Mobius support thread and Tech Guide (post all problems/help requests here) Read posts #1-8 first

dirkzelf said:
it is indeed weird that the green smear stays on the right side of the screen when flipped, implying that it is not the sensor doing this.
Click to expand...
I'm sure that's wrong, the green smear stays on the right side as the camera sees it, so it could be the sensor. It is opposite the time stamp. But you've eliminated the environment as the cause.
It could be some kind of calibration error (I don't know if the Mobius uses some kind of calibration/bad pixel map, but if so it could have been done badly.)

(P.s. I'm not seeing all posts, apologies if I'm covering what anyone else has said.)
 
jokiin said:
although that's an analogue format it's still a valid comparison, it's all about maintaining consistency of output or end result, dealing with knowns rather than unknowns, it's also one of the reasons why with retailers we only supply cameras with memory cards included so we can have a reasonable expectation of how it's being used, until such time as they buy some 128gb card off eBay for $8 and then blame the camera o_O
Click to expand...

Yes. That is precisely the point I was trying to make. And with the VHS tape analogy, similarly to microSD cards, you can only record on them just so many times.
 
You really need to physically flip the camera to see if there is an issue. There just isn't one to speak of.
 
Last edited:
Dashmellow said:
The question is not whether what you are seeing is UNSEEN. The question is what is causing you to see it. I think that you are seeing a color shift on one side due to something about your headlights but if you are not willing to do any definitive testing then it is a matter of what you "believe" over what "is". If you don't see any "Green Monsters" during the daytime, it is highly unlikely that you would have something other than an external light source causing it to occur at night only. I don't believe there is anything wrong with your Mobius other than perhaps unrealistic expectations.

You asked: "I'd like you to show us some example screenshots of other camera's that show this exact same phenomeneon as proof for your diagnostic conclusion." Well, I already did that two weeks ago but you seem to want to ignore that. Here is the image again. It was from @kamkar1's JooVuuX and in this case the so called" Green Monster" you see on the surface of the road is caused by a beam of light that has a CRI (Color Rendering Index) in the green area of the spectrum as it appears to the CMOS in your camera.

View attachment 21863

I keep trying to explain how different light sources will appear as different colors on camera. Here is another excellent example of the phenomenon. The entire right side of the road has a "Green Monster". Of course, on the left is the "Orange Monster".

View attachment 21864
Click to expand...

The only logical conclusion SEEMS to be the car's headlights, the green spot is there, and it's always on the same place, (it doesn't move with streetlighs passing by) but if you've seen the screenshots I posted earlier, then you can see the green spot even with the headlights off, so that still leads me to think that there is something else going on.

Headlights are actually pointing to the right side of the road though, to avoid blinding upcoming traffic, so maybe there is something in that too, although with headlights off I see the same, so..... I don't know man....

I could swivel the camera around a bit to left and right when at standstill with headlights on and off, to see if the spot stays in the same place on the screen or if it stays with the headlights or something else.

I didn't conciuously notice your earlier examples, but there the discoloring is indeed obvious with the streetlights, I have'nt noticed that with the mobius yet though.

Eletoponline365 offered me to investigate, but then I need to send my lens modules back once again ($$$$$), and they would crosspost me some new ones (which I have to pay $$$$$ for... and get credit when mine arrives at their office), I'm not doing that. The most logical way would be that they examine 2 or 3 lens modules to check if they are presenting the same kind of thing, and if not, send me the new ones for a swap.....

UNWILLINGLY I just have'nt gotten around to do the nightly whitewall test yet, but I will as soon as the chance comes up. (Is this also your idea of "definitive testing"?) I just need to find me some where I can test it ONLY with and without headlights at night, outside, ruling out streetlights.

Or is there another way to do a test? Indoor low light testing seems to be not a good reference to me.

I'd like to add that the camera is approx 20-25 cm from the winshield of the truck and the windshield is almost 90 degree vertical upright, so maybe placing the camera with the lens stuck right up to the windscreen might give a different result....
 
dirkzelf said:
UNWILLINGLY I just have'nt gotten around to do the nightly whitewall test yet, but I will as soon as the chance comes up. (Is this also your idea of "definitive testing"?) I just need to find me some where I can test it ONLY with and without headlights at night, outside, ruling out streetlights.

Or is there another way to do a test? Indoor low light testing seems to be not a good reference to me.
Click to expand...

By "definitive testing" I guess I was suggesting an attempt to obtain some objective information that would tell you once and for all if there is really a green tint on one side of the image at night or not. A white or neutral colored wall would probably tell you what you want to know. Often you can find a large blank wall in the back of large buildings like a big box department store or supermarket. Street lights shouldn't be a huge problem as long as the illumination is relatively even. On a white or neutral colored wall you will either see your "green monster" or you won't see it and you could try it with and without your headlights on.

On the other hand if finding such an outdoor location is too much of a hassle I do think an indoor low light test on a blank white wall would also tell you if the camera has a green tint on one side of the sensor or not. It would be an easy test and would certainly be worth a try. If there is no green tint visible under those conditions on a blank white wall then the likelihood is that the problem is indeed caused by the color temperature of an artificial light source such as your headlights and not a defective sensor.
 
Last edited:
Dashmellow said:
By "definitive testing" I guess was suggesting an attempt to obtain some objective information that would tell you once and for all if there is really a green tint on one side of the image at night or not. A white or neutral colored wall would probably tell you what you want to know. Often you can find a large blank wall in the back of large buildings like a big box department store or supermarket. Street lights shouldn't be a huge problem as long as the illumination is relatively even. On a white or neutral colored wall you will either see your "green monster" or you won't see it and you could try it with and without your headlights on.

On the other hand if finding such an outdoor location is too much of a hassle I do think an indoor low light test on a blank white wall would also tell you if the camera has a green tint on one side of the sensor or not. It would be an easy test and would certainly be worth a try. If there is no green tint visible under those conditions on a blank white wall then the likelihood is that the problem is indeed caused by the color temperature of an artificial light source such as your headlights and not a defective sensor.
Click to expand...

OK, got it. Finding a white wall should be no biggie, I just didn't get to it yet.....

I have been thinking though, all truck cabines have a small rear view window, since the camera is always on the left side of the dash, actually exactly in front of me, so the greening in the right side screen part might be a reflection of that small window, or maybe even the rightside window....... does that make any sense? Huh... I'll cover the rear window tomorrow evening to see if it brings anything. I might be able to do the outside car thingie as well.

I will also place the camera against the windscreen to see if that makes any difference....


Stay tuned......
 
Dashmellow said:
By "definitive testing" I guess I was suggesting an attempt to obtain some objective information that would tell you once and for all if there is really a green tint on one side of the image at night or not. A white or neutral colored wall would probably tell you what you want to know. Often you can find a large blank wall in the back of large buildings like a big box department store or supermarket. Street lights shouldn't be a huge problem as long as the illumination is relatively even. On a white or neutral colored wall you will either see your "green monster" or you won't see it and you could try it with and without your headlights on.

On the other hand if finding such an outdoor location is too much of a hassle I do think an indoor low light test on a blank white wall would also tell you if the camera has a green tint on one side of the sensor or not. It would be an easy test and would certainly be worth a try. If there is no green tint visible under those conditions on a blank white wall then the likelihood is that the problem is indeed caused by the color temperature of an artificial light source such as your headlights and not a defective sensor.
Click to expand...

3 screenshots when walking outside, I think we can rule out the car's headlights......

I had it mounted on my baseball cap, but didn't realize it's upside down.... But we can all se kermit very clearly. If need be I could post the video on yt.

vlcsnap-2016-03-29-21h27m32s602.png vlcsnap-2016-03-29-21h28m08s775.png vlcsnap-2016-03-29-21h28m54s478.png
 
The only thing I see here are different types of street lights with different color temperatures causing green and orange color shifts in exactly the same manor that I've tried my best to explain and demonstrate several times now.

dirkszelf.png

What we see in your baseball cap image above is the exact same green/orange street lamp color temperature phenomenon I've previously tried to demonstrate in this image below and elsewhere. Your "green kermit monster" simply does not appear to be caused by any obvious flaw in your Mobius. Unless you are willing to do some proper testing in a controlled environment there isn't much more to be said about it. This is getting pretty silly already.
colortemps.JPG
 
Last edited:
jokiin said:
it's a great value, reliable product, night video isn't its strong point though
Click to expand...
Agree completely.

FW 0.59 is pretty decent for my needs. Ive tried all the firmware up after with much testing and setting adjustments, its much clearer at night. I can stand the slight orange hue in the daytime.

As you say for the price point its amazing. Im after number plates and where other cars are in relation to me, not IMAX movie quality. I think some folks get a bit past why we have the cameras in the first place. If they aren't reliable like the mobius it doesn't matter how good the footage is if it didn't record your accident.

I think its spoilt us really, I remember when it first came out watching the YouTube footage and being amazed by the clarity of the video for a tiny camera.

As for me, Ive got my 5th one on order now as a spare unit in case I need it. :)
 
Last edited:
newton said:
Agree completely.

FW 0.59 is pretty decent for my needs. Ive tried all the firmware up after with much testing and setting adjustments, its much clearer at night. I can stand the slight orange hue in the daytime.

As you say for the price point its amazing. Im after number plates and where other cars are in relation to me, not IMAX movie quality. I think some folks get a bit past why we have the cameras in the first place. If they aren't reliable like the mobius it doesn't matter how good the footage is if it didn't record your accident.

I think its spoilt us really, I remember when it first came out watching the YouTube footage and being amazed by the clarity of the video for a tiny camera.

As for me, Ive got my 5th one on order now as a spare unit in case I need it. :)
Click to expand...

I couldn't agree more about remembering why we have dash cams in the first place and the expectations some people seem to have about cinematic image quality.
 
newton said:
Agree completely.

FW 0.59 is pretty decent for my needs. Ive tried all the firmware up after with much testing and setting adjustments, its much clearer at night. I can stand the slight orange hue in the daytime.

As you say for the price point its amazing. Im after number plates and where other cars are in relation to me, not IMAX movie quality. I think some folks get a bit past why we have the cameras in the first place. If they aren't reliable like the mobius it doesn't matter how good the footage is if it didn't record your accident.

I think its spoilt us really
Click to expand...

I agree too, we are spoiled. It's just that it appears now with the new c2 lens, and now that you mentioned it, I've just flashed one of them with 0.59 also, to see the difference with 2.41, just for the heck of it.

Do you see the discoloring like in my examples at all?

May I ask what settings you have chosen?
 
Dashmellow said:
I couldn't agree more about remembering why we have dash cams in the first place and the expectations some people seem to have about cinematic image quality.
Click to expand...

Come on dashmellow, you know by now it has nothing to do with my expectations, I saw a change for the worse in performance with a new lens, and it is just logical to be disappointed about that and to seek a solution.
 
jokiin said:
it's a great value, reliable product, night video isn't its strong point though
Click to expand...

I agree, I've used them for almost 4 years, never let me down, until the 2 failed within a week from one another.... and I still have a lot of faith in them, and still I have hope that other firmware might relieve me.....
 
I mounted my mobius cameras all central and facing down (so a third of the field of view is dash as an example) to reduce facing directly into the sky etc that confuses the sensor. Also it reduces glare and any reflection and colour distortion. I did get green glares and all sorts if I recall correctly, but changed the camera angle during set up as I identified that being the issue right away. I also have a slight tint to my glass that would create this.

Setting are just standard WDR on 50mhz light frequency, immediate disconnect etc (supercap settings just like @russ331 posted before), 5 min MOV clips, 1080p res with high video rate, wide FOV and 30FPS etc etc.

As its FW 0.59, no advanced settings because you cant muck around with it and mess it up.

dirkzelf said:
I agree, I've used them for almost 4 years, never let me down, until the 2 failed within a week from one another.... and I still have a lot of faith in them, and still I have hope that other firmware might relieve me.....
Click to expand...
Ive really torn my first mobius apart and rebuilt it several times fiddling and inadvertently making half of the situations at some point people have had. Ive always found whatever the issue, its either lens out of focus or lens ribbon cable, battery or 89% of the time memory card. The chip and board itself is shocking reliable.

But to be fair and unbiased, please note I have not got the C2 lens. I was completely satisfied with the B lens performance.
 
Last edited:
You must log in or register to reply here.
Back
Top