Sony A7SIII = $3500, 10-bit 4K120p internal, 16-bit 4K60p RAW out

Stuart Mannion said:
How do people feel about the potential of the raw out to the Atomos? Is there a chance it could near Redcode quality? I think it's a good sensor with good colour-science so why not? The only footage as far as I know is Bloom's Paddleboarder. I played with the grade on that and was happy with what I could get. I had concerns about the skin-tones in the close-up but then heard Bloom say he did a lot of secondaries on his face to remove a green tint refection from the trees.
Click to expand...

The quality will be amazing. It will be extremely high-quality video like with all of the Japanese cameras.

But it's not going to look like a RED unless you dumb down the RED to match it.

There is an intangible difference between those cameras and the Blackmagics vs. the others. (ARRI is for the 1%.)

It is impossible to explain.

(But most agree it's due to the minimal sharpening (if any), motion cadence, dynamic range and color science.)
 
Thanks for the response. Yes the sharpening, noise reduction and motion cadence are a big part of what gives the Sony's a different look than Red/Arri. I am hoping with the raw we can at least avoid the sharpening and noise reduction somewhat. The a7siii does seem to have a bit of that video-feeling motion blur but I'm hoping that with a slightly faster shutter-speed I can mitigate that.
 
Saw this the other week...this is pretty much their flagship for most as the Venice is for the 1% too.

FX9 ProRes RAW.

Just doesn't do it for me in a cinematic sense (and it's not the grade because I have seen similar stuff with BMs and REDs that was barely tweaked and you can just see and feel the difference).

High-quality video though, looks great for that reason...but can we expect the a7SIII to be any better?

https://www.youtube.com/watch?v=JVrg3pSyaZM
 
NorBro said:
Saw this the other week...this is pretty much their flagship for most as the Venice is for the 1% too.

FX9 ProRes RAW.

Just doesn't do it for me in a cinematic sense (and it's not the grade because I have seen similar stuff with BMs and REDs that was barely tweaked and you can just see and feel the difference).

High-quality video though, looks great for that reason...but can we expect the a7SIII to be any better?

https://www.youtube.com/watch?v=JVrg3pSyaZM
Click to expand...

Yeah it's not great. I think with a better (more cinematic) lens and better grade it would improve a lot. I don't like how the skin tones are compressed like they're about to clip. I've seen much better stuff from the FX9, and some of the a7siii footage looks better (https://www.youtube.com/watch?v=y7qOIXsxnN4) despite the 10bit limitations. I almost feel that the a7siii looks better for not being so clinical.

But yes even with all of those things being right there's still that motion-blur/motion cadence thing, especially with the slow-mo that feels 'video'. Will using a faster shutter help that?
 
IMO, using a faster shutter (too much) would make everything look more like video, and choppy.

For slow-motion, it's of course recommended...a lot of slow-motion I used to do got down to 90° and less sometimes.
 
NorBro said:
Saw this the other week...this is pretty much their flagship for most as the Venice is for the 1% too.

FX9 ProRes RAW.

Just doesn't do it for me in a cinematic sense (and it's not the grade because I have seen similar stuff with BMs and REDs that was barely tweaked and you can just see and feel the difference).

High-quality video though, looks great for that reason...but can we expect the a7SIII to be any better?

https://www.youtube.com/watch?v=JVrg3pSyaZM
Click to expand...

That footage looks promising to me. But his grade is all over the place.

Sometimes the greens have that sickly sony hue and a lack of subtlety and variation. Other times they were more mellowed and subdued.

The skin tones generally looked nice to me but I agree with Stuart that the highlights looked compressed and near-clipping.

Leads me to believe I would disagree with his exposure and grading choices.

I think the motion cadence looks okay. Most slow-motion video looks sort of blurry like that to me.

I've seen some A7SIII stuff on youtube that had terrible motion cadence. Then I played it at 0.25x and it became apparent that it had held frames and interpolated frames - it had been time remapped to an odd rate. Other samples I've seen looked fine.

I think the A7SIII is a fine camera and there's lots of color to work with. It wont hold back good lighting, exposure and grading, whether you're recording internally or ProRes RAW.

I've seen plenty of bad looking stuff shot on RED, arri and BM. They may have DR and/or color depth advantages over the Sonys but not by much.
 
This footage has been referenced before but I think it is probably the best overall look I have seen from this camera (Matti wakeboard footage). The image does not scream any manufacturer, just great color and motion. I guess his treatment of the footage is spot on as so much other stuff just looks affected. If I could guaranty I could reproduce this look easily, then I am sold. But most footage does not look this way...

So what is it that is creating the "feel" that Stuart is mentioning? And how can it be avoided?
 
Play with this footage and see if you're happy - https://redirect.viglink.com/?forma...drive.google.com/drive/folde...vrja5IrzGgRnhY

It's the only unedited slog-3 footage I've seen around

It really surprised me that matti said he graded from scratch with saturation and contrast. My attempts didn't look good at all doing that and I had to use Luts to remap the color. (Maybe he used a different gamut?)

I dont think the camera has a motion cadence problem. I think people are screwing up the shutter speed or editing
 
ahalpert said:
I've seen plenty of bad looking stuff shot on RED, arri and BM. They may have DR and/or color depth advantages over the Sonys but not by much.
Click to expand...

Plenty? I doubt it...

Usually it's one or two examples here and there, or that one famous ARRI movie everyone links to but never links to anything else.

But if you think the above looks promising then maybe a lot of the good stuff just looks bad to you.
 
NorBro said:
Plenty? I doubt it...

Usually it's one or two examples here and there, or that one famous ARRI movie everyone links to but never links to anything else.

But if you think the above looks promising then maybe a lot of the good stuff just looks bad to you.
Click to expand...

Some examples off the top of my head...

A lot of stuff shot on RED by hacks who can afford one and dont know what they're doing. Underexposed, over-sharpened

The netflix seasons of arrested development were shot on RED and looked like crap. Even moreso because they cropped in for coverage.

Side effects by Soderbergh. Shot on red. Looked like crap

I've seen so many many little shorts and stuff on youtube shot on BM cameras that looked bland or poorly graded or underexposed.

Arri is hard to look bad. But I saw a little youtube recipe video shot on a mini lf that looked like crap. I think it was because of the fluorescents they shot it with.

It's the same reason they could make a great-looking movie with the a7sii. Because lighting, lenses and grading are more important than your camera choice, at least once you have an adequately performing camera. Superlative performance will yield only a marginal improvement. Even an alexa can be made to look like crap
 
It's very difficult to find a bunch of inexperienced ignorant folks shooting on an Alexa. The camera is great but it also helps that only a few select people get to use it. I'm sure it would look pretty average if I used it XD
On the other end of the spectrum, people shooting with a $3k Sony and one $30 light (or no controlled lights at all) are basically all over the place.
 
ahalpert said:
Play with this footage and see if you're happy - https://redirect.viglink.com/?forma...drive.google.com/drive/folde...vrja5IrzGgRnhY

It's the only unedited slog-3 footage I've seen around

It really surprised me that matti said he graded from scratch with saturation and contrast. My attempts didn't look good at all doing that and I had to use Luts to remap the color. (Maybe he used a different gamut?)

I dont think the camera has a motion cadence problem. I think people are screwing up the shutter speed or editing
Click to expand...

Link is broken, BTW
 
NorBro said:
IMO, using a faster shutter (too much) would make everything look more like video, and choppy.

For slow-motion, it's of course recommended...a lot of slow-motion I used to do got down to 90° and less sometimes.
Click to expand...

One of the problems I've seen a lot with the F55(and I own one) on shows, is that whenever they do slow-mo with it, it almost always looks like horse****. It just looks like someone slowed down regular 60P "video". And more than likely, it's a combination of lighting and shutter speed.
 
Slowing down regular 60p video is how you create slow-motion in a 24p timeline. Or the camera can do it for you. But the end-result is the same.

How does the F55's look bad?

The only time in my life I absolutely butchered 60p slow-motion on a shoot was on a silly Panasonic camera where the shutter was allowed to be 1/50.

It was an oversight because of clicking around and switching back and forth between 60p and 24p all day. This problem is mostly solved with many cameras these days offering a 180° shutter option.
 
NorBro said:
Slowing down regular 60p video is how you create slow-motion in a 24p timeline. Or the camera can do it for you. But the end-result is the same.

How does the F55's look bad?

The only time in my life I absolutely butchered 60p slow-motion on a shoot was on a silly Panasonic camera where the shutter was allowed to be 1/50.

It was an oversight because of clicking around and switching back and forth between 60p and 24p all day. This problem is mostly solved with many cameras these days offering a 180° shutter option.
Click to expand...

I understand how to shoot slow-mo and how it works ; ) I was simply saying that the "look" of it just looked like "regular video", just slower. I don't see that "look" with most slow-mo I see coming out of Arri's or RED's.
 
Run&Gun said:
... It just looks like someone slowed down regular 60P "video". And more than likely, it's a combination of lighting and shutter speed.
Click to expand...
I have a question that you might known answer to. Or not.

When did the fluorescent lighting like Kino Flo become adequate for the indoor shooting in terms of the flicker free image?
 
DLD said:
I have a question that you might known answer to. Or not.

When did the fluorescent lighting like Kino Flo become adequate for the indoor shooting in terms of the flicker free image?
Click to expand...

I came into them a little late. I wasn't fully convinced in the early days, so I didn't get into the flo world until late '07/early '08, but a lot of guys had already started several years before. I'd say they really started getting popular between '00-'05(I actually remember a show I shot for saying we couldn't use them and this was around '03. They wanted tungsten only). I stopped using them around 2015.

And I've said this before, I thought flo's affected motion rendering when shooting 24p. I don't know if it's their refresh rate interacting weirdly with 24 frame/ 1/48s shutter on some cameras or what, but when I started using them, I could tell a difference on 24p(23.98) footage I shot with tungsten vs. flo's, when there was a lot of motion.
 
I recall watching Johnny Carson and, about 40 minutes into the show, they'd do a double break where the cameras would show the set/desk in the darkness before returning to another ad/promo. But that was the 1980's (well, Carson was around earlier but I wasn't) tungsten. Then Letterman was famous for keeping his set at 57F and that lasted into 2000's. Then there's a certain adult oriented film making where actors were quite sweaty but, one assumes, it was not a great idea to have them hosed down in the middle of a scene.

On the other hand, one could create a very high frequency fluorescent (20K-60K) going back into, at least, 1990's. Additionally, one could obviously vary the cam, both film and video to minimize it. But I don't think the film makers, including the big budgets, ever got rid of the old cathode ray tube TV flickering or even rolling. I'd have thought that'd be a question of timing both.

Thanks for the reply.
 
You must log in or register to reply here.
Back
Top