54IR-ZE-S3 vs 58IR-ZE-S3 Night quality

Aug 5, 2019
22
2
USA
I just added a new 58IR-ZE-S3 to replace an older camera and I couldn't help but notice my 54IR looks far superior at night. I wasn't expecting 58IR to be as good as the 54IR at night, but I'm wondering if I have something screwed up on the 58IR. The 58IR looks more blurry at night as you can see in the attached pictures.

Most day/night profile settings are default with the exception of the exposure settings:

Mode: Manual
Shutter: Custom 7-33ms
Gain: 0-50
Exposure Compensation: 20
Auto iris: enabled

I also realize a difference in ambient lighting could be coming into play here.

54IR:

20241124_205033.png

58IR:

10.158.91.251_ch_1_20241124_205023.png
 
Since they are both on the same size sensor, the 4MP will do better than the 8MP at night unless you have stadium quality lighting. It is why we say chase sensor size and not MP.

What are your settings for the 8MP - encoding, bitrate, codec, etc.? Is it the same as the 4MP?

See what the image does if you have:

H264H
12,0000 bitrate
CBR
General Codec.

But 33ms is way too slow to get clean freeze frame captures. In the absence of enough light, the camera will also move towards the slowest shutter speed you give it.
 
The 4MP:

H.265
Encoding strategy: normal
2688x1520
30 FPS
CBR
6144 bitrate
I Frame interval: 60

8PM:

H.264
Encoding strategy: normal
15 FPS
CBR
I Frame interval: 15

I'll try your recommended encoding settings. Should I lower the 33ms to something like 20 maybe? I'll see if that makes a difference.

Thanks!
 
With your pictures not covering the same area it's hard to have an apples-to-apples opinion. A while back I ran their predecessor models, a 5442 and 5842, side-by-side, and was pleasantly surprised that the 5842 wasn't as less-good at night than I had expected.
 
While the settings are extremely important, I think a lot of the perceived softness is due to the angle of the camera and the dull driveway surface, along with the 4MP cam pointing out at bright lights.

I think if you angled it up to test you'd see a big difference.