M2 Discussion

Yeah, the M6 certainly doesn't look very "ruggedized" compared with many other dash cam designs on the market. The shape though makes it look like it might be a good candidate for @SawMaster's rubber inner tube sheathing trick.
 
Just been capturing a little low light footage with the M2 FW i got on mail today ( 2.0.7 )

Here is 3 screen grabs from the 3 cameras in my windscreen.
9665GC_low_light.jpg

9665XS_low_light.jpg

And finally the M2
M2_low_light_(_30_fps_WDR_).jpg


I dunno if i missed something tried FW update using both method 1 & 2 but after that the GUI still say V 2.0.6.
But there is now a folder called lock on the sd card and the settings was default so i had to redo those, so i assume the firmware install was okay ???

This is with 30 FPS & WDR on. the same place with 60 FPS and WDR on look pretty much the same.
 
If dropped, I don't see the M6 fairing well.

With no bracket and that shape though, should fit in a pocket.

Like I said the yesterday, I wish you good luck in your dash camera endeavors but I'm of the humble opinion that this thread is not the place to be posting commentary about and a Techmoan video review of the DDPai M6 +.;)

Yes apologies for going off topic.
 
Just been capturing a little low light footage with the M2 FW i got on mail today ( 2.0.7 )

Here is 3 screen grabs from the 3 cameras in my windscreen.

Wow, I knew the M2 had issues, but I didn't realise they were that severe, thats really poor.

Its such a shame that the IQ is so poor, by the time it catches up it'll be outdated by other cameras that are doing higher resolutions at high frame rates..
 
To be honest i think night time performance have taken a dive as the daytime performance have been tweaked, we will have to see where it ends as low light now seem to be the focus point.
 
I just don't have much time to put together any sort of presentation at the moment but I hope to post something soon.

In the meantime, I am working with a very recent beta trying to make some definitive simultaneous comparisons between the M2 and the M1 as well as with the GC and the bottom line is that the IQ for the M2 has been very discouraging. Little by little things are improving but I am beginning to lose faith. At this point I can see where the camera "might" be suitable for RC FPV use (long distance daytime views) but for dash cam purposes the camera is sorely lacking. :(
 
In the meantime, I am working with a very recent beta trying to make some definitive simultaneous comparisons between the M2 and the M1 as well as with the GC and the bottom line is that the IQ for the M2 has been very discouraging. Little by little things are improving but I am beginning to lose faith. At this point I can see where the camera "might" be suitable for RC FPV use (long distance daytime views) but for dash cam purposes the camera is sorely lacking. :(

it's a shame to hear it's shaping up this way, I'm sure after having some closer involvement in this process over the past few months and given that the guys involved are quite experienced (certainly well above a lot of their competitors) you must have a greater appreciation of just how difficult it can be to make improvements
 
it's a shame to hear it's shaping up this way, I'm sure after having some closer involvement in this process over the past few months and given that the guys involved are quite experienced (certainly well above a lot of their competitors) you must have a greater appreciation of just how difficult it can be to make improvements

Yes, I can appreciate the difficulty involved, especially when a developer like Mobius takes on the task of having to essentially write their own firmware from scratch in Linux. I'm starting to think that the ambitious choice of components for this camera are proving to be much more of a challenge than originally anticipated. Obviously you're no stranger to this kind of thing but for the most part your team gets to work with a full fledged SDK.
 
Yes, I can appreciate the difficulty involved, especially when a developer like Mobius takes on the task of having to essentially write their own firmware from scratch in Linux. I'm starting to think that the ambitious choice of components for this camera are proving to be much more of a challenge than originally anticipated. Obviously you're no stranger to this kind of thing but for the most part your team gets to work with a full fledged SDK.

they also have access to a full fledged SDK but chose not to use it, on reflection perhaps that wasn't the best choice, it does perhaps highlight why working within the SDK framework is the preferred option, too many people make throw away comments like why don't you make this change or that, it's only a few lines of code, ten minutes work etc etc when the reality is they have no clue what they're talking about or more importantly what the implications are of (in their minds) seemingly trivial changes
 
they also have access to a full fledged SDK but chose not to use it, on reflection perhaps that wasn't the best choice, it does perhaps highlight why working within the SDK framework is the preferred option, too many people make throw away comments like why don't you make this change or that, it's only a few lines of code, ten minutes work etc etc when the reality is they have no clue what they're talking about or more importantly what the implications are of (in their minds) seemingly trivial changes

My understanding has been that there is an SDK but that it is fairly limited. At least that is the premise I've been going on. I'm trying to remember where exactly I read that now.
 
Well, I'm 'sposed to be working on some pressing paperwork, but what the hell, here are a few screen shots from yesterday.

Here is the M2. I have two units, both with similar but different lenses and both have a tendency to lose critical sharpness and acuity especially with moving objects.
This shot is H.264/30fps/WDR-off with all other settings at default. With WDR off, things tend to look a bit dark. It was recommended to me to shoot with WDR off to see if it made a difference with acuity but I'm not seeing it.
This happens to be the stock lens as currently ships with the camera.

M2highway.jpg


Here's a simultaneous screen shot from the GC.
GChighway.jpg
 
My understanding has been that there is an SDK but that it is fairly limited. At least that is the premise I've been going on. I'm trying to remember where exactly I read that now.

perhaps they felt there were limitations for what they hoped to achieve but the SDK is by no means limited from a video performance aspect, that is after all the primary purpose of the product, perhaps they had grander plans, I guess when they started this process close to two years ago the SDK may not have delivered at that time, could have been why they decided against it
 
perhaps they felt there were limitations for what they hoped to achieve but the SDK is by no means limited from a video performance aspect, that is after all the primary purpose of the product, perhaps they had grander plans, I guess when they started this process close to two years ago the SDK may not have delivered at that time, could have been why they decided against it

I guess whatever the explanation may be there is still a long road to travel until the M2 achieves it's original promise.
 
Yep not camera is every going to be totally drop proof. Even my Mobius M1 eventually landed lens down.
Maybe you should get one of these:
$_1 (1).JPG
This is a Mobius camera with the wide-angle lens. It's been fitted with a 3D-printed lens protector which effectively guards the lens from impact

Mobius HD Camera with Lens protector
https://www.ebay.co.uk/itm/192131427104
 
If development of the M2 is difficult, time consuming and ultimately expensive surely the best thing the manufacturer can do is to can the project and start on the M3 using a known chipset that has an SDK, it seems the M2 is just so far away from being viable its probably best to cut their losses ?

They can still sell them to the R/C folk that don't have an issue with them and then offer the M3 when its ready.
 
Well yeah it is true most RC people do their thing in the daytime, and the mobius cameras do have a more fortunate shape for most RC stuff.
Sure there are things where you can also strap your gopro on and be good, but the mobius shape just open up so many more avenues of recording your RC adventures.
With the mobius cameras i dont think there is a factory or office some place where a handful of engineers are hunched over computers while they bash away on the firmwares.
I am more inclined to think its a bunch of enthusiastic people with some skills, and if thats the case i think thats pretty cool daring to go down a path less traveled.
 
I think it's a case of them biting off more than they can chew. They were developing the Mini, the M2, and learning how to program with Linux all at the same time, as well as continuing support for the M1. I haven't kept up with the Mini but I think it's finished.

I expected slow development for the M2 but it's been slower than I imagined. Perhaps they'd do better by hiring someone with better programming skills to develop most of the potential then finish the remaining details on their own. At the current development speed the M2 might be obsolete by the time they catch up, hurting sales and making future projects less possible for lack of funding :( The action cam market is in overdrive and you either keep up or get left behind while the rest win :eek:

I appreciate the "do all you can yourself" approach, but sometimes the need for speed makes that a poor choice.

Phil
 
It remains to be seen whether it is a question of programming prowess or a mix of components that is especially and unexpectedly problematic. Each platform has its own challenges, as we all know.

For example, the SG9665GC is a great camera but it has always had issues with handling high dynamic range situations. After a very long wait and quite a lot of effort in reworking the firmware AE tables, the high contrast/high dynamic range issues that the camera has struggled with have largely been conquered yet when pushed to the extreme you'll still see the issue manifest. It's "that" camera's Achilles heel, I guess, even though it is otherwise terrific.

Despite all its current flaws, the M2 handles very wide contrast/dynamic range scenarios quite well, even with WDR-OFF. It is one of the few things I've found impressive about the camera.

Today was a brilliant sunny day after we got a lot of snow two days ago. This is when you tend to encounter those kind of high contrast situations where you'll really notice what I'm describing here.

These screen shots are from just a couple of hours ago and despite the flaws we are seeing here and the much superior IQ from the GC, the M2's ability to hold detail in the bright highlights impresses, and this is with WDR-OFF compared with WDR-ON with the SG9665GC.

M2
M2_3_17a.jpg

SG9665GC
SG_3_17a.jpg

M2
M2_3_17_b.jpg

SG9665GC
SG_3_17b.jpg
 
the challenge is not about getting a camera to perform under some particularly difficult circumstance, the challenge is trying to find a balance that works in the widest variety of circumstances, the M2 can look great in some situations yet totally fall to pieces in others, I do have some other cameras based on the same hardware they're using that probably aren't as good in some situations but also don't fall to bits like the M2 can do at times, they are developed on the standard SDK though so although the hardware is the same it's not really a level playing field

if they hit a brick wall with the product rather than dropping it and moving on as some have suggested perhaps they should consider open sourcing it and let the community have a go, look at some of the modded firmware some of the Russians have done with basically no tools and hex editor, would be interesting to see what's possible if they had more to work with
 
Back
Top