you out. Are you in the market for a camera? Do you have a FS7 and are trying to validate it in your mind? You keep starting all these camera threads and constantly refer to the FS7, so what's your angle? Unless you can answer that, this will be my only reply to all your posts. The camera helps with a cinematic look, but is second to things like lighting, camera movement, sound and acting.
 
No one knows and that's the bottom line.

ARRI is #1 in the world and RED is #2 in the world and everything is far below as far as cinema is concerned.

You'll hear that a million other things matter in production (and they absolutely are important), but the bottom line is these two companies are used the most in Hollywood and that's that.
 
Shoot all those cameras in the exact same settings and the illusion of magic will disappear. 100% lighting, lenses, diffusion/glow filters, and grading is the major factor of differences in those videos. Yes there are other differences between cameras, but they are not the major factors in the final look.

Watch more shootout videos in the exact same settings:



ARRI is the easiest tool to create predictably great results. Red (below Phantom speeds) has slomo covered. For Netflix, the F55 has become much more popular due to the 4K rules (which completely cut out ARRI until the (very limited) rental-only ARRI 65 and the newly released Alexa LF (over $100K, so not going to knock Sony and Red from the most used 4K cameras)). The new Sony VENICE could become the new top Netflix 4K camera given the cost vs. performance.

We can better recommend a camera system if you can post examples of your work so we can see your shooting & grading styles etc. I shot my first narrative short on the A7S (I), GH4, and (one shot) 5D Mark III (raw). Higher end cameras would not have made it any better. If you're shooting for yourself (vs. others), something like a Sony A7 III is more than sufficient for a narrative: 6K => 4K for True 4K (and no oversharpened edges that you don't like), decent skin tones, lightweight, low cost (so you can have many cameras), great low light, and so on. Yes, the A7 III looks better than the FS7. You don't need 10-bit or raw to get a cinematic look. Expect the A7S III to look even better (rumored to have the VENICE color science).

Here's a bunch of cinematic films shot on the iPhone: https://variety.com/2018/film/news/unsane-tangerine-films-iphones-1202730676/
 
El Ditector I am trying to raise my camera general knowledge level. I shot a movie 10 years ago with an Ex1 and Ldtus. Now I hqve a new 90 minute film to shoot in Africa for a theatrical releaseI started out wanting to buy a Red One on Ebay. But people talked me out of it as the Red One is heavy and unreliable. I did not have a plan B... so I took a second look at the FS5 but was not overly impressed albeit I think I could work with it. This said it felt too small and too simple to be a movie camera.. So I tested other cameras such as the EVA 1 and the C200. Both of these cameras yielded strange files which were unreadable on Premiere CS5. The C200 file was especially large and unrecognizef. I then rented thd FS7 and liked the image more than that of the FS5... but not by much. By then I had wasted a month trying to figure out this camera thing. Finally I played it safe and decided on the FS7... got a great deal for the FS7 mark 1 with two 128 cards and extra large battery for 7500 Euros. Just then another supplier offered the FS5 for 3500 Euros. 4000 Euros difference. I asked myself whether the FS7 was really necessary insofar as both cameras allow for quite a bit of menu options... this said somebody told me the Ursa mini was the right choice down from the Red. Ever since I have been asking myself why I wdnt from a 5000 Euro Red One ciminema camera to a 7500 Euro video camera... nog to mdntio the 3500 Grezt deal FS5 . SO WHENEVER I GHINK OF BUYING THD FS7 my suconscious sayw "no this is too expensive snw us a video camera" and when I ghink of the FS5 my suconscious says:"You started with q Rew and now want to make your movie with this? People will laugh when they see this camera." I suppose I can't makd up my mind because I am disappointew for not having gotten that Red.
 
Static studio shootouts have been useless since day 1.

As a former RED owner and having spent a small fortune on Japanese systems, my personal opinion is there is an intangible difference.

I have been asking for almost 10 years...show me (anyone) something cinematic from these other cameras...because I'm still looking myself.

As far as the iphone films, 'Unsane' is a good story...a good story that looks like it was shot on a phone.
 
By that same argument why are we bothering with high end cameras at all? You and I know there is a reason we dont shoot on IPhones.
 
Alexa LF looking like video:

Comment about the Alexa LF looking like video:

Dear Arri, I m a huge Arri lover ,and have been using ARRI camera s from film to Digital .....but compare to other cameras like Alexa XT and SXT ...somehow i feel the cinematic look is lacking in these sample footages.....thought low light capturing ability is splendid as per the footages i have seen here....(Really i would like to test it if it s availiable in India)Overall ..it can capture the market of music videos and commercials i think..but not CINEMA..
Click to expand...

ARRI's response:
The ALEXA LF uses the same photo site size, type and color science as ALEXA XT and SXT, it just has twice as many photo sites as ALEXA XT and SXT. So you can always shoot with the ALEXA LF and crop a Super 35 image out of the center, and you will get EXCATLY the same image as you would from an ALEXA XT or SXT. When using the larger sensor of the ALEXA LF (assuming the same T-stop, ASA setting and distance to subject as with a Super 35 ALEXA) you a shallower depth of field and a different perspective, which makes especially faces look better, more three dimensional. However, large format may not be for you, so we continue to provide ALEXA SXT W, ALEXA Mini and AMIRA cameras as well as a large range of Super 35 lenses. You choose what you would like to use for your production, and we are sure different markets will adopt large format for different type of productions.
Click to expand...

"more three dimensional" is something shallower DOF can help provide, however it's not about sensor size since smaller sensors with wider aperture lenses can provide equivalent shallow DOF. Lens choice has a big effect on "three dimensional" and filmic/cinematic looks: http://www.thehurlblog.com/lens-tests-leica-summicron-c-vs-cooke-s4-film-education/

The new Sony VENICE looks cinematic and is super clean with pristine 6K => True 4K detail. Along with full frame shallow DOF, it's creating a new kind of look (technically possible with Super35 sensors provided lenses with equivalent apertures (which currently don't exist)):
http://www.dvxuser.com/V6/showthread.php?360234-Sony-VENICE-Impressive-Skin-Tones
Using the same lenses and similar color work the A7 III could look very similar (also a 6K => 4K full frame sensor), though the VENICE color and detail will likely look better if compared side-by-side.

A useful exercise is taking whatever camera(s) you are using and making look like video on purpose, as well as making look as cinematic as possible, using the same clips. This will show you the contrast and color aspects of video/cinematic looks.

Here you can practice with ARRI's latest cameras, the Alexa 65 and LF (as well the older Super35 SXT): http://www.arri.com/camera/alexa/learn/alexa_sample_footage/
If just graded for basic contrast and color (which takes just a few seconds), the Alexa 65 footage looks more like video than cinematic (ARRIRAW). Do a more aggressive grade, perhaps add some grain, and it will look more cinematic/filmic. The LF footage uses a different lens, more cinematic lighting, and has more grain (lower light), and thus looks more cinematic with a basic contrast + color grade. However it still looks a bit like video and will take more work to look cinematic. Cinematic is a bit unreal- different highlight/shadow color, anamorphic bokeh/flares, grain, etc.: artifacts which lend to an emotional response, best used to set a mood to help tell the story. Which you'll experience if you watch the iPhone examples provided (more so if you didn't know they were shot on iPhone ahead of time). If you can't make your current camera look cinematic, perhaps worth practicing until you can until spending money on a more expensive camera. You can also practice on clips from ARRI's top cameras (both ARRIRAW & ProRes).

Another fun exercise is to make your footage look like old VHS video. Shot with the Canon 5D3 and Sony FS700:

You can apply these techniques to any camera: making footage look more cinematic as well as more like video, including old VHS. Once you do these exercises you'll see the magic is in the mind of the creator and not the camera.

Some tips where the 8K Panavision DXL (Red sensor) processed in post to look cinematic, like grainy 16mm film, etc. (from: http://www.thehurlblog.com/cinematography-resolution-is-not-sharpness/ ), as well as great info on lenses, looks & supersampling:

The Super35 / VistaVision (full frame+) comparison at 55:55 is wrong as they didn't use equivalence and had to move the camera which changed the perspective, completely invalidating the comparison of S35 vs. full frame (ARRI & Red/Panavision are still pushing the full frame look myth. Canon & Sony who make FF still cameras, interestingly are not pushing the myth).
 
Last edited:
Amadou said:
… I am trying to raise my camera general knowledge level. I shot a movie 10 years ago with an Ex1 and Ldtus. Now I hqve a new 90 minute film to shoot in Africa for a theatrical release …
…
Click to expand...
That doesn't answer your shooting style, which will determine what camera may suit your needs better.
Will it be narrative style, documentary style, wild life, or any combination of those?
Will it be one-man-band, limited crew, full crew?
Is 4K a must, or only nice to have?

Amadou said:
…
... So I tested other cameras such as the EVA 1 and the C200. Both of these cameras yielded strange files which were unreadable on Premiere CS5.
…
Click to expand...
Choosing a cinema camera that Premiere CS5 will handle its' files seems to me the wrong way to go about it.
Try DaVinci Resolve, either the freeware version, or the studio version for $300. Then your choice of camera wouldn't be limited by ab ancient NLE.
 
We use Nikon manual lenses with manual follow focus. I prefer to get audio directly into the camera. We have a crew but the Director does the cinematography. Editing and final export is done in Premiere 5.0 for a 2K DCP.
 
No matter what anyone says there’s mojo in different sensors and sensor sizes. All things are not equal between sensor sizes when adjusting apeture and FOV. Or equating lighting filters and grading.

Been through this discussion a hundred times on this forum. I swear I could see it and many says it’s just a better DP or lighting and point to scientific tests.

But the reality is I can spot the difference in my EVA vs GH5/5s footage in a second. Consistently. Even when using same adapter lenses. And I’m the same DP; I don’t suddenly get better going to the EVA. And I can spot the difference in c200 vs EVA1 in the way it renders the image. It is definitely different.

Yes, all the other details etc matter a ton but there are differences in sensors beyond the spec sheets etc. there are a ton of variables that go into capturing and processing the light on a sensor and so while no one knows the exact differences or what causss it, it’s there. Smoother edges, a “bigger” feeling, nicer gradient, more organic, whatever.

RED and ARRI have great mojo.

Lenses matter too, quite a bit, but you won’t always see it in scientific tests. I have a set of Zeiss glass and they do a little something magic to images that my sigma 18-35 just does not.
 
Amadou said:
We use Nikon manual lenses with manual follow focus. I prefer to get audio directly into the camera. We have a crew but the Director does the cinematography. Editing and final export is done in Premiere 5.0 for a 2K DCP.
Click to expand...

You're dooming yourself for a world of pain and hurt, even failure. Don't do this for a feature film.
 
I dont think those two red clips looked particularly cinematic... the hipster woodchuck chuck had a blown out sky and was slowmo hell...

The other one was a night shot.

Cinematic had more to do with how you handle light and the movement of the camera that what sensor you throw at the scene.

What you need is the exact same scene lit by a gaffer and then shot with various cameras to really see where the magic is. Maybe even handle the raw footage yourself.
 
You must log in or register to reply here.
Back
Top