M2 footage dash and action

[jokiin]

If the budget allows have a look at the new Innovv C5, it's made for motorbike use

 

[Eidmantas]

Thanks, I will . Though it seems to be as good as a waterproof Mobius 1 (@30 fps). The double setup is quite dear, though it looks very interesting. Well in any case, I will keep on searching. Cheers.

Hopefully the Mobius 2 gets better soon, though I heard that a new version is in the works?

 

[Dashmellow]

Hey, just finished reading this thread. Wanted to get two Mobius 1 with waterproof cases for motorcycle use. But really wanted that 60FPS. Though with the (Mobius2&) night video being as terrible as I've seen here, I am not sure what to do now. Though there's no case for the '2' I thought it'd be much easier to waterproof it with it's design and all! Doh! I guess It's back to being happy with 30 FPS and ordering the Mobius 1?

An alternative approach might be to consider a dedicated bicycle camera instead of a motorcycle or action camera. The nice thing is that they are waterproof without needing an external case and have some other attractive features such as image stabilization, large battery capacity (RideEye) and at least for the Shimano 60 fps @1920x1080, WiFi, GPS via smartphone, etc. Not sure about the current pricing and other factors but maybe one of these might work for you. (There a few others to look at as well. There's one @jokiin has recommended but I can't think of the name at the moment.)

http://www.shimano-sportcamera.com/en/product.html

http://www.rideye.com/

 

[jokiin]

. (There a few others to look at as well. There's one @jokiin has recommended but I can't think of the name at the moment.)

the Cycliq Fly6 and Fly12 (maybe the ones you've seen me mention) are very popular bicycle cameras, don't think they'd work well for motorcycle use though

https://cycliq.com/

 

[Dashmellow]

the Cycliq Fly6 and Fly12 (maybe the ones you've seen me mention) are very popular bicycle cameras, don't think they'd work well for motorcycle use though

https://cycliq.com/

Yes, that was it. (them)

 

[SawMaster]

Some interesting features in those Cycliq cams could transfer to dashhcams, such as the alarm system with cellphone notification and the post-crash auto-save They are over-priced but so is everything else to do with bicycling these days

Phil

 

[TonyM]

One thing I had not tried with the M2 was 720p@120fps, so I decided to give it a go this morning. I was hoping that the benefit of 120fps, albeit at a narrower FOV, might make the camera suitable for running alongside the GC to capture details that the GC misses due to motion blur. It was grey and overcast, so the difference between the slowest exposure duration of 120 vs 30fps was visible. The M2 does not appear to resort to frame doubling even at 120fps in poor light conditions.

First of all, this static scene overlay confirms that the 720p mode on the M2 is a 1280x720 sample from the centre of the 1920x1080 sensor. I calculate that the FOV has reduced from 130 to 110 degrees.


Driving at 40mph, there's less motion blur in the M2 frame than in the GC, but it still doesn't render any more useful details:



Looking at a different scene with low vehicle speeds moving away from a junction, it is evident that there is nothing to be gained from using the M2 in 720p120 mode for dashcam use.



I do think that there are potential benefits from running a camera at 120fps since the above examples show much less motion blur in the M2 frames. However the M2 simply isn't able to resolve enough detail in that configuration to make it worthwhile.

 

[kamkar]

Do it spit out true 120 FPS or do it double some frames as it cant do 120 FPS as it use a slower than 1/120 S exposure.

One would assume 120 FPS would mean considerable less motion blur and also darker footage in these less than favorable lighting conditions.

 

[Dashmellow]

Do it spit out true 120 FPS or do it double some frames as it cant do 120 FPS as it use a slower than 1/120 S exposure.

One would assume 120 FPS would mean considerable less motion blur and also darker footage in these less than favorable lighting conditions.

As @TonyM said above, "The M2 does not appear to resort to frame doubling even at 120fps in poor light conditions.".

 

[kamkar]

Okay skipped that

Seem like the camera got to use a much faster exposure then, or are reluctant to leave the slower exposures behind.
Could be fun to see if one was able to do some manual changes to ISO and exposure timings.
1/120 are still a pretty slow exposure for moving things, i would like to see at least 2X that.

 

[TonyM]

As far as I can tell, it sticks to 120fps and ramps up the ISO to maintain exposure. The M2 has never given 'clear' images like the M1, as they always seem to be a bit murky, and 720p120 seems just the same.

 

[TonyM]

I was wondering if the 720p120 video this morning was hampered by compression due to my settings of h.264 and standard bitrate (VBR approx 24kbps). So I decided to give the M2 one last chance, using the highest bitrate setting (VBR approx 36kbps) and h.265 for the least possible compression.

The results:
1. No improvement in video quality
2. Video recording became "unstable" - see clip below
3. RTC affected by video glitches The time skips forward during unstable sections. Clip duration should have been 5:00min. Actual durations 4:55, 4:40, 4:19, 3:32....

 

[kamkar]

I agree on the murky footage, i get the same feeling with the Panasonic sensor in the SJ6 AIR, its just not as vibrant like the Sony sensors in other cameras.

 

[estore009]

I was wondering if the 720p120 video this morning was hampered by compression due to my settings of h.264 and standard bitrate (VBR approx 24kbps). So I decided to give the M2 one last chance, using the highest bitrate setting (VBR approx 36kbps) and h.265 for the least possible compression.

The results:
1. No improvement in video quality
2. Video recording became "unstable" - see clip below
3. RTC affected by video glitches The time skips forward during unstable sections. Clip duration should have been 5:00min. Actual durations 4:55, 4:40, 4:19, 3:32....

Which SW on this M2?

 

[TonyM]

Which SW on this M2?

It's on 2.08

 

[estore009]

It's on 2.08

Peter said to us that its newer 2.10.6 FW to be much superior to the prior beta versions,its exposure and low light details are much superior to the M1, and visibly better than the samples shot with M2 2.08 FW. So,we wonder if you can email Peter directly for its newer FW,ok?

 

[jokiin]

Would have been smarter to send the newer version to those that had been beta testing already

 

[TonyM]

Peter said to us that its newer 2.10.6 FW to be much superior to the prior beta versions,its exposure and low light details are much superior to the M1, and visibly better than the samples shot with M2 2.08 FW. So,we wonder if you can email Peter directly for its newer FW,ok?

I'm not part of the testing team, nor do I have Peter's email.

 

[kamkar]

I have PM peters email to TonyM, and mailed peter myself

 

[TonyM]

I reinstalled FW2.08, reformatted the card and ran the same test at 720p120 and highest bitrate. This time there were no video errors, on either of two transcend cards, using both video modes to try with and without WDR.
I can't be sure whether reinstalling the firmware or reformatting the card was the solution. However the DSP seems to be ok with that 720p120 setting.