Interesting test of IPC-Color 4K-X

[tmxv4128]

Today I swapped out a 2.8mm IPC-Color 4K-X with a 3.6mm IPC-Color 4K-X in the same location. I wanted to see how much difference there was between the two cameras with each having the SAME settings. I assumed the 2.8mm would have better depth of field and the 3.6mm would allow a little more compression of the images. I set up the cameras at the front of my house just behind the fence. There is an intersection at the front of the house that gets a lot of traffic. Last year someone hit a car and knocked it through the fence, and during another incident, a stolen car was left in front of my house with the Highway Patrol in pursuit. My security camera footage from a camera mounted on my house was helpful to the police, but it was not very clear, which is why I decided to install a new camera out towards the street.

I was satisfied with the images of the 2.8mm Color 4K-X during both the day and night. Once dialed in, the colors and clarity were good. Today, when I swapped out the 2.8mm for the 3.6mm I expected the same; however, I was surprised by some differences. My settings for both cameras during the day: shutter 0-1, 0-35, 3D NR 36, 2D NR 36, WDR = 12. For night, I used the settings: 0-12, gain 0-40, 3D NR 39, 2D NR 39. Illuminators "off" but BLC "on."

I expected the 3.6mm to be much darker at night than the 2.8mm (with the same settings), but it was the opposite. Of course I will now go back and change the settings, but I just wanted to see what each camera looked like with similar settings. Also, I noticed there is a lot of lens flare from the street light with the 3.6mm, even though the camera is pointed down away from the light farther than the 2.8mm. I was expecting the opposite. If anything, I expected the 2.8mm to flare more. A few adjustments tomorrow will correct that, but I found it interesting.

Anyway, wondering if others who have used the 2.8mm and 3.6mm cameras have experienced similar results. Both of mine are running the 9/8/21 firmware. Great cameras and quite an upgrade from what I was using previously.

 

[CCTVCam]

That's a lot of difference at night. What are the fstop's comparitively of the 2 cameras?

Also, to my eyes your daytime pic looks a little dark. You're avoiding the sky being blown but at that time of day, I fell a little recoginition is being lost to keep the sky blue.

The shutter looks a bit too fast in the day but that's going to depend on whether you're trying to capture paltes, the typical speed of passing traffic etc.

 

[Flintstone61]

I've had several of Andy's camera's mounted in the exact same spot in the Condo driveway. Even though they were all 4MP turrets. one-3.6, one- 2.8, and one varifocal, I have not been able to use the same settings to achieve their best image, and each cam did a better job at different things.
It almost looks like the the 3.6mm has a smudge on the lens.
Is the camera shade pulled all the way out?
If it is, consider a 1/2" or 1" extension to keep that streetlight from flaring maybe?

 

[TonyR]

Wow....some things never change! I lived in CA for 29 years, left in '04, spent all that time in the roadway working on traffic signals and street lighting. "STOP" printed on a red octagon meant nothing then and I see it still has no real effect to many motorists 18 years later....their brain reads "YIELD" and even that is up for grabs.

 

[sebastiantombs]

I loved the guy walking the dog and glued to his cell phone. Situational awareness is not in his vocabulary at all.

 

[TonyR]

I loved the guy walking the dog and glued to his cell phone. Situational awareness is not in his vocabulary at all.

But he had a headlight on his head!

 

[CCTVCam]

But he had a headlight on his head!

With the car soon to follow?

 

[adamg]

My thoughts:
At daytime, the 3.6mm has like 10% less FOV and everything in the view looks 10% bigger, but I don't really see any difference in quality/fidelity of the image. Things are bigger, tiny details are 10% more visible, but nothing more.

At nighttime, the 3.6mm is making the dark areas much brighter, The effect is similar to a higher "WDR" value. It does make the dark portions of the image more visible, so I would rate it as better. If they are the same model camera except lens focal distance, I would expect that the 2.8mm camera could be configured to have the same dark area brightness.

 

[tmxv4128]

My thoughts:
At daytime, the 3.6mm has like 10% less FOV and everything in the view looks 10% bigger, but I don't really see any difference in quality/fidelity of the image. Things are bigger, tiny details are 10% more visible, but nothing more.

At nighttime, the 3.6mm is making the dark areas much brighter, The effect is similar to a higher "WDR" value. It does make the dark portions of the image more visible, so I would rate it as better. If they are the same model camera except lens focal distance, I would expect that the 2.8mm camera could be configured to have the same dark area brightness.

Excellent comments. I appreciate the feedback. Yes, after experimenting with both cameras for about a week, I switched back to the 2.8mm. There is not much difference in the field of view as you mentioned, but one interesting thing I found is that with LDC turned "on" in both cameras, there is actually less distortion from the curvature of the lens in the 2.8mm. My experience in photography tells me that this should be the opposite, but not in this case. The 2.8mm looks more pleasing to the eye. I was able to make each camera look almost identical with some tweaking of the settings. As Wildcat mentioned in his testing, changing the gamma and exposure compensation settings makes a world of difference when the camera is set to "manual." It is almost like the brains of the camera try to take back control of the settings when it is on "manual." Funniest thing I have ever experienced when setting up a camera. I am going swap out the 3.6mm camera with a 5442 I have in the backyard overlooking my driveway. It should work great out there with a little added light.

 

[wittaj]

You will find many things with these types of cameras are opposite what one sees with photography equipment LOL. Many come here with a photography background and struggle with these cameras.

In this case Lens Distortion Correction (LDC) digitally corrects optical distortion common with ultra-wide angle lenses like the 2.8 and 3.6mm, which depending on your use case may or may be a problem with IDENTIFY due to the way it digitally corrects the image.

 

[Mike A.]

Seems mostly what it does in these cams is to simply reduce the field of view to the center of the area and just doesn't display the areas to the outside where you tend to see the distortion more. It reduces the field of view by probably ~15%. I don't see much actual correction beyond that with mine. For example, if you have something in view that's a horizontal straight line where you can see the curving more, that distortion remains. It just shows you less of it toward of the edges where it does more. ; )

Going back to photography, I'm more used to seeing actual correction as applied by lens filters post-processing where generally you retain the full image and that line would be straightened.