Dahua WizColor 5x59-PRO and SmartLight 5x59-IL new series

So you will be posting videos only after full testing? No little clips before then?

Due to sensitive location of these cameras I will likely not be posting samples, this will not go down well as understandably people like to see samples! myself included. But, I am carrying out thorough testing and evaluation, my findings will be logged on here for others. My testing is fastidious, of that you can be assured, I am also testing this camera in zero light conditions, no ambient at all. Am currently running some IR tests, will then switch back to full colour for further testing. Currently I can say that a common theme is repeating, that is to say freezing motion with this camera is quite a challenge, more so in poor light or total darkness. Daytime I have managed to nail this down but requires shutter of 0-1ms, possibly this could be set to 0-2ms or even 0-3ms in better lighting. Weather here at the moment is overcast around 13000lux with some sunshine spells but not many! Gain is also better kept low with regards motion, it does not respond well to NR being pushed up either, this has been noted by others already, during daylight hours I have had NR at close to zero, literally around 4 for both. It should be further noted that vehicles are crossing this camera scene at approximately 45 degrees, between 30-40mph, just for reference, this is quite challenging for stopping motion I realise, but nonetheless a good test. My 5442 bullet can manage this task without issues, day and night.

For reference, my take on freezing motion is for moving objects to look equally as sharp as the static ones in the scene.

It is very early days with this new model range, so expect plenty of fine tuning to come via firmware.
 
Last edited:
Due to sensitive location of these cameras I will likely not be posting samples, this will not go down well as understandably people like to see samples! myself included. But, I am carrying out thorough testing and evaluation, my findings will be logged on here for others. My testing is fastidious, of that you can be assured, I am also testing this camera in zero light conditions, no ambient at all. Am currently running some IR tests, will then switch back to full colour for further testing. Currently I can say that a common theme is repeating, that is to say freezing motion with this camera is quite a challenge, more so in poor light or total darkness. Daytime I have managed to nail this down but requires shutter of 0-1ms, possibly this could be set to 0-2ms or even 0-3ms in better lighting. Weather here at the moment is overcast around 13000lux with some sunshine spells but not many! Gain is also better kept low with regards motion, it does not respond well to NR being pushed up either, this has been noted by others already, during daylight hours I have had NR at close to zero, literally around 4 for both. It should be further noted that vehicles are crossing this camera scene at approximately 45 degrees, between 30-40mph, just for reference, this is quite challenging for stopping motion I realise, but nonetheless a good test. My 5442 bullet can manage this task without issues, day and night.

For reference, my take on freezing motion is for moving objects to look equally as sharp as the static ones in the scene.

It is very early days with this new model range, so expect plenty of fine tuning to come via firmware.
Pretty solid No...
 
I'm finding the human detection of the 54PRO to be not-so-hot, very lame compared to the 5442 S2 that it's mounted next to. Here are 2 examples of somebody walking right through a tripwire with no trigger (both were caught by the 5442):

Capture.JPG Capture1.JPG

Yet it did a good detection at the very edge of the FOV that I might have forgiven it not catching.


Capture2.JPG

The target filter settings:
Capture3.JPG

I'm using the 7-31 firmware that came with the camera. So far, pretty disappointing. I'm open to the cause being operator error in the setup.
 
I'm finding the human detection of the 54PRO to be not-so-hot, very lame compared to the 5442 S2 that it's mounted next to. Here are 2 examples of somebody walking right through a tripwire with no trigger (both were caught by the 5442):

View attachment 230329 View attachment 230330

Yet it did a good detection at the very edge of the FOV that I might have forgiven it not catching.


View attachment 230331

The target filter settings:
View attachment 230332

I'm using the 7-31 firmware that came with the camera. So far, pretty disappointing. I'm open to the cause being operator error in the setup.
those are the largest thumbnails i have ever seen.
 
  • Like
Reactions: duplo
I'm finding the human detection of the 54PRO to be not-so-hot, very lame compared to the 5442 S2 that it's mounted next to. Here are 2 examples of somebody walking right through a tripwire with no trigger (both were caught by the 5442):

View attachment 230329 View attachment 230330

Yet it did a good detection at the very edge of the FOV that I might have forgiven it not catching.


View attachment 230331

The target filter settings:
View attachment 230332

I'm using the 7-31 firmware that came with the camera. So far, pretty disappointing. I'm open to the cause being operator error in the setup.
did you or can you enable the large-scale AI models ? Mine was greyed out on the 7/31 fw....had to go to the 8/27 fw. It works excellent on mine.
 
those are the largest thumbnails i have ever seen.
They're not thumbnails, rather screenshots off of smartPSS. Can't get the trigger lines displayed on BI.
did you or can you enable the large-scale AI models ?
I doubt it, since I can't find where it's done, grayed-out or not. Where is this setting found? (I admit to being a master of not seeing what's right in front of me).
 
Pretty solid No...

I will reserve full judgement until I have done some further testing, also keep in mind a lot of people will be using this in locations with some ambient light, also countries with more sunshine! those two things will make quite a difference in favour of the 54Pro, particularly with colour at night. Also not everyone will be trying to stop motion on vehicles travelling 25-35mph, again this will change final results. In those cases from what I've seen then this is looking to have good potential. It can be a bit Jekyll and Hyde, one minute it amazes me with image quality in very low light, similarly at night with IR, another I wonder how my 2MP starlight or 5442 produced a better output during daylight hours. For me at the moment freezing motion is the biggest hurdle, not a problem I have really encountered before, hoping this is a processing modification which can be made by the tech boffins back at Dahua. I also would not be the least surprised to see the bullet version performing slightly better, normally I pick bullet cams if I can, sometimes the location lends itself to the more subtle looking turret.

Yes, I am picky when it comes to image quality, I tend not to settle for second best if something superior can be achieved by tweaking. So far the biggest difference I see with this camera compared to previous models is the need for higher shutter speed and low noise reduction, the latter seems to help keep ghosting under control, similarly when the light levels begin to get challenging it doesn't react so well to increased gain, in fact, I have been trying to keep it below 45 if possible, currently 30 in daylight and 49 overnight in colour mode. I scheduled a period last night as the daylight faded where IR kicked in, this ran until around 9pm when the LED's switched on. What I am finding is at daybreak when light is low, the image is quite dark in colour mode, this is fallout from the very fast shutter, although image is still quite useable and motion is frozen.

Had about 3-4 hours IR testing last night, this was mostly as expected, shutter still needs to be pretty low for my light levels, this was 0-2 if possible, 0-3 or 0-4ms does give more light but begins to impact image quality too much. Maybe I am expecting too much at this point, but the IR seemed a bit underwhelming. But I am pushing this thing pretty hard, distance to target is 20 feet in pitch darkness, aside from vehicle headlights which of course assist greatly. I might try another mounted lower down where the IR light will have to do less work to reach target subject. The biggest problem is the 5442 bullet cameras are so damn good that anything else has a challenge straight out the gate! but these new cameras have lots of features, dual illumination and moments of brilliance, so I am keeping an open mind still.

Another major factor for many will be placement, if mine was facing oncoming vehicles and was lower down, it would pose much less of a challenge. Normally I am not a fan of high mounted overview cameras, unless they are just being used for that purpose, I tend to set mine just out of reach height wise, so around 10-12 feet, this not only gives a more useable view but also requires lower shutter speed. This one is more like 20 feet up so again, keep that in mind when reading my feedback. As mentioned above, I will likely mount another one lower down at much closer range to give a 'normal use case' scenario, if there is such a thing :)

To finish this post on a positive note, so far with the 54Pro I have been able to identify the colour of items being carried in a truck travelling at 30-35mph in total darkness, switch from white LED to IR mode on a schedule and freeze motion in low light colour before the sun had tried to rise. So I guess that eliminates the 2MP starlight's and it might be tough to pick a 5442 now these are on the market. A shortlist of suggestions for overcast/night conditions below for others maybe to try, more to come...

  • Keep shutter speed high, 2-3ms or less
  • Very low NR if possible, less than 10 ideally, mine is off during the day
  • Keep gain low, below 30 ideally during daylight, mine is 49 overnight and this is already too high
  • Get bitrate as high as you can, allow some headroom compared to fps settings
  • Exposure compensation down quite low, mine is 25-40
  • Iris is virtually 95-100% on mine during testing, not seeing much change with this lower
  • Gamma maybe 40-45, can help with motion and quality
 
Last edited:
  • Like
Reactions: RyanB
I will reserve full judgement until I have done some further testing, also keep in mind a lot of people will be using this in locations with some ambient light, also countries with more sunshine! those two things will make quite a difference in favour of the 54Pro, particularly with colour at night. Also not everyone will be trying to stop motion on vehicles travelling 20mph+, again this will change final results. In those cases from what I've seen then this is looking to have good potential. It can be a bit Jekyll and Hyde, one minute it amazes me with image quality in very low light, similarly at night with IR, another I wonder how my 2MP starlight or 5442 produced a better output during daylight hours. For me at the moment freezing motion is the biggest hurdle, not a problem I have really encountered before, hoping this is a processing modification which can be made by the tech boffins back at Dahua. I also would not be the least surprised to see the bullet version performing slightly better, normally I pick bullet cams if I can, sometimes the location lends itself to the more subtle looking turret.

Yes, I am picky when it comes to image quality, I tend not to settle for second best if something superior can be achieved by tweaking. So far the biggest difference I see with this camera compared to previous models is the need for higher shutter speed and low noise reduction, the latter seems to help keep ghosting under control, similarly when the light levels begin to get challenging it doesn't react so well to increased gain, in fact, I have been trying to keep it below 45 if possible, currently 30 in daylight and 49 overnight in colour mode. I scheduled a period last night as the daylight faded where IR kicked in, this ran until around 9pm when the LED's switched on. What I am finding is at daybreak when light is low, the image is quite dark in colour mode, this is fallout from the very fast shutter, although image is still quite useable and motion is frozen.

Had about 3-4 hours IR testing last night, this was mostly as expected, shutter still needs to be pretty low for my light levels, this was 0-2 if possible, 0-3 or 0-4ms does give more light but begins to impact image quality too much. Maybe I am expecting too much at this point, but the IR seemed a bit underwhelming. But I am pushing this thing pretty hard, distance to target is 20 feet in pitch darkness, aside from vehicle headlights which of course assist greatly. I might try another mounted lower down where the IR light will have to do less work to reach target subject. The biggest problem is the 5442 bullet cameras are so damn good that anything else has a challenge straight out the gate! but these new cameras have lots of features, dual illumination and moments of brilliance, so I am keeping an open mind still.

Another major factor for many will be placement, if mine was facing oncoming vehicles and was lower down, it would face much less of a challenge. Normally I am not a fan of high mounted overview cameras, unless they are just being used for that purpose, I tend to set mine just out of reach height wise, so around 10-12 feet, this not only gives a more useable view but also requires lower shutter speed. This one is more like 20 feet up so again, keep that in mind when reading my feedback. As mentioned above, I will likely mount another one lower down at much closer range to give a 'normal use case' scenario, if there is such a thing :)

To finish this post on a positive note, so far with the 54Pro I have been able to identify the colour of items being carried in a truck travelling at 30-35mph in total darkness, switch from white LED to IR mode on a schedule and freeze motion in low light colour before the sun had tried to rise. So I guess that eliminates the 2MP starlight's and it might be tough to pick a 5442 now these are on the market. More to come...

You know what is the best?
This problem is simply wrong automatic exposure algorithm, which select longer exposures in good conditions.. Can be fixed in firmware..

Worth to collect video documentation (video files showing problem + text description) and sends to Andy.. He can sends this to Dahua engineers..
 
You know what is the best?
This problem is simply wrong automatic exposure algorithm, which select longer exposures in good conditions.. Can be fixed in firmware..

Worth to collect video documentation (video files showing problem + text description) and sends to Andy.. He can sends this to Dahua engineers..

I am in touch with Andy, you can be assured :thumb:
 
I think @steve1225 is (yet again!) onto something there ... as I feel the exposure is "hotter" than the 5442-S3.

As @CamCrazy wrote (good stuff and I appreciate @EMPIRETECANDY providing our feedback to the Dahua Engineers), I also have seen that dropping the gamma (to 35-40'ish even) seems to help with motion and quality. But part of this is the overall exposure is dropped, so shutter speed should be decreased.

Also, I'm still trying to figure out what appears to be "Less sharpness in T54PRO-ZE (versus 5442-S3) possibly due to noise reduction being applied even if set to ZERO?"

I have a 2nd T54PRO-ZE arriving today which I'm going to test side-by-side to see if any significant difference in the units - would be surprised if there was.
 
Last edited:
I know this clip is not the perfect quality as I had my warning light schedule set a little too early for the cat visit. But the expression on the cat's face when he looks right up at camera....like wtf.
Installed a new motion light this morning on porch and more camera tweaking tonight :)

 
I think @steve1225 is (yet again!) onto something there ... as I feel the exposure is "hotter" than the 5442-S3.

As @CamCrazy wrote (good stuff and I appreciate @EMPIRETECANDY providing our feedback to the Dahua Engineers), I also have seen that dropping the gamma (to 35-40'ish even) seems to help with motion and quality. But part of this is the overall exposure is dropped, so shutter speed should be decreased.

Also, I'm still trying to figure out what appears to be "Less sharpness in T54PRO-ZE (versus 5442-S3) possibly due to noise reduction being applied even if set to ZERO?"

I have a 2nd T54PRO-ZE arriving today which I'm going to test side-by-side to see if any significant difference in the units - would be surprised if there was.
Lowering gamma day and night?
 
I know this clip is not the perfect quality as I had my warning light schedule set a little too early for the cat visit. But the expression on the cat's face when he looks right up at camera....like wtf.
Installed a new motion light this morning on porch and more camera tweaking tonight :)


Shutter speed ?
 
I know this clip is not the perfect quality as I had my warning light schedule set a little too early for the cat visit. But the expression on the cat's face when he looks right up at camera....like wtf.
Installed a new motion light this morning on porch and more camera tweaking tonight :)



Push that shutter speed up and that fur ball will look less like a fur ball :thumb:
 
I think @steve1225 is (yet again!) onto something there ... as I feel the exposure is "hotter" than the 5442-S3.

As @CamCrazy wrote (good stuff and I appreciate @EMPIRETECANDY providing our feedback to the Dahua Engineers), I also have seen that dropping the gamma (to 35-40'ish even) seems to help with motion and quality. But part of this is the overall exposure is dropped, so shutter speed should be decreased.

Also, I'm still trying to figure out what appears to be "Less sharpness in T54PRO-ZE (versus 5442-S3) possibly due to noise reduction being applied even if set to ZERO?"

I have a 2nd T54PRO-ZE arriving today which I'm going to test side-by-side to see if any significant difference in the units - would be surprised if there was.

Motion sharpness? if so, put your shutter as fast as it will go and report back... 0-1ms or even 0-0.5ms but certainly 0-2ms to test, if not tried already. I am seeing similar, same settings on the Pro will not mirror results from the 5442 series, it needs faster shutter.