FX3/FX30: FX3 banding or somethng else?

B

basil_555

Member
Hello,

When I shoot in low light with my Sony FX3 I get these rippling lines moving from bottom to top of the screen which are present in the actual footage. The problem goes away when I move to a well lit area. What is it and how to get rid of it? I shoot in S35 mode if that makes a difference. Thank you!
 

Attachments

  • wavy2.jpg
    wavy2.jpg
    588.1 KB · Views: 12
Sounds like a mismatch between the camera's shutter speed and the frequency of the lights. When you move to a different area, the frequency of those lights is different, so the problem goes away.
 
Definitely looks like it but it also looks like (hard to tell) the shutter angle is 180 and the framerate is 24 or 25, I think, which wouldn't produce such a heavy result, or shouldn't.
 
If the camera was set to 120p, a 180° angle would be 1/240, what may that do to some lights? (vs. 1/48, 1/50)
 
Sure, but as I said, the frame rate is totally irrelevant. Even in your example, it is the shutter speed that matters, not the frame rate.
Let's say the camera was set to 24p, or 30p, or 60p and the shutter speed was set to 1/240 -- the effect would be exactly the same.
 
DJ, but how come you say it's irrelevant if it's the one specification in the camera that is directly connected to the shutter if it's an angle.

If the framerate was 120, it would be the most relevant and important number here because 120 would be forcing a fixed and higher and unnatural shutter speed for lights.

But it doesn't look like it so it's moot. That is bad banding though; I don't think it's possible to get it like that at 24/25 and 180° but who knows.

[Sometimes it's also easy to think FPN but I've never seen it horizontally, at least not that much.]
 
I have something strange going on. The banding was present with Sony 16-50 lens in S35 mode. But when I used Sony 18-200 PZ lens the banding was gone. On the other hand, Sony 16-50 doesn't show any banding on my FX30. And it all happens not only in artificial, but in natural light too.
 
Last edited:
NorBro said:
DJ, but how come you say it's irrelevant if it's the one specification in the camera that is directly connected to the shutter if it's an angle.

If the framerate was 120, it would be the most relevant and important number here because 120 would be forcing a fixed and higher and unnatural shutter speed for lights.

But it doesn't look like it so it's moot. That is bad banding though; I don't think it's possible to get it like that at 24/25 and 180° but who knows.

[Sometimes it's also easy to think FPN but I've never seen it horizontally, at least not that much.]
Click to expand...
Shutter angle isn't shutter speed. They are two different things.

The unwanted bands in the video are caused by a mismatch between shutter speed and frequency of the lights. Shutter Angle does not come into play. I think you are confused because choosing a certain combination of shutter angle and frame rate will result in a certain shutter speed --- but shutter angle is not the same as shutter speed.

Think about this: 1/60th is always 1/60th regardless of any other camera settings. 1/60th is always 1/60th of a second for 24p. 60p, 120p, etc.
But if I tell you my shutter angle is 180 degrees, can you tell me what my shutter speed is? No, you can't answer that questions unless you also know the camera's frame rate. When you put the shutter angle and frame-rate together, then you can deduce what the shutter speed is, but on it's own, shutter angle tells you nothing about shutter speed.

So shutter angle is not the cause of the OP's problem. The problem is a mismatch between shutter speed and frequency of the lights
 
He is using a shutter angle on the camera.

Your semantics-approach is fine, but in this particular case we can adjust our problem-solving skills based on what we see.

We don't see the measurement for shutter speed in fractions of a second. And we're not making up a scenario in our heads where the camera is showing us 1/60 so we can defend why framerate would then be irrelevant.

Here, we can focus on 180°...becase he is using a shutter angle on the camera.

So now we can deduce SS for any framerate, and the primary objective would be to narrow down the framerate to pinpoint how great the frequency mismatch that you mention may be.

If the camera showed 1/120 or something higher in the bottom-left, this is a different conversation (framerate would not have been mentioned).
 
basil_555 said:
I have something strange going on. The banding was present with Sony 16-50 lens in S35 mode. But when I used Sony 18-200 PZ lens the banding was gone. On the other hand, Sony 16-50 doesn't show any banding on my FX30. And it all happens not only in artificial, but in natural light too.
Click to expand...
That doesn't sound normal...

I know it's dual but your ISO is high in that screenshot, when do you see those lines? All ISOs?
 
NorBro said:
He is using a shutter angle on the camera.

Your semantics-approach is fine, but in this particular case we can adjust our problem-solving skills based on what we see.

We don't see the measurement for shutter speed in fractions of a second. And we're not making up a scenario in our heads where the camera is showing us 1/60 so we can defend why framerate would then be irrelevant.

Here, we can focus on 180°...becase he is using a shutter angle on the camera.

So now we can deduce SS for any framerate, and the primary objective would be to narrow down the framerate to pinpoint how great the frequency mismatch that you mention may be.

If the camera showed 1/120 or something higher in the bottom-left, this is a different conversation (framerate would not have been mentioned).
Click to expand...

I've explained it to the best of my ability. If you don't understand the explanation, or if you want to continue to believe that the problem is something that it is not, that is up to you.
 
Do you have a lens cap on in that screenshot above?

[Maybe would also be good to upload a clip if you'd like further evaluation from anyone.]
 
Yeah, that's extreme, IDK...and you're getting chroma noise, too.

Maybe the sensor is not handling low-light as it should if you say it goes away in more light (or it's just really bad at it in S35).
 
NorBro said:
Yeah, that's extreme, IDK...and you're getting chroma noise, too.

Maybe the sensor is not handling low-light as it should if you say it goes away in more light (or it's just really bad at it in S35).
Click to expand...
But why does it happen with only that particular lens?
 
The only thing I can think of right now is the FX3 being a full-frame camera doesn't like that S35 lens on it in its S35 mode but that really makes no sense, would be an anomaly.
 
Trouble shooting is a process of controlled testing and elimination of variables until you narrow down the cause. You're not doing that.

Why don't you take the time to: 1) Actually put the camera on a tripod and shoot something besides junk. 2) Turn on full manual exposure control. 3) Light your test scene (or charts) with controlled and consistent professional lighting instruments. THEN maybe you can do some meaningful testing and get some answers. There are too many red flags in the metadata of your clip. It looks like you're shooting S-LOG3 with an 8-bit codec and then cranking the gain up to +26dB. Why? Nothing will ever look good under those conditions, and any problems will just be amplified. When you change lenes, are you being careful to maintain the exact same f-stop and exposure? Have you tried changing the shutter speed by changing the shutter angle and seeing if that affects the banding? Have you tried different lighting instrments, including natural daylight with no man-made light at all.
Where are you in the world, 50hz or 60hz?

There's a dozen other questions that need to be looked at, but at least you can start there.
 
You must log in or register to reply here.
Back
Top