Other: Is Your Sony FX6 Soft?

As it is with S-LOG footage. All detail/sharpening to be added in post. Like in the FS7 with the detail circuit turned on it still applies 'detail' when the detail value is set to '0'. It's very noticeable by just turning the Detail menu Off/On.

Chris Young
 
I had a shoot today and with the Sony 28-135 I got super sharp and detailed footage in S-Cinetone.
 
Hello,
for me, when I film large landscape with my FX6, set in UHD, I have a lot of aliasing and moire...It's too sharp.. (with detail off, of course) This is a big problem !
It seems there isn't a low pass filter.
In HD 1920 1080, there is no problem...
Except that, to answer the first question, with faces, the sharpness is good fot me, (with detail off) , better than the C300mk2... So, in UHD, the sharpness is too hard for me, and that's probably why I notice a lot of aliasing and moiré with wide shots !
 
Are you watching it at native resolution? Often scaling on a monitor can introduce lots of artefacts.
 
lf111 said:
Hello,
for me, when I film large landscape with my FX6, set in UHD, I have a lot of aliasing and moire...It's too sharp.. (with detail off, of course) This is a big problem !
It seems there isn't a low pass filter.
In HD 1920 1080, there is no problem...
Except that, to answer the first question, with faces, the sharpness is good fot me, (with detail off) , better than the C300mk2... So, in UHD, the sharpness is too hard for me, and that's probably why I notice a lot of aliasing and moiré with wide shots !
Click to expand...

sony has confirmed that there is a low pass filter on the fx6
 
Andy9 said:
""Thicker" is one of the terms that has been used over the years to ambiguously mean better-looking without being able to explain why."

Over a decade ago I started shooting and editing raw photos. The first time I opened a raw file and played around with it was a revelation. It looked amazing! And suddenly I could make major colour and exposure adjustments with everything still looking 'right'. After half an hour I realised that I wasn't in raw mode. I'd been playing around with a jpeg.
Click to expand...

should try that on a 10bit monitor
 
Thanks for your answer. You're right, I watching on a HD Plasma monitor with Resolve... But , I compared the same shots in UHD Between : FX6 /A7SII / C300mk2. The low Pass filter of the C300mk2 seems stonger, because there is almost Zero Artefacts... With the A7SII, some artefacts. And with Fx6, a lot of Artefacts. Even with detail off, it's so sharp and precise that there's a sort of noise in tree leaves for example, or in hair and bird. I would have liked the possibility of a softer image with the detail-off, like the Canon, or RED Helium, or Alexa Mini for example. It looks like there is still some artificial detail, despite the detail-off...
I'm going to do some tests with a soft FX 1/2 filter, or Promist 1/8...
 
I fully agree concerning the FX9's image.

To me, it's a visible difference between a sensor that's made for a DSLR that also takes pictures (FX6) and one that's made for video from the get go (FX9).
The cameras look different. Especially when pushed a bit.

I somehow don't have any connection to the word "thick" when it comes to comparing the two, but I can definitely relate to the opposite, and that would be "thin" when describing the FX6's image.

The FX9 just manages to present a better separation of things moving from the background. Things look more detailed, more lifelike. In both motion and when not in motion. Whether that is because of contrast or latitude, I do not know. I believe it's just the sensor plus the 6K to 4K downscaling.

This kind of separation (or for a loss of better words "rendering" of objects) makes all the difference to me.

The FX6's image and objects look like a collage (hence "flat) to me. Like two-dimensional objects just being moved in the same plane, so to speak. The FX9 manages to go way beyond that in my opinion.

I recently did a documentary film about people working in culture one year into the pandemic.
It was shot entirely on an FX9, using only the 24-70 GM.
As a one man show (lockdown rules) having to do everything myself from lighting, camera, sound, and afterwards cut and post in Resolve, I can honestly say I couldn't have done this with any other camera (Auto Internal ND, Eye Auto Focus, 2x XLR on the body...).
After using it for the film, I loved the camera even more.

Here is the link, if anyone is interested (as always - please use the VIMEO link):

http://2meterstillstand.de
 
Last edited:
Early in the camera's release, I used to think there was a difference. I mean there should be, right? It's almost twice the price.

But after watching more footage, new footage, there's no great visual difference anyone would ever notice, and usually only we do because we love our gear and spend a lot of time looking at the motion pictures, IMO.
 
I used to own an FX9 and had used another for work at a network since launch. The downsampled 6k, with my eyes, is much sharper that the 6. It makes sense since the final image is true 4K. Now I own a 6 and sold the 9, and I don’t regret it. I used to have to soften the FX9 image a bit in post because it was too crispy, especially with older subjects. The 6 is still plenty sharp, although when shooting straight 1080 OT is a tad soft in my opinion. What I gain with the 6 is worth it to me, which is a much lighter load in the field, and the ability to shoot slow motion at the touch of a button like with the FS7 (the 9 really is a pain in the ass in this regard).
 
Anecdotal stories that claim that one camera is better than the other are of very little use to me, especially if the person sharing their opinions has not spent quality time with both cameras and has some demo footage to back up their claims -- from BOTH cameras. Someone needs to put the FX6 and the FX9 in a 3D-style rig with the same lens, S-LOG3, and do some side-by-side shooting in various situations. That is the only way you'll be able to accurately demonstrate what the differences are between the two cameras.
 
Side-by-side, with same lens, CINE EI S-LOG3, same exposure, same WB, and the same codec eliminates 99% of the variables. Any differences after that can be attributed to the camera themselves.
 
Plus you'd need two of each camera to make sure there isn't a variance between two supposedly identical models, as there are between plenty of other supposedly identical electrical items.


I've done two different types of shoot this week with the FX6.

One Cine-EI with zero NR on and one Rec-709 livestream. Both using a 50 year-old lens. Both were absolutely pin-sharp and I would not give questions over the camera's sharpness another thought.
 
Ben Scott said:
Both using a 50 year-old lens. Both were absolutely pin-sharp and I would not give questions over the camera's sharpness another thought.
Click to expand...

I think some of the thought that might need to be given is if the camera is too sharp.. or rather the pre chip filtration is not strong enough. So the FX9 might have more of a tendency to moiree which is bad.. too sharp. One way of blurring the image before the sensor is to use a 50 year old lens.
With the FS7 I was aware of moiree with modern canon cine glass and not with my 50 yo lenses (well 40) .. those older lenses giving a clear advantage over 'modern/sharp' glass which tended to moiree and display purple edges.


===

On designing anti moiree filters it is a compromised situation on cameras that offer '4k' and HD as good 4k filtration will over blur HD aquisition. This problem was properly solved in the f55 with two filters interchangable at a price of time and money
 
Doug Jensen said:
I'll bet you can't name any serious camera shootout that ever used two models of the same camera.
Click to expand...

Of course not. It was an observation on how if you are going to be anal to the nth degree about a controlled experiment you need to see if there is variance between any two of the same model before checking it against another model. There is no end to the rabbit hole one can go down.

Certainly guitar effects pedals - even digital ones - have varied from pedal to pedal in my experience.


Or you can just have a quick check and decide if it's good enough for what you need it to do and not sweat the small stuff and just go shoot stuff.

EDIT: Certainly I have had 2 x FS7s with the same WB and lenses not matching.
 
Last edited:
Ben Scott said:
Of course not. It was an observation on how if you are going to be anal to the nth degree about a controlled experiment you need to see if there is variance between any two of the same model before checking it against another model. There is no end to the rabbit hole one can go down.
Click to expand...

Yeah, you're right, why eliminate some really simple variables when doing testing? If you want to be an ass, go ahead.

Exactly which of these variables do you think can be ignored and still produce a valid comparison between cameras?
1) Lens
2) Codec
3) Framing/shot
4) S-LOG3
5) Exposure
6) White balance
 
Last edited:
Doug Jensen said:
Yeah, you're right, why eliminate some really simple variables when doing testing? If you want to be an ass, go ahead.

Exactly which of these variables do you think can be ignored and still produce a valid comparison between cameras?
1) Lens
2) Codec
3) Framing/shot
4) S-LOG3
5) Exposure
6) White balance
Click to expand...


You're having an argument with someone that isn't arguing with you or disagreeing with you. But if you want to be an ass, go ahead.
 
You must log in or register to reply here.
Back
Top