Vantrue N5 - 4CH Dash Camera Test & Review - RCG

[Karagandinez]

Will that be sufficient for your needs? Will it actually record anything related to the impact event?

It all depends on what the hit will be. If another car has to stop or turn around, then something might get into the frame.
If the blow is on a tangent, for example, the culprit blows your side mirror, then he may not even slow down and leave before the recording begins.

 

[Jeff_Vantrue]

Collision Detection Parking Mode

The collision detection parking mode reduces the power draw from the N5 to a very low level (15 mA @ 12.6V or 0.19 Watts) by powering down the N5 to a state where it's monitoring only for impacts. This greatly extends the amount of time that a vehicle battery or dash camera battery pack can power the N5 in this parking mode.

The really big downside to this parking mode (IMO) is that it takes 11 seconds (based on my 4-channel testing) for the N5 to power up an resume recording after an impact has been detected. It records a 60 second video. The video will contain the activity 11 seconds to 71 seconds after the impact event. Will that be sufficient for your needs? Will it actually record anything related to the impact event?

Is Vantrue N5's Buffered Motion Detection used?​

Vantrue N5 Buffered Motion Detection：​

In the 10s pre-recording buffered Motion Detection mode, once any motion around the car is detected, all 4 cameras will automatically start recording and will save the video for 10 seconds before and 30 seconds after the event occurs.

 

[rcg530]

Is Vantrue N5's Buffered Motion Detection used?

Vantrue N5 Buffered Motion Detection：​

In the 10s pre-recording buffered Motion Detection mode, once any motion around the car is detected, all 4 cameras will automatically start recording and will save the video for 10 seconds before and 30 seconds after the event occurs.

I understand that the N5 motion detection parking mode will have the 10 seconds before the triggering motion event and 30 seconds after the triggering motion event. The cost of using that parking mode is the very high power consumption while the N5 is waiting for a triggering motion event. In my testing, the N5 consumes 9.89 Watts while in motion detection parking mode (on average) waiting for a triggering motion event and it jumps up to 10.27 Watts while recording video due to a triggering motion event. The projected motion detection runtime when powered by a 96 Wh battery pack is 9 hours 42 minutes.

The post you replied to of mine was discussing collision detection parking mode. That mode has the benefit of using very little power since it places the N5 into a "deep sleep" low power consumption mode while waiting for a triggering impact event. The average power consumption in this mode is around 0.19 Watts. The trade off is the fact that it takes the N5 around 11 seconds after the triggering impact event to start recording 60 second videos from all 4 cameras. The projected collision detection runtime when powered by a 96 Wh battery pack is 21 days 1 hour 15 minutes.

The N5 present a screen as you enter the parking mode configuration screen suggesting to use the collision detection parking mode during hotter summer months.

 

[Jeff_Vantrue]

Collision Detection Parking Mode

The collision detection parking mode reduces the power draw from the N5 to a very low level (15 mA @ 12.6V or 0.19 Watts) by powering down the N5 to a state where it's monitoring only for impacts. This greatly extends the amount of time that a vehicle battery or dash camera battery pack can power the N5 in this parking mode.

The really big downside to this parking mode (IMO) is that it takes 11 seconds (based on my 4-channel testing) for the N5 to power up an resume recording after an impact has been detected. It records a 60 second video. The video will contain the activity 11 seconds to 71 seconds after the impact event. Will that be sufficient for your needs? Will it actually record anything related to the impact event?

Hi RCG530
About Collision Detection Parking Mode
Currently it is recording 1 minute, how many minutes do you suggest is best? I will have a discussion with the engineer.

 

[rcg530]

Hi RCG530
About Collision Detection Parking Mode
Currently it is recording 1 minute, how many minutes do you suggest is best? I will have a discussion with the engineer.

It’s not the length of the video recording that is my concern, it’s the 11 seconds it takes to start recording video that is the problem.

 

[Jeff_Vantrue]

It’s not the length of the video recording that is my concern, it’s the 11 seconds it takes to start recording video that is the problem.

Collision detection is to start recording 4 cameras at the same time, so the time will reach 11 seconds; the best way is to choose motion detection or low bit rate detection.

 

[kamkar]

I am wondering, could you start cameras one at the time, if that was the case i would start the front camera then the rear camera, and then after that the remaining cameras.
Chance are at least in a parking guard event that the footage from the outside cameras will be most important, but of course if someone enter you car in some way, then you of course also want a interior camera or cameras to be up and running.
But i am not aware if something like that are technically a option.

I still feel that triggered parking guard are a challenge no matter how good it is made, hence why i prefer to use low bitrate myself.
Maybe with that ( low bitrate ) have outside cameras record low bitrate, and then with G-sensor trigger the interior camera / cameras would come on too.

I have long asked for G-sensor sensitivity to be even higher, when in parking mode, like where a big truck or bus at speed passing by close would trigger, or say hail falling on the car.
Actually i have also brainstormed glass break sensors, and radars like some alarms use to detect people stopping near the car

 

[rcg530]

Collision detection is to start recording 4 cameras at the same time, so the time will reach 11 seconds; the best way is to choose motion detection or low bit rate detection.

Hi Jeff,

I agree that it will take some time to start recording after an impact event when using the collision detection parking mode, but any amount of time over 5 seconds (or so) to wake up the cameras and start recording greatly diminishes the helpfulness of the dash camera in catching on video what actually triggered the impact event. The tip presented by the dash camera when configuring parking mode "Please select collision detection if in summer exposure environment" is telling the user not to use motion detection or low bitrate parking modes during the summer. I realize that there's not going to be a perfect solution for a low power collision detection parking mode, but every second that can be shaved off of the boot time in that mode will make it a far more viable solution.

Other improvements could be made in motion detection and/or low bitrate parking mode to reduce the amount of power consumed. It might be beneficial to allow the interior cameras to be excluded (via a setting) from the motion detection or low bitrate recording to reduce the amount of power and hopefully the amount of heat those dash camera units will generate while in parking mode. That change could make those parking modes far more viable for hotter months of the year and extend the N5's parking mode times when powered by a dash camera battery pack.

 

[rcg530]

Thermal Pictures

I ran a test for one hour with the N5 configured with its default video resolution of 1440p x 1080p x 1080p x 1080p. The room temperature was 81F/27.2C. There was an overhead fan on running on a medium setting which should replicate the air movement in the passenger compartment while driving. I placed both the front and rear cameras on a camera stand I have so that airflow could reach all sides of the dash cameras.

Front Dash Camera (Front & Front Cabin Cameras):

The front camera case didn't get too hot during this test. My previous testing of the N5 resulted in the microSD card getting very hot. The microSD card resides under a plastic cap/button assembly and the electronics in the front camera housing around the microSD card generate a large mount of heat (156.8F/69.3C). This test was with no sunlight on the dash camera so I would expect that temperature to rise when exposed sunlight while driving.

 

[rcg530]

Rear Dash Camera (Rear & Rear Cabin Cameras):

The rear cameras generated far less heat than the front cameras. The max temperature was 113.8F/45.4C.

 

[Jeff_Vantrue]

Very good idea, I will discuss with the engineer to start 1 camera first, so that the startup time can be shortened.
Vantrue recently developed the automatic switching function of parking mode; For example, choose motion detection or low bit rate, and while the temperature inside the car continues to rise under sun exposure, the parking mode will automatically switch to collision detection to adapt to the temperature environment inside the car for normal work.

 

[kamkar]

Together, we might get lucky and get a good idea

 

[kamkar]

I was also thinking the interior cameras, well if no one get in the car they dont really have to record, so was it possible to put those on a trigger of their own, say for instance the cabin light.
That way those will only come online if someone open the door.
But i am not sure if cameras can be compartmentalized like that.

 

[Jeff_Vantrue]

I was also thinking the interior cameras, well if no one get in the car they dont really have to record, so was it possible to put those on a trigger of their own, say for instance the cabin light.
That way those will only come online if someone open the door.
But i am not sure if cameras can be compartmentalized like that.

I will discuss this suggestion with the Vantrue engineers.

 

[rcg530]

While testing out the N5 in my filming studio, I kept noticing that the N5 front camera is quite heavy when compared to other dash cameras. The N5 front camera (196 g / 0.43 lbs) and and GPS windshield mount (44 g / 0.097 lbs) weigh a combined total of 240 grams (0.53 lbs). That's only 8 grams under the weight of my DJI Mini 2 drone (with battery installed). I hope the magnetic mount and the swivel joint of the GPS windshield mount for the N5 successfully handle (no shaking / no vibrations) the amount of weight they'll be carrying.

 

[kamkar]

Maybe Vantrue use a copper heat sink, the heat sink on my threadripper CPU ( water-cooled ) is 900 grams of copper.

 

[rcg530]

Maybe Vantrue use a copper heat sink, the heat sink on my threadripper CPU ( water-cooled ) is 900 grams of copper.

That's one large chunk of copper!

 

[rcg530]

Today, I tackled the task of working on yet another dash camera jigsaw puzzle. Removing and/or moving some dash cameras to make room for the Vantrue N5. The rear window install wasn't that bad, but the front install was challenging due to the large size of the N5 front camera unit. My choices are limited on where I can place a front camera that has the interior camera as part of the front camera body. I can't place it too low on the windshield since that will obstruct my line of sight while driving. My only option is to place it somewhat behind (meaning between the mirror and the windshield) the right side of the center rear view mirror so the interior camera can see the passenger compartment. After everything was installed, I found the front interior and rear exterior camera video images to be mirror images of reality, so I found the setting in the firmware settings to turn off the "mirror" setting for those two cameras. Tomorrow, I'll see how the first drive goes with the N5 installed in the car.

Front camera installation:

The "white" marks on the outside of the front windshield are chips in the outside windshield glass. The police agency that previously owned this car must have driven this car on dirt roads with a lot of rocks hitting the windshield.

The USB cable supplied to connect the N5 front camera to the N5 rear camera can be a concern when trying to move the N5 front camera to the upper portion of the windshield. The USB-C cable end stands quite tall on the top of the N5 front camera. It would be nice to see the USB-C connector (firm plastic cable section) be a bit shorter allowing for more flexibility of the cable close to the N5 front camera.



Rear camera installation:

Due to how I route the rear camera cables through my car for these test installations, the USB cable for the N5 rear camera is routed across to the other side of the vehicle. I used a couple of cable retention clips to keep the cable out of the field of view for the rear interior rear camera.

 

[rcg530]

Thermal Pics From Test Drive

I now have the N5 4-channel dash camera installed in my car. The resolution selection is 1440p (front) + 1080p (front interior) + 1080p (rear interior) + 1080p (rear).

Today, I drove my weekly shopping trip route and the outside temps were in the mid-60sF/18.3C. I used my TOPDON TC004 thermal camera to capture temps from the N5 in two situations. First situation was at the end of a "normal mode" driving session. The car had been driven/parked three times. At the end of the fourth normal mode driving segment (which is a little over two hours into this trip), that is when these thermal pictures were taken. The fourth normal mode driving segment lasted 15 minutes.

The "HOT" spot was on the microSD card end of the front camera at 195.2F/90.7C right after the round cover/button assembly was moved away from the front camera body. The other sections of the front camera were very hot as well. After the round cover/button assembly had been removed from the N5 front camera body for 95 seconds, the temp in that area of the front camera body dropped by 15.1F/8.4C to 180.1F/82.3C. It looks like that cover/button assembly holds in the heat a bit right where the microSD card is located on the N5.



This next set of thermal pics were taken after the car had been parked for 40 minutes and the N5 was operating in low bitrate 1080p 15fps parking mode. The outside temp was 66F/18.8C and the ambient air temp inside of the car was 100F/37.7C. The bottom of the N5 front camera body was super hot with a temp of 201.8F/94.3C. That is the hottest temperature I've ever observed with any dash camera in any mode of operation (even during the much hotter summer months). The other dash cameras in the same car/situation were at least 20F/11.1C degrees cooler than the N5.

 

[rcg530]

Dropped/Duplicated Video Frames - Front & Front Interior Camera Footage

I found that the front camera is either dropping or duplicating a video frame about once every second in the video. I first observed this behavior while playing the video on my Windows 11 laptop. I imported the front camera video into Adobe Premiere Pro. I confirmed the video was dropping one video frame or duplicating a video frame about once every second for the front camera. I found similar examples in the front interior video, but not necessarily every second in the video. The resolution selection was 1440p (front) + 1080p (front interior) + 1080p (rear interior) + 1080p (rear) and the front camera's WDR feature was enabled and the rear camera's HDR feature was enabled (which the default values for those settings). I've reported the issues noted in this post to my Vantrue contact.

In this screenshot from Adobe Premiere Pro, I placed a marker at each occurrence of a dropped or duplicated frame.



The rear interior and rear camera video files don’t appear to have any dropped or duplicated frames. The rear interior and rear cameras have focus problems.

In general, the rear interior camera is a bit out of focus for the entire field of view. The front interior camera has everything in focus (see below).



The rear camera is out of focus (badly) except for a small amount on the extreme left side. If something appears on the extreme left of the field of view, it’s in focus but when it moves out of that small section to the main field of view it goes out of focus. Yes, I've confirmed the protective film has been removed from the rear camera lenses and they are clean and the rear window is clean.



Here's what the front camera footage looks like:



Here's what the front interior camera footage looks like:



I’ve uploaded the raw video files for the A (front) / B (front interior) / C (rear interior) / D (rear) cameras from the examples included in this post to my Google Drive: https://drive.google.com/drive/folders/1wPUM_-PzF8CaBzbfEHC39iDg3IUou5My?usp=sharing