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

[alekk]

OK ... since you insist (but again, I want to reshoot the footage), here's a comparison of the 5442-S3 and T54-PRO-ZE at 10msec (1/100s) with ZERO Gain and NR.
Note I intentionally underexposed the picture with a fast shutter speed as I'm trying to compare the difference.

BTW, I don't know if the "base ISO" settings of the two cameras are the same, but setting Gain=0 is about the best I can do.

Also, for those wondering what the "light level" is ... and for the camera buffs out there, I pointed my Canon telephoto lens at the rocks/pavement in the lower center and (for CORRECT exposure - brighter than shown here), at F/2.8 and ISO 100, I was getting a 1/2 shutter speed.

Want a faster shutter speed? I'm already at my maximum aperture of F/2.8, so I can only dial-up the ISO/gain ... and you can either do the math (each doubling of ISO is twice as fast shutter) ... which I confirmed as the camera meter'ed at 1/125s at ISO 6400.

Remember that this is with the flood light ON ... if I turned that OFF, then the camera was ONE SECOND exposure at ISO 6400 (which would be 32 seconds at ISO 100) ... and remember that's semi-close to the lights on the house, so it will drop off more the further you get away. I just went out this morning and it's overcast ... at ISO 100, the camera is metering at 1/640s ... which (check my math) is 15 stops difference. Doing a quick WAG, (check my math again), that going from (about) EV12 to EV-3.

Wiki talks about the lux lights level from sunlight (about 100,000/EV15) to nighttime (0.002/EV-7 on a clearly moonless night).
LOL compare this to the specs "0.0006 [email protected] (Color, 30 IRE)" ... but in fairness, it seems like the whole industry does this.
LOL again does that mean at 0.0007 lux, you'll get a image that is just barely above totally black?!? ;-)

 

[CamCrazy]

OK ... since you insist (but again, I want to reshoot the footage), here's a comparison of the 5442-S3 and T54-PRO-ZE at 10msec (1/100s) with ZERO Gain and NR.
Note I intentionally underexposed the picture with a fast shutter speed as I'm trying to compare the difference.

BTW, I don't know if the "base ISO" settings of the two cameras are the same, but setting Gain=0 is about the best I can do.

Also, for those wondering what the "light level" is ... and for the camera buffs out there, I pointed my Canon telephoto lens at the rocks/pavement in the lower center and (for CORRECT exposure - brighter than shown here), at F/2.8 and ISO 100, I was getting a 1/2 shutter speed.

Want a faster shutter speed ... then dial-up the ISO ... and you can either do the math (each doubling of ISO is twice as fast shutter) ... which I confirmed as the camera meter'ed at 1/125s at ISO 6400.

Remember that this is with the flood light ON ... if I turned that OFF, then the camera was ONE SECOND exposure at ISO 6400 (which would be 32 seconds at ISO 100) ... and remember that's semi-close to the lights on the house, so it will drop off more the further you get away. I just went out this morning and it's overcast ... at ISO 100, the camera is metering at 1/640s ... which (check my math) is 15 stops difference. Doing a quick WAG, (check my math again), that going from (about) EV12 to EV-3.

Wiki talks about the lux lights level from sunlight (about 100,000/EV15) to nighttime (0.002/EV-7 on a clearly moonless night).

View attachment 229809


View attachment 229810

Second image looks brighter at the extreme back, more processed somehow though and less detail on the t-shirt.

 

[alekk]

Second image looks brighter at the extreme back, more processed somehow though and less detail on the t-shirt.

Yea, the "less detail on the shirt" is not the first time I've seen the T54PRO-ZE show less sharpness than the 5442-S3 ... and I see some of this during the daytime too.

So I'm not sure to what to make of this - I think @steve1225 said that the sharpness algorithm/results of the 5442-S3 are better.

I actually have a 2nd T54PRO-ZE on order to compare to see if maybe something is slightly off on the copy I got.

 

[bigredfish]

Yea, the "less detail on the shirt" is not the first time I've seen the T54PRO-ZE show less sharpness than the 5442-S3 ... and I see some of this during the daytime too.

So I'm not sure to what to make of this - I think @steve1225 said that the sharpness algorithm/results of the 5442-S3 are better.

I actually have a 2nd T54PRO-ZE on order to compare to see if maybe something is slightly off on the copy I got.

I think its the nature of the beast. I'll bet if you cut that test difference in 1/2 from the cameras we'd see facial detail differences. I'm guessing you're what 20-25ft?

 

[CamCrazy]

Yea, the "less detail on the shirt" is not the first time I've seen the T54PRO-ZE show less sharpness than the 5442-S3 ... and I see some of this during the daytime too.

So I'm not sure to what to make of this - I think @steve1225 said that the sharpness algorithm/results of the 5442-S3 are better.

I actually have a 2nd T54PRO-ZE on order to compare to see if maybe something is slightly off on the copy I got.

Well I hope the 54Pro is not too bad! and guess it is very early days yet, surely it can surpass the S3 or at least match it. I've got a good selection coming my way now to test including 4K-T and 5442-S3 bullet, along with the 54Pro ZE, see if there is any noticeable sample variation. Then wish I'd just bought a bunch of S3's

I find all of this very interesting, do you guys try bumping up the bit rate to see the results, similarly frame rate potentially, I have seen on my older 5442's that this can show a noticeable difference. I mean bit rate above 8000kb, in fact more like 10000-12000kb. This may depend on speed of people or vehicles re the frame rate but this is just my opinion on final results. I have some which look side on to a roadway and this gives good chance for bleeding edge testing if you want to call it that, or worse case should I say!

Looking forward to testing these new cameras out, preparing myself for realising one or two will probably shine more - still we can hope the 54Pro series will mature like the 5442 did, personally I am mostly excited about that and also trying the 4K-T

 

[tmxv4128]

Well I hope the 54Pro is not too bad! and guess it is very early days yet, surely it can surpass the S3 or at least match it. I've got a good selection coming my way now to test including 4K-T and 5442-S3 bullet, along with the 54Pro ZE, see if there is any noticeable sample variation. Then wish I'd just bought a bunch of S3's

I find all of this very interesting, do you guys try bumping up the bit rate to see the results, similarly frame rate potentially, I have seen on my older 5442's that this can show a noticeable difference. I mean bit rate above 8000kb, in fact more like 10000-12000kb. This may depend on speed of people or vehicles re the frame rate but this is just my opinion on final results. I have some which look side on to a roadway and this gives good chance for bleeding edge testing if you want to call it that, or worse case should I say!

Looking forward to testing these new cameras out, preparing myself for realising one or two will probably shine more - still we can hope the 54Pro series will mature like the 5442 did, personally I am mostly excited about that and also trying the 4K-T

Interesting that you mention the bit rate. When I first installed the 54PRO, I wanted to see if there was a noticeable difference when switching between compression and bit rates. I always use H.264.H and about 12288 for the bit rate. The 54PRO is the first camera that I ever seen the huge blocks of alternating compression in the H.265 setting, no matter the bit rate. Generally, I only saw that in my other cameras if I were to zoom in on something, like a street sign. Switching over to H.264.H cleaned that up immediately.

 

[Chokolinho]

How can I fix that ? 54pro Ze

Zoomed in 2000.

Shutter : 7 , Gain : 50 , NR : 40

 

[bigredfish]

Hard to say what we’re looking at. Show the full uncropped image

 

[Chokolinho]

Hard to say what we’re looking at. Show the full uncropped image

 

[bigredfish]

Horrible IR reflection

Gotta reposition that camera. Try first to aim/focus is so there is no ceiling in the view, point it down more.

 

[Chokolinho]

Horrible IR reflection

Gotta reposition that camera. Try first to aim/focus is so there is no ceiling in the view, point it down more.

The problem is that I need to be able to see the door.

 

[bigredfish]

How can I fix that ? 54pro Ze

Zoomed in 2000.

Shutter : 7 , Gain : 50 , NR : 40

Ok so that second pic is better. I’d look at adding an IR Illuminator that can help light that far corner

 

[Chokolinho]

Ok so that second pic is better. I’d look at adding an IR Illuminator that dan help light that far corner

Is zoomed in 2000. I'll try to lower them further tomorrow and zoom out more.

 

[bigredfish]

The problem is that I need to be able to see the door.

That roll up door?

Hmmm can you move the camera 5 ft to the left ?

 

[Chokolinho]

That roll up door?

Hmmm can you move the camera 5 ft to the left ?

The garage door in red. I'll have to take a look tomorrow.

 

[CamCrazy]

Interesting that you mention the bit rate. When I first installed the 54PRO, I wanted to see if there was a noticeable difference when switching between compression and bit rates. I always use H.264.H and about 12288 for the bit rate. The 54PRO is the first camera that I ever seen the huge blocks of alternating compression in the H.265 setting, no matter the bit rate. Generally, I only saw that in my other cameras if I were to zoom in on something, like a street sign. Switching over to H.264.H cleaned that up immediately.

Is running H264 an option in that camera or only H264H? normally I find H264 seems to give more consistent results, again, just my findings. Seems that for my 5442's 10000-12000kb is a good place to be.

 

[tmxv4128]

Is running H264 an option in that camera or only H264H? normally I find H264 seems to give more consistent results, again, just my findings. Seems that for my 5442's 10000-12000kb is a good place to be.

Yes, you can run H.264. I have found H.264.H to be better and tested it by zooming into a speed limit sign on my street. In one of bigredfish' posts he said, "Use h.264.h for the best quality and least problems," so I have kept my settings at that since then.

 

[bigredfish]

H.264h or High profile is the best quality, thats why I prefer it.

H.264 profiles and levels | Inside & Outside MediaCoder

The H.264 Baseline, Main, and High profiles differ primarily in their compression capabilities and quality:

• Baseline Profile: This is the simplest and most efficient profile, designed for low-cost applications like videoconferencing and mobile devices. It uses basic compression techniques and can achieve a compression ratio of about 1000:1.
• Main Profile: This profile includes all the features of the Baseline profile but with improvements in frame prediction algorithms, making it suitable for SD digital TV broadcasts. It can achieve a compression ratio of about 2000:1
• .High Profile: This is the most advanced profile, primarily used for high-definition broadcasting and storage. It offers superior video quality and can achieve a compression ratio of about 2000:1, making it suitable for HD broadcasts and Blu-ray disc storage.
  

 

[CCTVCam]

That is a very nice image for 4ms shutter, the obvious question, how does this look with the onboard lights, is that what you mean by flood light I am assuming, hence triggered by the IVS, more pictures please would be fantastic

The two locations I am planning these for they can have the onboard twin LED's running all night so that gets me out of trouble, they both will have flood assistance around 4000 lumens and one has two 300 lumen bulbs which could be turned on if necessary. Couple of other cameras I plan to use the 54Pro models and only have LED's running when IVS gets triggered hopefully from the IR image.

No not internal LEDS. A separate LED floodlight with a separate PIR sensor. (I like to keep sensor and light separate so I can change just one when something fails). I'm not keen on internal LEDS as they're not as bright as a separate light and once they go dim over time, it's new camera time (or external flood). I prefer just to go internal flood in the 1st place. With the new 54Pro you can of course reduce light wear by having them only switch on when triggered. The catch 22 here though, is there needs to be enough light for the camera to see the trip wires without the LED's on otherwise it won't trigger. The answer there is a background light or run IR.

This is how my lighting is set up. Note I replaced the 15w and 20w flood lights with a single 40w flood light where the 15w used to be. The camera is to the right of the dusk to dawn light and the dusk to dawn light shade now has some black vinyl added on the camera side to reduce glare (not that it was a problem, I just wished to improve image quality). The dusk to dawn sensor is mounted higher than the floodlights so the sensor part is outside the light spill and remains in a dark zone.

The main PIR sensor is positioned out of reach and is almost impossible to avoid as it points downwarsd thus casting detection rings not only left right but front to back (what would be veritcal if it was mounted more horizontal). You can't even reach this area in addition without going past my other camera. It is also mounted in a dark zone.

 

[CCTVCam]

Horrible IR reflection

Gotta reposition that camera. Try first to aim/focus is so there is no ceiling in the view, point it down

I agree but I'd also consider lowering it's position. It looks a bit high (around 8-9 feet?) and mounting it lower will reduce the ceiling reflection as well as giving a better facial shot.