NVR and Camera Roles with IVS / Motion

scrumi

n3wb
Joined
Oct 15, 2020
Messages
3
Reaction score
0
Location
USA
I have an Empire Tech / Dahua NVR5208 and several Dahua cameras, IPC-HDW5831R-ZE & IPC-HDBW3441F-AS-M. While I get that motion and IVS can* be done on the camera or the NVR and if done on the camera then the event is sent to the NVR for storage if needed, and if set on the NVR I'm guessing the information is pushed out to the camera so processing is done at the nodes rather than the hub. I guess I wanted to see if my guessing is correct, or how this is handled. Is it best to configure those specific items out at the camera level rather than pushing them from the NVR?
 

Shockwave199

Known around here
Joined
Mar 13, 2014
Messages
1,016
Reaction score
550
Location
New York
It's generally advised to do it in camera I believe. If memory serves, when I first set up motion detection in the nvr and enabled it for a channel, it still wasn't working. Turns out that even though I enable it in the nvr it somehow didn't push that setting to the cameras and I had to do it in camera. After that if I tweaked motion zones in the nvr, the camera would update any changes without issue. I've only ever set ivs rules in camera and that's because I find the camera UI better to navigate for that. On the nvr side, there have been instances where I had to 'refresh' in the page for it to recognize a setting change I made in the camera. But mostly I have seen either side handshake between the two regardless of where you do the settings for detection. I only do exposure settings in camera, never the nvr.
 

scrumi

n3wb
Joined
Oct 15, 2020
Messages
3
Reaction score
0
Location
USA
Ok what about recording? The cameras send the event information to the NVR and the NVR still makes the ultimate decision on what to actually do to that information?
 

Shockwave199

Known around here
Joined
Mar 13, 2014
Messages
1,016
Reaction score
550
Location
New York
I believe the cameras are the front end of processing the triggers. The nvr is recording them. Or if you have an SD card in the camera it gets recorded to that. I could stand corrected though...
 

catcamstar

Known around here
Joined
Jan 28, 2018
Messages
1,659
Reaction score
1,193
Ok what about recording? The cameras send the event information to the NVR and the NVR still makes the ultimate decision on what to actually do to that information?
If you look in the "Storage" Settings on your NVR, you'll see WHAT your NVR will do WHEN any event happens (eg: MD / IVS || 24/7 recording). You can even specify per day/timeslot on what to do when.

Hope this helps!
CC
 

bigredfish

Known around here
Joined
Sep 5, 2016
Messages
17,016
Reaction score
47,477
Location
Floriduh
The camera decides when to activate based on rules for events. The NVR simply records it.

Regardless of what the manual may say, recommend you make all image, event, encoding settings on the camera.
 

soyrunner

Young grasshopper
Joined
Feb 6, 2020
Messages
37
Reaction score
7
Location
Georgia
My 4 cameras stream but only 3 can store to an sd card or ftp the alarms and recordings out. I'm told that NVRs can't accept ftp's files from cameras unless the NVR is modified? The NVR would override settings in the cameras. Accessing the files on the NVR is a problem so for now I ftp from the cameras and ftp from the NVR to a server and modify the files to view. Modifying the files takes hours so this really isn't workable long term.
 

KawaNinja

n3wb
Joined
Oct 28, 2020
Messages
15
Reaction score
3
Location
ISRAEL
The camera decides when to activate based on rules for events. The NVR simply records it.

Regardless of what the manual may say, recommend you make all image, event, encoding settings on the camera.
I'm trying to achieve the same thing, meaning using the Dahua camera to do the Motion Detection and trigger the Dahua NVR to do the recording.
I plan to use the MD part to faster locate events in a long footage.
I don't want the NVR to analyze 12+ camera images all the time just to trigger MD recording.
As you are suggesting I use WEB interface to set the Motion Detection on the camera, I can also see it in the camera log that the MD is working fine.
Even though I set the storage schedule in the NVR for both continuous (General) and MD 24x7 only General (green) is recorded.
Do you have any idea what am I doing wrong ?
 

bigredfish

Known around here
Joined
Sep 5, 2016
Messages
17,016
Reaction score
47,477
Location
Floriduh
What camera model?
What NVR model?
NVR 4.0 or 3.0 interface?

Are you using MD or IVS/Tripwire?

Check
1- Record is checked on the camera (IVS or MD)

cameraIVS-Record.jpg cameraMD-Record.jpg

2- On NVR both Schedule and Record screens are set

02.jpg 04.jpg
 

Shockwave199

Known around here
Joined
Mar 13, 2014
Messages
1,016
Reaction score
550
Location
New York
What camera model?
What NVR model?
NVR 4.0 or 3.0 interface?

Are you using MD or IVS/Tripwire?

Check
1- Record is checked on the camera (IVS or MD)

View attachment 73598 View attachment 73599

2- On NVR both Schedule and Record screens are set

View attachment 73596 View attachment 73597
The zigzag trip wire - that makes for better trip detection in some scenes? Any idea why that helps? Actually I'm finding intrusion areas to work better than trips and when needed, adding trips within.
 

KawaNinja

n3wb
Joined
Oct 28, 2020
Messages
15
Reaction score
3
Location
ISRAEL
What camera model?
What NVR model?
NVR 4.0 or 3.0 interface?

Are you using MD or IVS/Tripwire?

Check
1- Record is checked on the camera (IVS or MD)

View attachment 73598 View attachment 73599

2- On NVR both Schedule and Record screens are set

View attachment 73596 View attachment 73597
Dahua cameras
IPC-HFW1435S-W-S2
and
DH-IPC-K42
IPC-A26HS

NVR v4.0 Using SmartPSS GUI
Dahua NVR4116HS-4KS2

Rebooted several times

The logical connection from the camera MD trigger to NVR Channel recording is not clear to me
Manuals are terrible, they just repeat what you see on the GUI
As far as I understand if I select Motion Detection - ENABLE on the NVR side for the channel, the processing is done on the NVR side
SO how do I yell the NVR to start recording on the in-camera MD ? ?
 

bigredfish

Known around here
Joined
Sep 5, 2016
Messages
17,016
Reaction score
47,477
Location
Floriduh

bigredfish

Known around here
Joined
Sep 5, 2016
Messages
17,016
Reaction score
47,477
Location
Floriduh
Dahua cameras
IPC-HFW1435S-W-S2
and
DH-IPC-K42
IPC-A26HS

NVR v4.0 Using SmartPSS GUI
Dahua NVR4116HS-4KS2

Rebooted several times

The logical connection from the camera MD trigger to NVR Channel recording is not clear to me
Manuals are terrible, they just repeat what you see on the GUI
As far as I understand if I select Motion Detection - ENABLE on the NVR side for the channel, the processing is done on the NVR side
SO how do I yell the NVR to start recording on the in-camera MD ? ?

In theory either should do the same. But my experience is that using the NVR to make settings such as Image, Exposure, Event, etc works best if you never touch those settings on the NVR ad make all of those on the camera Web interface. Once you start flipping back and forth making changes on both, bad things happen.

How are you viewing playback to make the assumption MD events arent being recorded? Have you tried using SmartPSS? It could be a browser issue that you're just not seeing the events on the timeline... on the 4.0 interface some features and controls in playback simply dont show up unless using regular IE for example.
playback1.jpg
 

bigredfish

Known around here
Joined
Sep 5, 2016
Messages
17,016
Reaction score
47,477
Location
Floriduh
The zigzag trip wire - that makes for better trip detection in some scenes? Any idea why that helps? Actually I'm finding intrusion areas to work better than trips and when needed, adding trips within.
The other benefit to the zig zag line is it provides multiple points of opportunity for IVS to activate.

Picture a normal single tripwire line. IVS has only one opportunity to catch the motion. Now look at my zig zag line. There are 4-5 points of contact on the same “single” line
 

Shockwave199

Known around here
Joined
Mar 13, 2014
Messages
1,016
Reaction score
550
Location
New York
The nvr channel is mirroring the camera settings.
The other benefit to the zig zag line is it provides multiple points of opportunity for IVS to activate.

Picture a normal single tripwire line. IVS has only one opportunity to catch the motion. Now look at my zig zag line. There are 4-5 points of contact on the same “single” line
Yeah I see. Thanks. You know your comment to think of drawing in 3D rather than 2D actually helps a lot too. I started getting that concept earlier when drawing a difficult scene that wasn't triggering. Sometimes it's not just a horizontal plane but instead adding some vertical to it as well which can help better triggering.
 

catcamstar

Known around here
Joined
Jan 28, 2018
Messages
1,659
Reaction score
1,193
The nvr channel is mirroring the camera settings.

Yeah I see. Thanks. You know your comment to think of drawing in 3D rather than 2D actually helps a lot too. I started getting that concept earlier when drawing a difficult scene that wasn't triggering. Sometimes it's not just a horizontal plane but instead adding some vertical to it as well which can help better triggering.
My suggestion is to try-before-you-buy I mean implement: ask a friend/hubby/neighboor to take position on the curb/driveway/patio and open SmartPSS. You'll see the blue drawbox (object detection) being drawn. That's your "aim". This helps you drawing your minsize and maxsize frames too. But more importantly, let you draw the tripwires/intrusionboxes as they would work best for YOUR scenario.

Happy Camming!
CC
 

Shockwave199

Known around here
Joined
Mar 13, 2014
Messages
1,016
Reaction score
550
Location
New York
Make sure on the playback timeline you are checking (enabling) yellow motion detection playback. If playback is at 24hrs sometimes you can miss small yellow motion events on the timeline. Try 2hr or 1hr zoom and when you locate something, go for 30 min for a better event zoom size.

Setting everything in the NVR is really tweaking the camera too. The nvr isn't 'isolated' from the camera. The nvr and camera are synced to push changes you make on either side. But as been pointed out, it's usually better to enable motion detection and IVS and regions/rules in the camera. Once you save it and look back in the nvr it should mirror the settings. Sometimes the sync to camera can get screwed up if you do it all in the nvr. I haven't really had that problem in a long while but it can happen- mostly with exposure and encoding settings. Usually all is well with motion settings though.
 

Shockwave199

Known around here
Joined
Mar 13, 2014
Messages
1,016
Reaction score
550
Location
New York
My suggestion is to try-before-you-buy I mean implement: ask a friend/hubby/neighboor to take position on the curb/driveway/patio and open SmartPSS. You'll see the blue drawbox (object detection) being drawn. That's your "aim". This helps you drawing your minsize and maxsize frames too. But more importantly, let you draw the tripwires/intrusionboxes as they would work best for YOUR scenario.

Happy Camming!
CC
Thanks. I've had many targets- cars, neighbors just happening along, animals and of course for the bulk of it- me lol. It takes time to dial it all in for sure. One thing I certainly agree on- IVS is WAY better than motion. Since setting up IVS in every channel, I've turned off motion altogether. We've had such fog here overnight for the past couple days that if regular motion detection would have been on it would have been useless.
 
Top