F65: Is this how you would test for noise?

B

bitnaut

Active member
I used f/5.6 at EI 640 as my base/proper exposure then varied the aperture to get +1, +2, -1 and -2 stops of exposure then matched them all to the base exposure in Da Vinci Resolve 16.

Is this a valid way of testing for noise? Asking for a friend. :)

https://vimeo.com/347433980
 
When I test for noise, I try to shoot something as close to what I want to shoot as possible. That is the main way I can tell if the noise is too much, or if I can push it further.

I might shoot a chart for analysis.
 
Ed,

What are you trying to understand?

Noise level at different iso’s? The test you performed is valid, although most people would change the shutter angle to change exposure and leave the lens at the same aperture. But it does give you an idea. Even better, you could use a meter to shoot the chart, and use that as your base for matching. Make sure that you are getting your slog (1,2,3?)curve to best fit your middle grey. in the chart.

If you are trying to understand how much noise exposure change can bring to the image, don’t compensate with your iris, or ND, or ISO. Just vary exposure with let’s say shutter angle and then just open the files up in Resolve and compare noise levels across at different ire levels. I personally think this is a more important test. With the other test... iso changes affect noise, but they also shift your middle gray making it harder to make decisions. And who cares about noise if you ya e to get the shot and you don’t have the lighting, you are going to bump it up anyway. Tests look very clean, though.
 
bitnaut said:
What sort of chart?
Click to expand...

Depends on my goals.

A standard color chart is helpful. But they can cost $200 more or less. Sometimes a lot more.

But if you are just testing for noise, find a dark scene you know you like the look of. Don’t just shoot something dark, because if you know it is too dark, then why shoot it? So find things you like and if the camera ever starts getting in the way of what you like then you have two choices:

- adapt to the camera and figure out how to make what you see work in the camera.

- get a different camera that can do what you want.

And thirdly, if those two things don’t work, then join me in spurring on better sensors! We need better sensors!
 
The times i use charts is when i want to be able to replicate results. They help in getting more definitive numbers, and the test results are generally repeatable and more easily understood after longer periods of time. But they are mot intuitive and not specific to your work or voice. They are more quantatice than qualitive??? Everyone figures their own learning curve.
 
bitnaut said:
I was just trying to see what my F65 is capable of qualitatively.
Click to expand...

In that case, i just go shoot things i am going to shoot more often. See how far the camera will follow my eye. Eventually the camera gives out at some point. I make note of when it starts becoming a nuisance.

Once a camera stops being able to do what i want, then i know i am either asking too much, or the camera is not the right camera for me.

With digital cameras, i just adapt to what they want to do. I know digital looks a certain way If you push it too hard. It falls apart. It isn’t as raw a chemical process as film, so there is nothing that interests my in the limits of a digital camera, it is just a bunch of s*%t at the extremes.

So i know, if it is digital, i will have to learn to see like the camera does.

The only reason i care that digital is s*%tty, is because digital killed film. So, digital has to get better than it is today. Otherwise, what was the use in killing film?

The F65 is still one of the best images you can get. And because it is well designed, you are going to be able pull out more nuance than most any other video camera. Sony cameras tend to be on the neutral side, so it is up to the shooter to put something good in front of it.

Frankly, i wish i had one.
 
Last edited:
You must log in or register to reply here.
Back
Top