Mobius Telephoto Dashcam

[Deleted member 37895]

Here's what the Public Lab 8mm looks like pointing straight forward

 

[TonyM]

Mobius 1 with 4mm F1.5 and 4mm F1.2 lenses

Initial low light comparison between these two similar lenses, mounted on identical cameras with the same settings

I'm happy with the focus on the F1.2. The F1.5 does not seem quite as sharp as the F1.2 anywhere in the frame, either close up or far away. However there's not much difference between them in terms of detail or motion blur.

F1.2


F1.5


Lens D


F1.2


F1.5


Lens D

 

[Dashmellow]

I'm happy with the focus on the F1.2. The F1.5 does not seem quite as sharp as the F1.2 anywhere in the frame, either close up or far away. However there's not much difference between them in terms of detail or motion blur.

Speaking of sharpness, I've been testing out the varifocal 2.8-12mm ƒ/1.4 since installing the IR-cut filter and with or without the IR-cut filter I am continuously impressed by the acuity and sharpness this lens is capable of compared with the 4mm ƒ/1.2, the 6mm ƒ/1.2 or any of these other lenses we've all been experimenting with. This lens keeps producing images that have a certain clarity, presence and fine detail the other lenses just can't seem to touch. Photographers describe images like this as having "snap".

This morning I did a quick focusing job and ended up with the focus set a bit closer than it should have been so the farther away traffic footage wasn't so good but here are a few examples of what I ended up with closer in.
Here the varifocal lens is set to its maximum magnification of 12mm.



Note that you can easily read the expiration date on the registration sticker on the lower right hand corner of the license plate, especially if you view this image full size. The other telephoto lenses can't seem to resolve small details quite this well.


I found myself behind an elderly out-of-town tourist couple who drove off not realizing their rear hatch was open. I tried to get their attention by beeping my horn, flashing my lights and gesticulating wildly but they ignored me because they seemed to think they were being followed by some crazy man. Suddenly, a teenager came running up and slammed their hatch shut without saying a word to them and that really freaked them out. I thought they were both going to have heart attacks. Finally, they realized what had just happened and gave a grateful, friendly wave. I think they were relieved to discover they hadn't ventured into a town full of manics after all.

 

[Deleted member 37895]

That's Bernie and Jane Sanders

 

[Dashmellow]

That's Bernie and Jane Sanders

Nope, these guys had Connecticut plates. I know Bernie actually and that ain't him.

 

[Radius8]

From all of you who have done extensive testing on focus, is there a recommended "most useful" focus distance for dash cam applications? I realize that that depends on the lens, but say for the popular F1.2, 4mm for example.

I've installed my F1.2 the back camera because for the front it was a bit too narrow as compared to the stock A lens. In my case image quality was not the highest priority since the optical path goes through 3 layers of glass. Rather I needed something a bit more narrow to actually capture an useful image, other than just my engine bay (yes, the engine is in the back in that car ).

 

[Harsh]

I focussed the 4mm, 8mm and 12mm looking at a red brick wall about 20-30 metres away. A railing that comes into the view at 5 metres and other buildings + trees helped...

 

[Deleted member 37895]

From all of you who have done extensive testing on focus, is there a recommended "most useful" focus distance for dash cam applications? I realize that that depends on the lens, but say for the popular F1.2, 4mm for example.

I've installed my F1.2 the back camera because for the front it was a bit too narrow as compared to the stock A lens. In my case image quality was not the highest priority since the optical path goes through 3 layers of glass. Rather I needed something a bit more narrow to actually capture an useful image, other than just my engine bay (yes, the engine is in the back in that car ).

In landscape photography you generally try to focus a bit closer than infinity just at the edge of where distant objects are in focus. If you focus at infinity perfecty there's no benefit because the sensor or film can't resolve that much information anyway. I guess the point here is to capture license plates, so you have to choose a point where a plate is distant but readable in the video. Then when it gets closer even if it's out of focus slightly it should still be readable just because it's bigger in the frame

 

[Radius8]

Ah, that makes sense . Thank you for the great suggestion.

 

[Dashmellow]

From all of you who have done extensive testing on focus, is there a recommended "most useful" focus distance for dash cam applications? I realize that that depends on the lens, but say for the popular F1.2, 4mm for example.

Each lens is indeed different and these faster ƒ/1.2 & ƒ/1.4 lenses are especially challenging to focus for dash cam purposes due to their short depth of field. These very fast lenses are different than lenses that can focus well out to infinity from a point of near focus.

The trick is to choose the optimal hyperfocal point. Hyperfocal focusing is based on the fact that depth of field (DOF) typically extends 2/3 behind the point of focus and 1/3 in front.
Moving the hyper-focal distance forward or backward will provide you with an optimal zone of focus (DOF) for your needs and preferences. With these lenses it usually requires some trial and error until you find the sweet spot, especially for dash cam use.

 

[TonyM]

4mm F1.5 vs 4mm F1.2

EDIT: Post move to different thread: https://dashcamtalk.com/forum/threa...ecording-with-4mm-f1-2-lens.31145/post-393198

 

[TonyM]

4mm F1.5 vs 4mm F1.2

EDIT: Post moved to different thread: https://dashcamtalk.com/forum/threa...ecording-with-4mm-f1-2-lens.31145/post-393199

 

[kamkar]

Haha Harsh, your reverse video totally freaked me out, that white car pulling in behind you i was like OMG that worman are not looking where she are going, that is until i remembered to look at the "wrong" side of the car for the driver.

And not to forget at 2:16 the moped driver going the wrong way up a 3 lane road

 

[Dashmellow]

@TonyM, the two lenses look quite similar, even at speed. If anything the plate numbers on the ƒ/1.2 seem a bit sharper to my eye. I'll need to look more carefully at the videos when I have time as my crappy bandwidth makes viewing YouTube a bit of a hassle.

I did happen to notice a CYAN color shift after a RED object enters the FOV on the ƒ/1.5 lens similar to what I experienced with the A2 lens on the Mobius. Observe the ƒ/1.5 video carefully at 38 t0 40 seconds just after the two red cars go out of view. The effect is subtle and doesn't appear quite as serious as I encountered with the A2 lens Mobius but it's definitely there and doesn't really seem to appear on the ƒ/1.2 lens video.

 

[TonyM]

@TonyM, the two lenses look quite similar, even at speed. If anything the plate numbers on the ƒ/1.2 seem a bit sharper to my eye. I'll need to look more carefully at the videos when I have time as my crappy bandwidth makes viewing YouTube a bit of a hassle.

I agree the F1.2 appears slightly sharper. But not by much.

I did happen to notice a CYAN color shift after a RED object enters the FOV on the ƒ/1.5 lens similar to what I experienced with the A2 lens on the Mobius. Observe the ƒ/1.5 video carefully at 38 t0 40 seconds just after the two red cars go out of view. The effect is subtle and doesn't appear quite as serious as I encountered with the A2 lens Mobius but it's definitely there and doesn't really seem to appear on the ƒ/1.2 lens video.

The clips from both cameras start with a cyan shift. The F1.5 goes back to normal at around 27s, then shifts cyan again as you noticed around 38s. Both cameras shift back to normal at around 55s.

 

[TonyM]

One of the rattling noises you can hear is my A119 Until reviewing the A119 video to compare to these lenses, I didn't realise it was bouncing around so much on its GPS mount.
EDIT - Adding an EVA pad to the mount has fixed the vibration

 

[Dashmellow]

I agree the F1.2 appears slightly sharper. But not by much.


The clips from both cameras start with a cyan shift. The F1.5 goes back to normal at around 27s, then shifts cyan again as you noticed around 38s. Both cameras shift back to normal at around 55s.

Which default lens setting are you using? The "A" setting tends more towards blue and cyan than the others which willl be warmer.

 

[TonyM]

Which default lens setting are you using? The "A" setting tends more towards blue and cyan than the others which willl be warmer.

I was just about to say that I'm using profile A for all my lenses, for the sake of comparison. I had noticed that these 4mm lenses were cooler than the stock A and A2 lenses. My 6mm F1.2 is also cool, whilst the 25mm f2.4 that I tried is much warmer.

 

[Dashmellow]

I was just about to say that I'm using profile A for all my lenses, for the sake of comparison. I had noticed that these 4mm lenses were cooler than the stock A and A2 lenses. My 6mm F1.2 is also cool, whilst the 25mm f2.4 that I tried is much warmer.

The nice thing with the Mobius is that one can adjust for these lens color characteristic quirks. Dash cams and actions cams are generally color balanced in the firmware for the lens that happens to ship with it.

There clearly seems to be a quirky color shift sensitivity with the Mobius lens module depending upon which optic is installed. Unfortunately, when I tried to bring the issue to Mobius' attention and provided such irrefutable evidence of it they declined to even acknowledge the problem.

 

[TonyM]

I've moved my 4mm F1.2 & F1.5 comparisons over to this thread: https://dashcamtalk.com/forum/threa...ht-recording-with-4mm-f1-2-lens.31145/page-17