EmpireTech PTZ425DB glitchy background?

TuckNTruck said:
Maybe I'll get out in the cold tomorrow and try out a new ethernet cable.
Click to expand...
I'm inclined to think it's the camera not your network likely hitting some internal resource limit. I just wanted to explore the possibility as intermittent and especially temperature dependent cable problems can be extremely frustrating to troubleshoot.
 
  • Like
Reactions: bigredfish and TuckNTruck
It is bad implementation of h.264/h.265 encoders in Chinese SOC chips (HiSilicon?) used by Dahua/HIK..
Normally this is more a problem on h.265, downgrading to h.264h helps a lot.
on some models this is bigger problem, on some not - depending of SOC (chipset) used in camera and firmware...

if in video there is too much details, video encoder goes crazy - allocate to much bandwidth for I-frame (full frame, that one sharp every 2 seconds) and to little for p/b-frames (differential ones, usually 49/59 frames after full I-frame - which create 2 second video).

With small bandwidth for differential frames & high details in source video encoder decrease internal 'details level' to fit frames in bandwidth.. When internal 'details level' is low, encoder start losing details a lot for big uniform areas like grass, trees, walls, roads, water etc..

Increasing bandwidth and decreasing frame rate should help.
But I found on some cameras that increasing iframe interval to 4 or 8 * frame rate helps...

You can decrease sharpness (or disable WDR at day) to decrease details in video.

Also You can try VBR with highest bandwidth and details level.
 
  • Like
Reactions: guykuo
steve1225 said:
It is bad implementation of h.264/h.265 encoders in Chinese SOC chips (HiSilicon?) used by Dahua/HIK..
Normally this is more a problem on h.265, downgrading to h.264h helps a lot.
on some models this is bigger problem, on some not - depending of SOC (chipset) used in camera and firmware...

if in video there is too much details, video encoder goes crazy - allocate to much bandwidth for I-frame (full frame, that one sharp every 2 seconds) and to little for p/b-frames (differential ones, usually 49/59 frames after full I-frame - which create 2 second video).

With small bandwidth for differential frames & high details in source video encoder decrease internal 'details level' to fit frames in bandwidth.. When internal 'details level' is low, encoder start losing details a lot for big uniform areas like grass, trees, walls, roads, water etc..

Increasing bandwidth and decreasing frame rate should help.
But I found on some cameras that increasing iframe interval to 4 or 8 * frame rate helps...

You can decrease sharpness (or disable WDR at day) to decrease details in video.

Also You can try VBR with highest bandwidth and details level.
Click to expand...
Unfortunately, I think you've nailed it. This morning I tried a new ethernet cable and even a different switch and get the exact same issues - no change whatsoever.

If the water is smooth, I see minimal issues (especially after changing the iframe interval to 4* frame rate - at least with that setting when the background macroblocking is noticeable it only happens every 4 seconds vs every 1 second).

When the water gets choppy, it's like the image is so busy the camera can't handle it and it turns into a blocky mess. I can zoom in on a dock or land and the issue goes away. Watching the camera, it's kind of interesting because I can see the camera process certain sections of the water and I get a perfect image in just that small section with the surrounding water being blocky/pixelated. But as the wind/wave pattern on the water changes, the image processes differently and it renders all the water as blocky almost like viewing the substream, and then moments later another chunk of the water is processed perfectly.

This mini PTZ is the only camera on the dock that does this. In addition to the mini PTZ I have a 5442, SD5A445XA-HNR, IPC-HDW2431T-AS-S2, and IPC-T2431T-AS-3.6mm-S2 running on the dock and none of them have any issues processing images with wind/waves/whatever. I also had a SD49225XA-HNR in this exact mounting location for about a year with no issues from it either.

So I wonder if my camera is defective or if I'm using the camera in a setting that it's just not designed for and this is as good as it gets? I don't understand why the image is processed so terribly on this camera while none of my other Dahua cameras have an issue even though they're also in the same environment?
 
  • Like
Reactions: bigredfish
TuckNTruck said:
This mini PTZ is the only camera on the dock that does this. In addition to the mini PTZ I have a 5442, SD5A445XA-HNR, IPC-HDW2431T-AS-S2, and IPC-T2431T-AS-3.6mm-S2 running on the dock and none of them have any issues processing images with wind/waves/whatever. I also had a SD49225XA-HNR in this exact mounting location for about a year with no issues from it either.

So I wonder if my camera is defective or if I'm using the camera in a setting that it's just not designed for and this is as good as it gets? I don't understand why the image is processed so terribly on this camera while none of my other Dahua cameras have an issue even though they're also in the same environment?
Click to expand...

it is camera model / used chipset / firmware thing. Dahua / hik use different SOC chipsets and image sensors for different lines and models.

for every combination you need to put some work to finish firmware. Some received a lot of work from Dahua engineers (usually main lines like wizmind-s 5xxx or sd5 ptz), other not so much (short / low quantity models created to fill product catalog / compete with hik).

Sometimes they use too weak / old chipset for needs / use of exact camera model. And this cannot be corrected by firmware
 
  • Like
Reactions: TuckNTruck and bigredfish
TuckNTruck said:
Interesting. Mine is different. When I do a manual check, it says I am using the latest version.
System Version:2.821.0000006.1.RBuild Date:2023-03-16
PTZ Version:V2.401.0000001.47.RHNDG_220906_42754

No variation of compression, frame rate, or bitrate has made any real difference, unfortunately.
Click to expand...
I have never seen the manual check work in any camera.

Firmware is here: Download Firmwares
 
  • Like
Reactions: TuckNTruck and bigredfish
^^^^^
Ditto me either
 
  • Like
Reactions: looney2ns
looney2ns said:
I've had numerous original Dahuas, that it has never worked in.
Click to expand...

Dahua USA from what I known, is changing products ID for local market.
So no DH-HFW5442E-ZE-S3 like rest of the world, but in USA You have: N45FB7Z
OEM's have IPC- as replacement for DH-, sometimes they put (or order from Dahua) own product_id in camera flash drive.

Firmware upgrade works based on product_id which is on the camera flash drive.
If it is changed for OEM or specific country (like USA) firmware upgrade will no work.
 
Try using H.264H and AI Coding. I had similar regular pulsing in a grassy area with mine and that was the only thing that I found that helped it. I've run it that way for a while now and haven't seen any downsides. There may be some in some way but I don't see them in my use.
 
  • Like
Reactions: bigredfish
steve1225 said:
Dahua USA from what I known, is changing products ID for local market.
So no DH-HFW5442E-ZE-S3 like rest of the world, but in USA You have: N45FB7Z
OEM's have IPC- as replacement for DH-, sometimes they put (or order from Dahua) own product_id in camera flash drive.

Firmware upgrade works based on product_id which is on the camera flash drive.
If it is changed for OEM or specific country (like USA) firmware upgrade will no work.
Click to expand...
Most of us here, do not use USA models, we use International models.
 
If scene complexity is overloading the video encoder, it would be worth a trying further down in FPS. 12 or 10 FPS would further reduce the demand.
 
You must log in or register to reply here.
Top Bottom