T54PRO-ZE versus ... another ... T54PRO-ZE ... ;-)

A

alekk

Getting comfortable
Aug 13, 2018
358
533
Republic of Boulder
Why would there be anything interesting about comparing a T54PRO-ZE versus ... another ... T54PRO-ZE?!?

Well ... it might provide some insights into unit variance ... albeit it's only two cameras. Plus when I had 'em mounted several feet apart (vertically), I thought I was some softness in the periphery ... so now we can do some serious pixel peeping! ;-)

FYI FWIW is that the first one came directly from China and arrived on 10/3/2025 with the second (logo in the bottom left) ordered on Amazon and arriving on 10/18/2025 - both ordered from @EMPIRETECANDY who (as typical) handled things well and I would recommend. Looking at the serial numbers (that are in Hex), the "newer" one is higher by "17D70" ... which is 97,649. But the difference in the MAC address is only "3F" which is 63 ... and the latest camera being LESS ... so who knows what that means?!? ;-)

Both running 0827 Firmware from Dahua with identical (fairly generic) settings.

One thing I noticed earlier that still exists here (as I suspect it would) it that the both camera's are set to zoom=550 and have very similar field of view (see GIF below comparing them) ... but the focus numbers are about 1645 versus 1745. I was kinda surprised to see that much difference ... albeit I don't think that's a big deal.

I'll have this up for another day or two ... so if anyone has any suggestions for "interesting" things to look at, let me know.

2025_10_20_TWIN_T54PRO-ZE.jpg



2025_10_20_1716_Two_Cams.gif
 
Last edited:
  • Like
  • Wow
Reactions: Clark Griswald, JDreaming and Arjun
Unfortunately, I don't have a managed switch that would show me power draw ... but that's kinda funky one was a few watts more ... because they are only pulling about 5 Watts (no lights, etc.) to begin with.

Yea, I'm not expecting any "earth-shattering" insights ... but figured I'd toss it out there is anyone has any clever tests they want run.

Picture below is about an hour before sunset ... lots of dynamic range ... but ... the T54PRO's do some shadow boosting, so those aren't as dark as from the 5442.

Settings are 10msec shutter (1/100s) with 100 Iris and ZERO gain and ZERO 2D & 3D NR - gets dim if I go down to 1msec. The metering looks pretty identical and general image qualify looks similar ... albeit it's pretty windy, so I'll have to do my pixel peeping tomorrow for sharpness when it gets calm.


2025_10_20_1716_Two_Cams.gif
 
Last edited:
  • Like
  • Wow
Reactions: Clark Griswald, JDreaming and Arjun
alekk said:
Unfortunately, I don't have a managed switch that would show me power draw ... but that's kinda funky one was a few watts more ... because they are only pulling about 5 Watts (no lights, etc.) to begin with.

Yea, I'm not expecting any "earth-shattering" insights ... but figured I'd toss it out there is anyone has any clever tests they want run.

Picture below is about an hour before sunset ... lots of dynamic range ... but ... the T54PRO's do some shadow boosting, so those aren't as dark as from the 5442.

Settings are 10msec shutter (1/100s) with 100 Iris and ZERO gain and ZERO 2D & 3D NR - gets dim if I go down to 1msec. The metering looks pretty identical and general image qualify looks similar ... albeit it's pretty windy, so I'll have to do my pixel peeping tomorrow for sharpness when it gets calm.


View attachment 230604
Click to expand...

Look same to me too

I’m running 0-3 ms day on the AS 3.6
 
  • Like
Reactions: JDreaming
No wind this morning ... and not cloudy ... so enough light to run at 1msec (full iris, ZERO gain) and good that the two images looks pretty identical.

There's quite a bit of dynamic range between the bright sun and shadows ... the camera is doing a decent job IMHO.

There a separate/common issue that I think some NR is being applied (despite being set to ZERO) that results in the street/driveway looking overly smooth ... but @EMPIRETECANDY commented about this recently and sounds like some new firmware may address it.


2025_10_21_0840_1-0-0-0_twoT54.gif
 
  • Like
Reactions: JDreaming, Clark Griswald and bigredfish
Yep - there are dancing pixels ... but in this thread, I'm just focusing on comparing two of the same cameras and seeing how "identical" they are - so far, so good ... which bodes well for Dahua manufacturing/quality control.

Having said that, here's 90 seconds from this morning - shows me getting the newspaper (Mrs. Bigredfish has MUCH better hair!), some neighbors walking by, and then a car.

I should add that the encoding is H264H, AI Coding, 10 FPS, VBR(!), Quality 5, Max Bit Rate 8192 ... so that may contribute to it. Although if I change it to CBR/16384 bit rate, it doesn't change much.

Remember that I am at ZERO gain (iris 100) ... so if I go much faster than 1msec (1/1000s), the exposure starts to drop as I don't have enough photons getting to the sensor. If I go with a longer shutter speed speed and add some gain, there's aren't as many dancing pixels ... but they are still there.

I'll say that having two identical cameras next to each other (that I'm now confident are behaving the same) is REALLY nice when comparing stuff!

 
  • Like
Reactions: JDreaming, Clark Griswald and bigredfish
That IS a very difficult scene (even without Mrs. Bigredfish's hair in it to assess sharpness! ;-) with lots dynamic range, lotsa detail, and focus from near to far.

Having you tried any HLC ... just do ONE ... as it seems to knock down the highlights more than you think it would.

This also seem to result in a different "path" in the image processing software (@steve1225 alluded to something like this) that I find results in a more pleasing image ... at least for my scene.
 
  • Like
Reactions: JDreaming and bigredfish
Good info @bigredfish ... but can I suggest maybe posting this on the "main thread" since I just meant to spin off a discussion about comparing two identical cameras.

Having said all that(!), I tried turning HLC off, but still see some dancing pixels - it's hard to quantify how much of a difference it is ... but certainly something that I'll be looking at and I appreciate your feedback.

BTW, when you say "you lose the shed door" by turning HLC ON (even just ONE), that ideally should NOT be happening ... since it should be knocking down the highlights, so why the shadows?!?

But what I've seen is with HLC OFF, there's a boost in the shadows that is artificial and basically just raises the black baseline - kinda like I discussed in this thread.
 
BTW, since the two camera's are darn close to identical, I'm going to take 'em down in a few hours since I need to paint the wall/camera's and get it permanently mounted ... as my wife is getting back in a few days and she is not as supportive as Mrs. Bigredfish when it comes our camera addiction! ;-)

So last call if anyone wants a quick side-by-side comparison.
 
  • Haha
Reactions: JDreaming, Clark Griswald and StevenFromTexas
alekk said:
Good info @bigredfish ... but can I suggest maybe posting this on the "main thread" since I just meant to spin off a discussion about comparing two identical cameras.

Having said all that(!), I tried turning HLC off, but still see some dancing pixels - it's hard to quantify how much of a difference it is ... but certainly something that I'll be looking at and I appreciate your feedback.

BTW, when you say "you lose the shed door" by turning HLC ON (even just ONE), that ideally should NOT be happening ... since it should be knocking down the highlights, so why the shadows?!?

But what I've seen is with HLC OFF, there's a boost in the shadows that is artificial and basically just raises the black baseline - kinda like I discussed in this thread.
Click to expand...

Sure.

BTW They still look the same
 
  • Like
Reactions: EMPIRETECANDY, JDreaming, Clark Griswald and 1 other person
You must log in or register to reply here.
Top Bottom