Mobius A2 lens experiences & observations

@jokiin, to further address your question, here are two examples of the same exact scene under more similar lighting conditions with the camera set to A lens and C lens.

"A lens" setting
cyan.jpg

"C lens" setting
balanced.jpg


Also, here's another screen shot from today with the A2 camera set to the C lens setting with the sun behind the clouds and a red building in the FOV.

red_building.jpg
 
Last edited:
do you have two the same so you can do side by side, the difference is quite noticeable but the lighting is also quite different which may be further influencing the results

BTW, I only have one A2 lens Mobius camera, so "two the same side by side" is not an option.
 
Last edited:
the colours certainly do look better, realistically just because they're calling this lens an A2 is no reason to expect that the A lens settings would be appropriate anyway, the reality is the various field of views of the different lenses is not something they've been tailoring things to in the firmware but rather the other lens characteristics, the A2 lens is perhaps more closely related to the C lens than the original A lens
 
the colours certainly do look better, realistically just because they're calling this lens an A2 is no reason to expect that the A lens settings would be appropriate anyway, the reality is the various field of views of the different lenses is not something they've been tailoring things to in the firmware but rather the other lens characteristics, the A2 lens is perhaps more closely related to the C lens than the original A lens

Maybe my experience here might be helpful for you in sorting out the now well documented and quite similar color shifting issues with the SG9665GC that @Feitelijk has been reporting?
 
Yes, every developer should be showering me with multiple free dash cams to test! :)

when you want to test changes the best way to do that effectively is side by side so that all else remains equal and the test environment is as controlled as possible so it's only the specific adjustment you are looking at, when you try and do A-B testing with one camera it's impossible to replicate the rest of the test unless it's in a lab environment, the real world testing makes for far better source material than what you can do in a lab
 
when you want to test changes the best way to do that effectively is side by side so that all else remains equal and the test environment is as controlled as possible so it's only the specific adjustment you are looking at, when you try and do A-B testing with one camera it's impossible to replicate the rest of the test unless it's in a lab environment, the real world testing makes for far better source material than what you can do in a lab

I agree but the results here are hard to dispute. Frankly, I'm a bit puzzled why you are even making such a fuss over this considering what I've had to show here.
 
I agree but the results here are hard to dispute. Frankly, I'm a bit puzzled why you are even making such a fuss over this considering what I've had to show here.

you're puzzled because you're reading things into it that I'm not saying
 
the colours certainly do look better, realistically just because they're calling this lens an A2 is no reason to expect that the A lens settings would be appropriate anyway, the reality is the various field of views of the different lenses is not something they've been tailoring things to in the firmware but rather the other lens characteristics, the A2 lens is perhaps more closely related to the C lens than the original A lens

True. But users should be make aware of this. Without @Dashmellow, you and other enthusiasts on this forum a newbie will be totally lost. Cannot be doing @Mobius sales any good.
 
you're puzzled because you're reading things into it that I'm not saying

You said that, "it's impossible to replicate the rest of the test unless it's in a lab environment, the real world testing makes for far better source material than what you can do in a lab". What exactly are you "not" saying"?

My results here unequivocally demonstrate the issue. What more do you expect or want to see? Is this "real world testing or not"? This side by side business you're talking about is a red herring here.
 
True. But users should be make aware of this. Without @Dashmellow, you and other enthusiasts on this forum a newbie will be totally lost. Cannot be doing @Mobius sales any good.

I'm sure when they finally work out what settings are required they'll add an appropriate option in Msetup to deal with it, the Mobius is very much a community supported device so I think there's an expectation from the developers (right or wrong) that the users will have to dig around a bit to find the info to get what they want out of it, I'd think anyone that looks into a Mobius has done at least some amount of online research into it and has a bit of an idea of what to expect, not like it's a blind purchase at Wal-Mart or similar, that would need a very different approach
 
You said that, "it's impossible to replicate the rest of the test unless it's in a lab environment, the real world testing makes for far better source material than what you can do in a lab". What exactly are you "not" saying"?

My results here unequivocally demonstrate the issue. What more do you expect or want to see? Is this "real world testing or not"? This side by side business you're talking about is a red herring here.

you've managed to work out after a period of trial and error that the C lens option is a better fit than the A lens option, is that going to be the end of it and they'll just rename that or are they going to fine tune it to suit the actual lens, if so when they get to the fine tuning that is better done with side by side testing, if you can't test any further adjustments side by side then it will only be subjective as the the conditions won't be the same, real world testing works better for this
 
you've managed to work out after a period of trial and error that the C lens option is a better fit than the A lens option, is that going to be the end of it and they'll just rename that or are they going to fine tune it to suit the actual lens, if so when they get to the fine tuning that is better done with side by side testing, if you can't test any further adjustments side by side then it will only be subjective as the the conditions won't be the same, real world testing works better for this

Yes, you are right about that and I see what you are getting at now. What I've done here is point things in the right direction. As I said earlier today there really needs to be a specific setting tweaked for the A2 lens (and the D lens too) but that is up to the Mobius team to fully analyze and refine in a more controlled setting.

To tell you the truth there has been somewhat of a reluctance from Mobius to acknowledge this issue (and certain others - (M2 related) and I am patiently still waiting for a response to my reports and documentation. The only thing I've heard back quite some time ago about the A2 color balance shifting is "we're trying to reproduce the problem". This brings up an interesting point as so far I have not seen any other reports of this A2 lens white balance I've documented anywhere on the internet. I would love to hear from anyone else who has purchased an A2 lens.
 
Yes, you are right about that and I see what you are getting at now..

I can appreciate the written word is often open to interpretation, or misinterpretation quite easily, all good

To tell you the truth there has been somewhat of a reluctance from Mobius to acknowledge this issue (and certain others - (M2 related) and I am patiently still waiting for a response to my reports and documentation. The only thing I've heard back quite some time ago about the A2 color balance shifting is "we're trying to reproduce the problem".

can be a bit frustrating when problems can't be replicated, again that's where side by side examples with identical hardware is useful so even if they can't replicate it you can show them A-B tests under the same conditions of alternate settings etc, they'd be well served to give you another A2 lens camera to facilitate that type of testing if they really want to get it sorted to the best it can be, it might get to the point where you need to suggest they send you another cam so you can collect the side by side samples for them, they'd be crazy not to
 
I can appreciate the written word is often open to interpretation, or misinterpretation quite easily, all good



can be a bit frustrating when problems can't be replicated, again that's where side by side examples with identical hardware is useful so even if they can't replicate it you can show them A-B tests under the same conditions of alternate settings etc, they'd be well served to give you another A2 lens camera to facilitate that type of testing if they really want to get it sorted to the best it can be, it might get to the point where you need to suggest they send you another cam so you can collect the side by side samples for them, they'd be crazy not to

Yeah, providing me with another identical camera could be very helpful but the fact is that I've supplied more than enough evidence to Mobius via raw clips on MEGA to demonstrate that this is a real issue and the fact that changing the setting to the "C lens type" fixes the problem certainly seems to be a confirmation. I really think at this point that it might be very helpful to hear from another independent party using an A2 lens Mobius whether they are seeing this same color balance/color shifting issue or not.
 
I'll add some videos later tonight, will leave it on the A lens setting.
 
Last edited:
Let's drive Dashmellow crazy keeping track of cameras :eek: The ideal test setup would be:

-Older production cam w/ old "A" lens for establishing a known performance baseline
-Older production cam w/ new "A2" lens set at current test settings for comparison of sensor/hardware differences always matching below cam
-New production cam w/ new "A2" lens set at current test settings for discovering best settings always matching settings with above cam
-New production cam w/ new "A2" lens set at default settings to discover where issues are to be found

Running a set-up like this gives you every lens/cam combination needed to identify exactly what/where a problem is, and whether it's in the lens, sensor, hardware, or firmware. Thus the Mobius staff wouldn't have 'cannot replicate' issues if those were being caused by minor differences in the cams themselves somewhere and not the software, all the while having a well-known solid performance baseline to judge changes and differences against.

Unfortunately it might drive Dashmellow crazy keeping track of it all sending him to the local insane asylum where there is no internet :rolleyes: We will miss you here on DCT Dashmellow but I promise to visit you there occasionally :p

On a more serious note, while I can understand how a 'cannot replicate' problem could happen, I honestly think we're seeing too many of them. If a raw vid shows it, then it must exist. If it isn't the same with the Mobius staff when they use identical settings then that shows it is not the settings causing the problem but something else. When the issue being worked on is serious enough, Mobius could send the person having the problem a loaner cam/card and a prepaid shipping box then get the offending cam with the problem in their hands where the problem must then be found. And if then they still cannot replicate it, then becomes clear that there is something in the install causing it ie: PS variations, windshield angle, type of glass or it's tinting, cosmic radiation at the owners location etc. It is certainly extra effort, cost, and risk to Mobius but it also fixes things instead of leaving the cam owners with their problems extant until Mobius happens across the problem by chance themselves later on ;)

I do appreciate that a lot of this work is being done by unpaid volunteers who have limited time to invest into the effort- love you guys for doing that :D But there is a possibly better way here which could make life better for those few who do experience problems with this unique and wonderful little cam. Now back to my coffee :cool:

Phil
 
Well, that probably would drive me a bit crazy. Fortunately, the older production "A" lens wouldn't fit into the new production case (by .5mm!, it turns out). OK, now THAT drove me crazy!..... but it was probably a good thing though as a permanently installable "A" lens in the "A2" lens camera would probably have distracted me from exploring the different lens setting options in mSetup and discovering that this is more of a firmware issue than an optical issue or sensor problem.

If anything is driving me crazy here it is the M2 camera which I am nearly about to throw in the towel over but that's another story for another time and place, except that there have been some parallels with my experience providing test results and getting info about both cameras.

The specific answer I got from Mobius to my initial report of the white balance problems last February when I sent screen shots and posted RAW clips to MEGA, was, "We are still trying to duplicate the M1 A2 lens issue." At the same time I asked several specific questions and asked for the specs on the different lenses I've been working with and was told I would receive a reply "early next week" but didn't. More recently (about a week ago) I asked the same questions again and provided a detailed follow-up on this problem with screen shots and short MEGA clip update but am still awaiting a reply. For some reason there has been somewhat of a reticence to fully acknowledge certain issues I report or to provide specific details when queried which is a bit frustrating but word on the street is that the developer is involved in another project at the moment and that may be part of the explanation.

In any event, I think what we need here is some kind of independent confirmation from anyone else with a new production Mobius A2 camera who may be experiencing white balance issues or color shift problems (or not). Let's see what @harsh reports.

In the meantime if anyone is interested in the short clip I referred to here is the YouTube version and a link for the RAW clip on MEGA. It was unusual in this instance because all of my previous experience with the color shifts involved a shift to an overall CYAN cast from a normal color balance whereas here it starts out CYAN (on a rainy, snowy day) shortly after starting my vehicle and then the scene abruptly shifts back to a normal color balance when my vehicle is facing in a different direction.


https://mega.nz/#!t1EkkSiK!b4MgldYjLedfJfZZ-mBvvKLzhctOW69HM5AmVK28R4g
 
Well, that probably would drive me a bit crazy. Fortunately, the older production "A" lens wouldn't fit into the new production case (by .5mm!, it turns out). OK, now THAT drove me crazy!..... but it was probably a good thing though as a permanently installable "A" lens in the "A2" lens camera would probably have distracted me from exploring the different lens setting options in mSetup and discovering that this is more of a firmware issue than an optical issue or sensor problem.

If anything is driving me crazy here it is the M2 camera which I am nearly about to throw in the towel over but that's another story for another time and place, except that there have been some parallels with my experience providing test results and getting info about both cameras.

The specific answer I got from Mobius to my initial report of the white balance problems last February when I sent screen shots and posted RAW clips to MEGA, was, "We are still trying to duplicate the M1 A2 lens issue." At the same time I asked several specific questions and asked for the specs on the different lenses I've been working with and was told I would receive a reply "early next week" but didn't. More recently (about a week ago) I asked the same questions again and provided a detailed follow-up on this problem with screen shots and short MEGA clip update but am still awaiting a reply. For some reason there has been somewhat of a reticence to fully acknowledge certain issues I report or to provide specific details when queried which is a bit frustrating but word on the street is that the developer is involved in another project at the moment and that may be part of the explanation.

In any event, I think what we need here is some kind of independent confirmation from anyone else with a new production Mobius A2 camera who may be experiencing white balance issues or color shift problems (or not). Let's see what @harsh reports.

In the meantime if anyone is interested in the short clip I referred to here is the YouTube version and a link for the RAW clip on MEGA. It was unusual in this instance because all of my previous experience with the color shifts involved a shift to an overall CYAN cast from a normal color balance whereas here it starts out CYAN (on a rainy, snowy day) shortly after starting my vehicle and then the scene abruptly shifts back to a normal color balance when my vehicle is facing in a different direction.


https://mega.nz/#!t1EkkSiK!b4MgldYjLedfJfZZ-mBvvKLzhctOW69HM5AmVK28R4g

If the developer is giving you the runaround, it is obvious his mind/whatever is somewhere else, it seems he/they have been that way for a while now. If I were you I would address him directly and find out if your assistance in helping him is being used to refine the camera to a conclusion or what? I can't see you wasting your valuable time on someone else's venture/camera without some active participation on his part to report back with how he used your findings/report.
This is like buying a cow because you desperately need the milk, but not telling the milker where the cow is:rolleyes::D!
 
Back
Top