Posts made by Philippe

Hello, since the firmware update v3.1.0.956_22041511 the detection of the camera has evolved and I had to adapt the detection settings. Everything is working perfectly now and the camera manages to detect a human far enough in my driveway without taking any vehicles or pedestrians on the road, a joy.

However, the camera still has trouble recognizing at times a vehicle that is reversing into my driveway, I think the AI should be worked on to more easily recognize a vehicle that is reversing.

Here is the current camera setup:

Have an option to put the RLC-823A camera on standby and have a global push notification, because currently there are three types of notification (Human, Animals, Vehicles) and therefore if a car and a human are detected at the same time then we will receive two notifications at the same time, one for the vehicle and a few seconds or minutes later another for the person who has been detected. This causes notification spam. The best would be to leave the choice to the user to receive a notification for each type of detection or simply a single notification of detection during a chosen period of time.

Example :

1. Enable an option to combine detection types into one (Camera Detection)
2. Put an interval timer between two push notifications (2 minutes in my example)
3. A vehicle comes in front of the camera at 10:30 a.m. and a person comes out of it, the camera will send a single notification to notify the owner of the camera, and if the camera detects any new movement during the interval 2 minutes, the camera will not send no further notification.
4. After the 2 minutes interval, if the camera detects new motion, then it will send push notification again.

Because currently, it is enough that I have a car to park in my driveway, if a person arrives on foot, then I will receive a notification for the person and 1 minute later I will receive a notification for the vehicle. Which causes notification spam for nothing at all. A single notification would be really nice.

@daniel-platz_501920706814188 I agree with you, no improvement with this beta, the camera can't keep up with a person walking slowly below it.

@reolink-fiona Video : https://drive.google.com/drive/folders/1TcHMQUZuCdZiU60H4Je7Z4Pz-J2hNn-j

@reolink-fiona Indeed the camera returns to its point of guard now, however the camera still has difficulty following a person who passes below

And the camera does not return to its point of guard....

For my part it does not change anything, the camera does not turn fast enough to follow a person who walks towards the camera and passes below. The camera can't keep up.

@kevo548_508782064763008 Yes I agree with you, it needs to be improved.

Yes, adding a sleep mode would be very useful to deactivate the camera when we have done work or when we are in the garden in good weather without the camera following us constantly.

@reolink-fiona I have the Reolink rlc-823a camera and I have the interval delay option to enable. However, here is an example of the problem I am having:

In the Reolink app, I set the push notification delay to 120 seconds.

It is 9:00 am, a vehicle arrives in my driveway and a person gets out of the vehicle. at 9:00 am and 5 seconds I receive a notification that a car has been detected, then at 9:02 am I receive a new notification to tell me that a person has been detected at 9:00 am...

So it's completely ridiculous to receive two notifications for the same detection timecode.

It would be easier to have only one push notification for the same detection timecode and put a delay of your choice during the moment when the camera should not send a notification again. And after this time, if there is a new detection, then the camera sends a new push notification.

Because the camera currently sends a notification separately for each category for the same timecode, so if there is a car coming and a person gets off with an animal, then I receive 3 notifications for the same detection. 1 instant notification for the car, 2 minutes later for the detected person and 2 minutes later for the animal which has also been detected.

It's really tiring in the end, instead of making a single notification that encompasses all three categories (animals, people, vehicles)

It's so tiring to receive push notifications for the same detection simply because the camera monitors if it detects a vehicle or a person.

It would be easier to send a single push notification when detecting multiple categories (people, vehicles) occurring in a close interval of each other.

Because currently my camera detects vehicles and people in my driveway. When a car arrives I get a push notification, and when the person gets out of their car, I get another push notification telling me there is a person... So my phone rings twice in 20 seconds for the same thing and in the end it is tiring.

Let us choose to receive only one push notification for the same detection period that we have defined (30,60,120 seconds) to avoid push notification spam.

@joseph-chircop_497308027822318

I have exactly the same problem with my RLC-823A camera, in "Low" and "Mid" quality I have no problem, but when I switch to "high" quality I sometimes have the image freezing and the audio choppy or even cutting out completely. I don't have a bandwidth issue either and I'm also on the same network. And in WAN I am also limited to 1.2 Mo/s for FTP downloading and mobile downloading, while in LAN I download up to 10 Mo/s.

The engineering team needs to look at where the problem is coming from, because we can't take advantage of the full potential of the cameras currently in high quality...

@joseph-chircop_497308027822318

In WAN I am connected in Optical Fiber 5 Gb/s with the French operator Free. On the LAN side, the camera is connected to a PoE+ switch of 100 mbps from TP-Link with an ethernet cable of 20m in cat 6 F/UTP which is connected to the router of the operator Orange France with a symmetrical speed of 300 mbps.

When I connect the ethernet cable which is connected to the camera to a laptop and I do a speedtest on speedtest .net I have a speed of 100 mbps in download and 100 mbps in upload and a ping of 2ms. When I connect the cable again to the camera, in LAN I have the 100 mbps because I have no live lag and I download the videos which are recorded on the SD card at 10 Mo/s. I switch to WAN, if I am in High quality the live broadcast starts to freeze, I go from 06:30:10 to 06:30:14 with a loss of a few seconds. And when I download a video recorded on the SD card I am restricted to 1.4 Mo/ s.


Yes there are still lots of things to improve and add, and the Reolink team are listening.

@joseph-chircop_497308027822318

It is true that there is a problem between the LAN and WAN connection. In LAN I have a speed of 100 mbps while in WAN the camera stops at 10 mbps. I do not understand why...

I just installed the beta version 4.29.3 for android, and indeed the playback of recorded videos is very slow now Looks like frame-by-frame playback instead of having smooth video playback

Hello, I have two RLC-823A, Horizontal auto tracking works fine but vertical tracking doesn't work well once the person gets below the camera. This one stops following instead of turning to continue following the person. The firmware v3.1.0.804 of January 20, 2022 are installed on my two cameras.

Hello, I wanted the opinion of the community, we agree that the button for manually activating the siren is useless as it is. Because the siren only goes off for 3 seconds... It would still have been more useful to be able to choose from an option of the desired activation time. The support tells me that some users find this disturbing for the neighborhood, but the purpose of a siren is to scare an individual and warn the neighbors... So if the siren only rings for 3 seconds, everyone nobody cares and the option of being able to release it manually for only 3 seconds is useless either.

I ask and hope the community too, that a maximum siren activation duration option can be configured in the camera options. Or that we can activate and deactivate the siren manually as we see fit