Menu
What's new

Mini-Review - 5442-Z4E-S3 AKA B54IR-Z4E S3 - Replacement to the 5442-Z4E

wittaj

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
25,389
Reaction score
49,453
Location
USA
alekk said:
Thanks @wittaj for your expert comments.

Per my earlier post, really impressive that the F/1.8 and 1/1.8" sensor Z4E camera has comparable (visible) low light capability as the the IPC-Color4K-T ... which is F/1.0 and 1/1.2" sensor.

I think I know what to ask Santa for ... ;-)
Click to expand...
I think they learned a lot in the firmware for the 4K/T about maximizing low light performance and brought it into this model.
 
Parley

Parley

Known around here
Joined
Dec 19, 2015
Messages
5,673
Reaction score
16,210
Location
Cypress, California
The Automation Guy said:
Sorry, it is quite possible the mode is listed as "manual" and not custom. I am going off memory here (and my memory is terrible)!

Here are a couple of shutter speed decimal equivalents.....
1/2000 = .5 ms or .0005 seconds
1/1000 = 1ms or .001 second
1/500 = 2ms or .002 second
1/333 = 3ms or .003 second
1/250 = 4ms or .004 second

Please note, you can also set the shutter speed as a range from fastest (0 ms) to the desired slowest shutter speed. So entering 0-4ms will allow the cameras to select any shutter speed 1/250th or faster (so 1/500th or 1/1000th, etc). I generally set a range instead of demanding a single shutter speed be used (although honestly it probably doesn't make a difference in real life). The slowest shutter speed will be different for everyone, but likely needs to be 2ms (1/500th) or faster to ensure images have no motion blurring in them. If you have fast traffic you might need to increase or if the traffic is slow (for example I'm at the end of a street in a cul-de-sac, so traffic cannot go fast past my house) you might be able to use a slower shutter speed. I'd recommend you use 0-2ms to start and see how that works.

The gain setting will appear as a range with the max probably at the default "50" (out of 100). That isn't a bad number to use, but you can adjust it higher if you need the brightness to increase because you plates are too dark to read, or lower if the plates are still getting over exposed and you don't want to change the shutter speed. It usually takes some playing around with both the gain and shutter speed values before you land on the "perfect" compromise that allows motionless captures that are bright enough to be useful.
Click to expand...
Do you have a site as to where I can find the expanded shutter speed decimal equivalents? I did a Google search but did not come up with the right chart.
 
Parley

Parley

Known around here
Joined
Dec 19, 2015
Messages
5,673
Reaction score
16,210
Location
Cypress, California
I just had a thought from all of this. The 60 hertz power cycle from the streetlamps cause the light to dim and brighten somewhat. I noticed that if the shutter speed is set to 1/120 it does not do that. So, I am wondering if you were to set the shutter speed of 1000ms/240 equals 4.1666ms if that would also work being it is a multiple of 60. Also, 1000ms/480 equals 2.0833 and 1000ms/960 equals 1.04166ms. I would think that this has been tried before.
 
Parley

Parley

Known around here
Joined
Dec 19, 2015
Messages
5,673
Reaction score
16,210
Location
Cypress, California
wittaj said:
I don't think a firmware will fix it as it is likely a frequency induced phenomenon of the lights.
Click to expand...
That is exactly what it is. The power to the lights is on 60 Cycle and I thought multiples of that would work. By the way I will be ordering two more of these cameras for my LP1 and LP4 positions. Now if the streetlights were LED's this might not happen.
 
T

tigerwillow1

Known around here
Joined
Jul 18, 2016
Messages
3,878
Reaction score
8,603
Location
USA, Oregon
Since the cameras aren't sync'd to the power line, the odds of them being able to exactly math the grid frequency is zilch. My best guess if that you'd see a slow fading in and out.
 
David L

David L

IPCT Contributor
Joined
Aug 2, 2019
Messages
8,188
Reaction score
21,402
Location
USA
Parley said:
That is exactly what it is. The power to the lights is on 60 Cycle and I thought multiples of that would work. By the way I will be ordering two more of these cameras for my LP1 and LP4 positions. Now if the streetlights were LED's this might not happen.
Click to expand...
Add a BB Gun to your order, lights out!!!
 
J

jrbeddow

Getting comfortable
Joined
Oct 26, 2021
Messages
379
Reaction score
490
Location
USA
Parley said:
That is exactly what it is. The power to the lights is on 60 Cycle and I thought multiples of that would work. By the way I will be ordering two more of these cameras for my LP1 and LP4 positions. Now if the streetlights were LED's this might not happen.
Click to expand...
Have you tried various non-multiples of 60hz as well? Intuitively, I would think you would want to avoid potentially syncing with the 60hz frequency, not align with it.
 
wittaj

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
25,389
Reaction score
49,453
Location
USA
tigerwillow1 said:
Since the cameras aren't sync'd to the power line, the odds of them being able to exactly math the grid frequency is zilch. My best guess if that you'd see a slow fading in and out.
Click to expand...
It is absolutely the lights. I have seen lights do the slow fade in and out like the Christmas Thread above. I have seen cars go by on my LPR cam that look like the headlights are off and they aren't when seen with my overview cam. I have seen it look like random off and on like they are flashing their lights.

Generally the cheaper the bulb or the power supply, the more likely a camera with a faster shutter is likely to catch this phenomena that our eyes can't see.

It is why if you select anti-flicker at 60Hz it limits the fastest shutter speed to 1/120 and at 50Hz it limits the shutter to 1/100 - any shutter speed faster and it can catch the weird cycle/frequency/modulation/whatever electrical folks call it.

1701967483389.png

With higher quality bulbs and power supply, I have been able to do odd shutter speeds to minimize that phenomena. Like if I run my LPR at 1/867 (1.15ms) instead of 1/1000, the car that looks like the headlights are off can be seen.

Now if the image is more of a wave going thru the whole image, that is more of the firmware trying to deal with trying to run the camera either with not enough light or a too fast of a shutter. Adding more light or slowing the shutter can eliminate it. Some folks have had success changing the saturation as well to minimize that wave.
 
Parley

Parley

Known around here
Joined
Dec 19, 2015
Messages
5,673
Reaction score
16,210
Location
Cypress, California
jrbeddow said:
Have you tried various non-multiples of 60hz as well? Intuitively, I would think you would want to avoid potentially syncing with the 60hz frequency, not align with it.
Click to expand...
If I am reading what you are saying, then non-multiples of 60hz can be a large number. For reading license plates it is not a factor for me. However, looking at the background it would be nice to eliminate the light wave. Just something I wonder about.
 
wittaj

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
25,389
Reaction score
49,453
Location
USA
Parley said:
If I am reading what you are saying, then non-multiples of 60hz can be a large number. For reading license plates it is not a factor for me. However, looking at the background it would be nice to eliminate the light wave. Just something I wonder about.
Click to expand...
Can you post a video clip? If it is the whole image showing the "wave" as opposed to just the portion nearest the light, that is just a function of running too fast a shutter and not enough light.

This phenomena was first found in the 4K/T thread and for some reason folks were having decent success reducing that wave by dropping saturation.
 
You must log in or register to reply here.
Top