git2 poor video quality

right of course, lower mp sensor produce better image in low light, just like git1>git2 in low light

im considering using a dashcam as walk around action cam, if there's one that beats git2 in video quality in day light
 
Ajusshi said:
right of course, lower mp sensor produce better image in low light, just like git1>git2 in low light

im considering using a dashcam as walk around action cam, if there's one that beats git2 in video quality in day light
Click to expand...
Dashcams don't have gyro stabilisers, they will never beat the Git2!

Also the batteries are too small, they set the exposure for the road which isn't always what you want in an action cam and they and try to store detail rather than accuracy or niceness.

You can get good results using a dashcam as an action cam, and the Git2 can give good results when used as a dashcam but they are different tools for different jobs.
 
Ajusshi said:
oh ok, I thought it would be better from youtube videos. What dashcam/ no screen cam compares to git2 video quality?
Click to expand...
you can't judge a camera's true quality from what you see on youtube, because youtube and other streaming sites compress the heck out of video, and as a result, you get blockiness, color banding, and other weird artifacts that aren't there in the raw video.

always try to find downloads of the ACTUAL video files from the cameras, so you can see for yourself how the camera really performs.
 
Observing with my HD monitor I won't see the full res. anyway.
 
gitup said:
Normally the dashcam can do better than Git2 in low light condition as use 2MP, 3MP or 4MP sensor.
Our company @viofo will release a no screen dash cam with 2MP sony sensor next month.
Click to expand...

Gitup and Viofo is one company ? Didn't know that.

I have the A119 Dashcam (Novatek NT96660 + OV4689) and I love the level of detail (day and night) and low light performance is great on the dashcam !

When I got the Git2, I was kinda dissapointed that the level of detail on Git2 captured compared with the A119. The focus on the Git2 appears like it is only focused one feet a way. Everything after that looks very smoothen and lack detail. Low light performance on the Git2 is literally night and day compared with the A119 dashcam but of course much less noise on the Git2 (So that is understandable the dashcam may have been tuned for visibility rather than noise concerns)

I will try to post the comparison video for you guys to take a look.
 
Your A119 have much fewer and larger pixels on the OV4689 ( 2 μm ) than the sensor in the git 2 ( 1.34 μm ), so thats why you dont see the same level of low light performance in the 2 cameras.
If anything you should compare that dashcamera to the git 1 camera that have a more similar pixel count, and then i think you would see the git1 being better.

It could be your git2 camera have been focused a little too short, this happen once in a while, even with the best of brands in dashcams and action cameras.
 
kamkar1 said:
Your A119 have much fewer and larger pixels on the OV4689 ( 2 μm ) than the sensor in the git 2 ( 1.34 μm ), so thats why you dont see the same level of low light performance in the 2 cameras.
If anything you should compare that dashcamera to the git 1 camera that have a more similar pixel count, and then i think you would see the git1 being better.

It could be your git2 camera have been focused a little too short, this happen once in a while, even with the best of brands in dashcams and action cameras.
Click to expand...

Thanks. I guess the question is then whats the use of more pixels if they don't do a better job in capturing more detail. ?😕 Understand on the larger pixels more light bit though.

When you say focus is a little too short and happen once in a while, do you mean it is as a defect or a feature ? and can it be corrected ?

Thanks
 
In daytime if you use all the pixels on the sensor then a camera with a larger pixel count should off course have a more detailed picture. but if you only use a little of those pixels say 1080p on the Git 2 camera then the camera will only use a few of the pixels on the sensor to make that resolution.
So a camera like the git2 first really shine if you use the largest resolutions where most of the sensor is in use.
Problem is to use the image stabilization of the Git 2 camera you cam only do that with 1080P resolution, no action camera can do image stabilization with the full size of the image sensor in use.
Image stabilization can also ruin the footage a little bit i understand.

The larger the pixels in on the sensor the better they are in capturing light, some sensors in CCTV and other cameras have pixels thats around 5 μm in size so they can really capture a lot of light.

You can change the focus on most dashcam or action cameras, but this normally mean you have to take the camera apart and so void warranty, so best to talk with seller first.
Videos on how to do this on Gopro and other cameras on the youtube.
But in general what you do is unscrew the camera lens, often they are secured in place with a little hot glue so they can be hard to get loose.
And then you just unscrew the lens more or less to change focus, the sweet spot can be hard to hit as its only a little turn of the lens thats needed to be in focus or not so good focus.
Good thing as i re call is the Git 2 have a HDMI port, so you can connect it to a TV to check the focus on the fly.
After refocus of lens you have to lock lens in place again, dont use Cyano glue for this as it will ruin coatings on lens glass.

To see if focus is bad on a camera it is best if you place it where it can not move, and then film something with a lot of not moving things at different distances to the sensor, and off course do that test recording in daylight.
 
You must log in or register to reply here.
Back
Top