| Author |
N8īs Major Camera Desing Flaw - Poor Flare Performance |
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
On 2010-11-11 18:04:20, false_morel wrote:
Guys, I'm not very sure about this, but the way I see it, those excessive flares are a downside of the wide lens optics rather than that sliver ring around the lens!
Actually, physically, or geometrically, teh ring is bended away from the lens. This means it should reflect light away from the lens not vise-versa! At extreme cases, a light ray would be reflected with 0° angle, but never negative angle towards the lens..
This is how I see it..
And actually I find it really awkward for the team developing the camera to oversee this flaw if it exists!
And at the end of day, any camera, specially with wide-angle lens, will suffer from those flares at certain angles when facing a bright light source!
Of course, the question here is whether that ring is helping worsening the situation, but in my opinion, it should be rather helping improving the conditions.
And that finger-shading effect that helps remove those flares, is actually stopping excessive light from entering the lens rather than stopping it from refelcting off the ring..
About the position of the flares, it doesn't have to be on the side of the light-source, specially if the sun is that source..
Correct me if I'm wrong..
My two cents.
A Sharp Edge doesnīt reflect the light evenly.
[ This Message was edited by: Vit on 2010-11-12 15:51 ] Every Picture Tells a Story | |
|
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
On 2010-11-11 14:57:06, mlife wrote:
I believe the bad flare issue is indeed caused by the silver ring and is easy to dupliate... in the photo below you can clearly see a curved shape flare which IMO is the silver ring around the lens (my N8 is orange).
If I simply place my finger to the right of the lens (seperating it from the light) you get this...
*my finger entering the camera lens to left is only to show HOW MUCH light is reflecting right there.....
Good (and also bad) to know that I am not the only one who thinks that.
People that say itīs merely a lens issue donīt seem to know the issue that well.
Iīve had a Pixon12, which has 30mm of focal lenght. Iīve never seen such a great amount of lens flare with it.
Besides that, when we talk of critical lens flare with wide angle, we usually talk about super wide lenses. For instance, 14mm lenses.
28mm is not that wide for this lens to behave that badly.
That 'ring' has a sharp edge, and this kind of edge does not reflect the light evenly.
My guess is that it has been a designerīs solution. I donīt think it has a functional hole.
[ This Message was edited by: Vit on 2010-11-12 15:59 ] Every Picture Tells a Story |
false_morel
Joined: Feb 24, 2010
Posts: 375
PM |
Vit, could you shed some more light on your sharp edge theory?
You don't have to go all geeky like mlife and I did, illustrating everything. Just some more words would do..
But again, I may agree that the ring could worsen the flare effect only if some bright light source was within the frame of the camera! Also, at certain angles, not all the time..
Otherwise, in case of Etaab's two photos, where the sun was out of the frame, there should be no way the ring could reflect some light into the lens!
In this case, the flares there are a mere wide lens downside..
|
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
At both the bottom and the top of that surface, sharp edges may not reflect the light evenly, giving way to diffuse reflection.
If that lower part were convex and the upper part were concave, then we would guarantee that nothing would come into the lens, but due to the finishing job imprecisions that can be present, the light might not reflect evenly there, and we could not precise to where light rays are directed.
Besides that, youīve also missed something in your analysis.
First of all, youīre assuming that the inclination angles that those reflective surfaces form with any perpendicular line to the reference plane coincide with half the viewing angle, and that is mostly not true.
Secondly, you also havenīt taken the distance between that plane and the sensor into account.
Thus a light source may well be out of the range of the lens and still have an effect at the opposite side of the frame.
Take a look at these pictures:
Those two etaabīs photos may well have been affected by this brushed metal ring.
[ This Message was edited by: Vit on 2010-11-12 20:23 ] Every Picture Tells a Story |
jake20
Joined: Jan 23, 2009
Posts: > 500
From: USA
PM |
ok Vit are you a talented painter? just paint that ring black 
have you tried 30FPS video yet?
|
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
On 2010-11-12 21:24:57, jake20 wrote:
ok Vit are you a talented painter? just paint that ring black
have you tried 30FPS video yet?
I guess Iīd have to be more than talented to do that...  Not worth the risk. Itīs easier to paint my fingers black and pretend that they are a lens hood.
Regarding the video hack... I was about to ask you the same question.
Iīve read some brazilian guyīs comment at the hx page reporting that heīs bricked his phone after applying that hack. i guess Iīll wait for the official update to come out.
[ This Message was edited by: Vit on 2010-11-12 20:32 ] Every Picture Tells a Story |
jake20
Joined: Jan 23, 2009
Posts: > 500
From: USA
PM |
yea, i am going to give it a little more time to see if it comes out in the next firmware.. i hear it may be coming very soon.
http://twitter.com/Nokia_N8 |
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
On 2010-11-12 21:50:38, jake20 wrote:
yea, i am going to give it a little more time to see if it comes out in the next firmware.. i hear it may be coming very soon.
http://twitter.com/Nokia_N8
Despite this serious design flaw, I am enjoying my phone right now.
It feels rock solid, the anodyzed paint is a pleasure to touch and the capacitive screen with pich-to-zoom is just gorgeous.
The best part was the price. As itīs been made in Brazil as well, it costs 3/4 of the price of a Satio. Can you believe that?
Every Picture Tells a Story |
false_morel
Joined: Feb 24, 2010
Posts: 375
PM |
Ok, this is getting even more interesting..
Anyway, some clarifications first please:
- What does the distance between the lens and sensor have to do with anything with our case?!
- What's with the second figure you drew?
- And where exactly are those bottom and top parts of that ring where those sharp edges should be located?
- How are you referring to that metallic silver sheet with some lens shape properties? Concave and Convex? How could a metallic sheet surrounding some lens could be made concave or convex?! And at different parts too?
These points are really confusing!
Anyway, the main part here, in your first figure, how did you actually calculate teh angle of view?
Just drew randomely or approxitemaly to some calcutions you made?
As the main issue here, if the source is within the frame or just out of it, those excessive rays will come in and cause some flares at certain angles. With or without the rings!
Now, if that ring would be some major flaw if and only if it could result in some flares despite the source being far away off the frame! That is those excessive rays wont be reaching the lens directly but rather reaching that ring and reflecting in indirectly.
This is the case that I'm trying to prove wrong.
And if I'm correct, then that ring if helping produce some flares, it wont be the major factor in that case, but rather the wide lens..
Now, as to diffusion, this is a bit complicated.
I already hinted about this implicitely.
The material the ring made off, should cause diffusion not only at those edges which I'm still having trouble locating, but all over the surface.
However, in case of a diffusion, those diffused rays coming in the lens should cause visible artifacts rather than normal flares..
Besides, those rays should be weak enough that their effect would be negligible and hence unrecognizable..
But anyway, here it's a much complicated topic if we are to go in. And to be honest, I can't go that far in this aspect.
If you have some simple direct info on this, feel free to share.. |
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
- What does the distance between the lens and sensor have to do with anything with our case?!
You wrote that there was no way that a left light source could ever have an effect at the right side of the frame. The way you drew, you assumed that the sensor would be at the same place as the lens. Itīs not. So that reflected ray can perfectly reach the right side of the sensor.
- What's with the second figure you drew?
Front, top and side views. Simple like that. It requires some imagination from you. We are able to draw only 2D, but the case was 3D. Thatīs why Iīve provided 3 views.
- And where exactly are those bottom and top parts of that ring where those sharp edges should be located?
- How are you referring to that metallic silver sheet with some lens shape properties? Concave and Convex? How could a metallic sheet surrounding some lens could be made concave or convex?! And at different parts too?
Anyway, the main part here, in your first figure, how did you actually calculate teh angle of view?
Just drew randomely or approxitemaly to some calcutions you made?
Simple. Iīve done the same thing that you did when you categorically stated that the light source MUST BE WITHIN THE FRAME. In other words, I did nothing.
But when I say that half the viewing angle is mostly smaller than the angle from those surfaces is reasonable.
Those angles are equal, AT MAXIMUM.
If half the viewing angle was greater, then we would be able to actually have those rings appearing within the frame.
If it was designed to be exactly equal, in theory, then the tiniest manufacturing error would permit that a part of the ring appeared within the frame.
So it is reasonable that the viewing angle is narrower than the angle from those surfaces. Itīs simple like that.
As the main issue here, if the source is within the frame or just out of it, those excessive rays will come in and cause some flares at certain angles. With or without the rings!
I agree. Flare is an issue with any lens. Iīve never said the contrary. I said that those rings may worsen the effect.
Now, if that ring would be some major flaw if and only if it could result in some flares despite the source being far away off the frame! That is those excessive rays wont be reaching the lens directly but rather reaching that ring and reflecting in indirectly.
This is the case that I'm trying to prove wrong.
And if I'm correct, then that ring if helping produce some flares, it wont be the major factor in that case, but rather the wide lens..
Well, who knows?
Now, as to diffusion, this is a bit complicated.
I already hinted about this implicitely.
The material the ring made off, should cause diffusion not only at those edges which I'm still having trouble locating, but all over the surface.
However, in case of a diffusion, those diffused rays coming in the lens should cause visible artifacts rather than normal flares..
Besides, those rays should be weak enough that their effect would be negligible and hence unrecognizable..
But anyway, here it's a much complicated topic if we are to go in. And to be honest, I can't go that far in this aspect.
If you have some simple direct info on this, feel free to share..
Yeah.
But one thing weīve got for sure. That brushed ring should definitely not be placed near the lens.
[ This Message was edited by: Vit on 2010-11-13 13:53 ] Every Picture Tells a Story |
false_morel
Joined: Feb 24, 2010
Posts: 375
PM |
On 2010-11-13 14:49:16, Vit wrote:
You wrote that there was no way that a left light source could ever have an effect at the right side of the frame. The way you drew, you assumed that the sensor would be at the same place as the lens. Itīs not. So that reflected ray can perfectly reach the right side of the sensor.
You mean a refracted ray which hit the left area of the lens would reach teh right area of the sensor..
Very true..
I didn't say otherwise!
But this doesn't relate to our case.
What I'm suggesting, that the sunlight located just outside of the right side of etaab's photos, could have never made it to hit and bounce off the right side of the ring!
It resulted in flares caused by excessive sunlight directly hitting the the wide area of the lens and refracting towards the far left area of the sensor.. A mere wide lens downside.
To make it all clear, here are the possible cases:
- Flares caused with a light source within the frame. In theory, the flare could be worsened by the ring at certain angles, however, the flares are to happen in this case with or without the ring.
- Flares caused with a light source just out the frame. In theory, this is more of a downside of the wide angle lens. The ring could play a negative role in here, but at a very tight angle range.
- Flares caused while a light source relatively located at a wide angle with respect to the angle of view (angle > angle of inclination of the ring). Here the ring can't cause any negative effects.
Two points to highlight:
- The ring can't cause flares all by itself. It could only worsen some already unavoidable flare effects.
- The second case is the one we're debating on:
Is there a specific angle where a light source could bounce off the nearer side of the ring and into the lens while that source is out of teh frame. With your illustration, you made it clear it's possilbe. And to be honest, this is the point I missed. You are right in here. Here is another illustration where everything is thoroughly explained:
There are four different areas labled with colors, black, pink, orange, and red.
Three different light sources, located at the black, pink, and red areas.
Three angels, i the angle of inclination of the ring, D is angle difference between the angle of view and i, and L teh angle represnting the area inside teh frame from which light could bounce off the ring and hit the lens.
Angle of view is represented by the bold cut line..
Analysis:
- From the red area, no light could bounce off the ring and into the lens.
- From the orange and pink area, light could bounce off the ring and into the lens.
i should be measered, same as angle of view highlited by that bold cut line.
L + D = 45°!!!!
- From the black area, teh ring couldn't play any negative role.
Which leaves us with the following conclusion:
Regardles of how much the angle of view and angle of inclination are, there is a total of 45° of 180° in which the ring could play a negative role worsening the flare effect. Note that from that range, flares are due to happen with or without the ring.
How much of 45° degrees are outside the frame, depends on D, the angle difference between angle of view and i.
I believe it's very tight. That's why i concluded that in etaab's case, the sun was in the black area!
So it all depends on the real angle of view (which is wide on the N8 compared to other normal lens such as the Satio for instance),and the inclination angle of the ring itself.
To figure this out, we have to measure the two angles accurately and make some experiments where it would difficult to specifiy whether the flares are worsened by the ring or are all natural!
That's not to cover the diffusion issue, where everything becomes too complicated to figure out.
Front, top and side views. Simple like that. It requires some imagination from you. We are able to draw only 2D, but the case was 3D. Thatīs why Iīve provided 3 views.
I thought so. But I guess that figure turned out so poor that you could have make it without.. 
---------------------------
Ok, but do you really think this is necessary?! And what about the surface in between the two edges?!
That is, as to diffusion happening on those edges, if happening, it should be the same diffusion all over the that ring as well..
Simple. Iīve done the same thing that you did when you categorically stated that the light source MUST BE WITHIN THE FRAME. In other words, I did nothing.
But when I say that half the viewing angle is mostly smaller than the angle from those surfaces is reasonable.
Those angles are equal, AT MAXIMUM.
If half the viewing angle was greater, then we would be able to actually have those rings appearing within the frame.
If it was designed to be exactly equal, in theory, then the tiniest manufacturing error would permit that a part of the ring appeared within the frame.
So it is reasonable that the viewing angle is narrower than the angle from those surfaces. Itīs simple like that.
This is the part that made me figure out teh point I missed.
The context you used to explain teh difference between the angle of view and angle of inclination is quite confusing..
That's why I didn't get you from the first time..
Language! It would have been nice if all the world just spoke one language don't you think..
Now, if that ring would be some major flaw if and only if it could result in some flares despite the source being far away off the frame! That is those excessive rays wont be reaching the lens directly but rather reaching that ring and reflecting in indirectly.
This is the case that I'm trying to prove wrong.
And if I'm correct, then that ring if helping produce some flares, it wont be the major factor in that case, but rather the wide lens..
Well, who knows?
But one thing weīve got for sure. That brushed ring should definitely not be placed near the lens.
I think it's better that Nokia got some answers.
What is certain at this moment, is that they have to know about this, and to provide some explanatoins.
We'll see how things evolve then..
But not to be too harsh on them, this can't be considered a major flaw. As that rings can't produce some flares all by itslelf. It only worsens unavoidable ones and at a specific range of angles ( the range is 45° wide)..
Still, they better come up with a good excuse.. |
Vit
Joined: Nov 13, 2007
Posts: 279
From: Brazil
PM |
L + D = 45°!!!!
Which leaves us with the following conclusion:
Regardles of how much the angle of view and angle of inclination are, there is a total of 45° of 180° in which the ring could play a negative role worsening the flare effect. Note that from that range, flares are due to happen with or without the ring.
How much of 45° degrees are outside the frame, depends on D, the angle difference between angle of view and i.
I believe it's very tight. That's why i concluded that in etaab's case, the sun was in the black area!
L+D = 45°? Did you measure it or youīre just assuming it?
If it is indeed 45, then the total (out of 180°) would not be 45, but 90. You need to consider the other side.
Besides that, itīs not out of 180, because due to the raised area, rays emited from light sources located near the phoneīs body donīt reach the lens as well.
I would say that 2x(L+D) would be of the order of 60°. Out of 170°, perhaps. So weīve got 35% of the frame that could be affected directly by this awful finishing element. But percentage is not everything here.
Almost nobody shoots directly against a strong light source, so we shall not consider a great amount of the red area for our comparison. Thus, letīs neglect 40°, for instance, of that area. That reduces to 60° out of 130°, almost 50% of the effective frame.
We also need to consider that a great amount of shots have a light source located at the combined ORANGE + PINK area (coming from above -> street gas lamps, incandescent lamps, fluorescent lamps, sun light).
Thatīs why this issue is so critical.
Yes, that area may be tight. Nevertheless, the flare effect is critical (in the sense that it becomes unberable) at more or less one pretty tight region as well. Thatīs why I suspect that it has to do with that ring.
Itīs frustrating having to deal with this issue that frequently.
[ This Message was edited by: Vit on 2010-11-13 19:09 ] Every Picture Tells a Story |
false_morel
Joined: Feb 24, 2010
Posts: 375
PM |
On 2010-11-13 19:14:57, Vit wrote:
L+D = 45°? Did you measure it or youīre just assuming it?
It's geometry. However, I did the calculations again and figured out I had a major flaw in my calculations as well. 
Somhow I had previouslly in my calculations i fixed at 45° degrees without knowing..
Anyway, this time I'm all sure:
L + D = i!
Don't ask how I did come to this equation.. It's not that complicated, but could get really hard to explain..
Anyway, analysis of the equation:
The angle range in which teh ring could have an effect, should always be equal to i. In other words, it depends on how much the ring itself is inclined!
Now how much of the range is within teh frame and how much out of it, depends all on the angle of view.
It could be, if D is relatively big, that L tends to 0°.
This means within the frame the ring wouldn't have any effect.
Unprobable though. Still, to get all things straight, one has to measure i and to know how much the angle of view is..
If it is indeed 45, then the total (out of 180°) would not be 45, but 90. You need to consider the other side.
If we are to switch to probability here, some other aspects should be taken into consideration.
You could count both sides. But you should also know, that at most of the cases, only one light source located at only one of those areas should be there. Not two on both sides!
Besides, we have to take all four sides into account then, and also the corners..
This makes things a bit complicated.
But I think getting exact numbers wont mean a lot..
We need at this stage to know the angle range so that everyone would learn how and when to avoid those flares..
Besides that, itīs not out of 180, because due to the raised area, rays emited from light sources located near the phoneīs body donīt reach the lens as well.
True.. But lets not overdo it here...
It goes like this:
Having the sun behind or at either side of the N8 isn't a problem at all..
Now facing the sun could a porblem.. Whether sideways or head-on..
Head-on shots the ring has nothing to do with..
Side-ways could be of a small problem at certain angles..
Yes, that area may be tight. Nevertheless, the flare effect is critical (in the sense that it becomes unberable) at more or less one pretty tight region as well. Thatīs why I suspect that it has to do with that ring.
Itīs frustrating having to deal with this issue that frequently.
Yes, but I don't think the flares in etaab's photos are caused by the ring for instance..
I tend to believe that only photos suffering excessive flare effects do have the ring involved..
Then, we have N8's photos all over the net by now, including all the photos on this forum..
How many of those do suffer flare effects?!
Actually, there are many photos with the sun involved in and other night shots with some bright light sources included and all turned out excellent..
I wouldn't say this is frustrating..
A downside for sure though.. A downside Nokia better give some good excuses for.. |
etaab
Joined: Jan 23, 2004
Posts: > 500
From: UK - South Yorkshire
PM |
Alright ive been away for a couple of days and all this has happened. Its turned all Carol Vorderman !
SO, basically are we saying it is the silver rim YES or NO ?
And if the answer is yes, who is going to mod theirs first by doing something about it ??
Check me out on Instagram ! search for etaab ! |
false_morel
Joined: Feb 24, 2010
Posts: 375
PM |
On 2010-11-16 16:50:18, etaab wrote:
Alright ive been away for a couple of days and all this has happened. Its turned all Carol Vorderman !
SO, basically are we saying it is the silver rim YES or NO ?
And if the answer is yes, who is going to mod theirs first by doing something about it ??
Well, despite all that discussion and geometrical as well as optical aspects discussed at certain depth, we couldn't come up with a clear conclusion!
What's certain is that teh ring should have some effect. How radical and at what situations exactly and what angles need some more measuring and experimenting..
But surely Nokia got to have some explanations.
As I receive my N8, I shall give them some headaches. I'm good at that. | |
|
Access the forum with a mobile phone via esato.mobi
|
Posted: 2010-11-12 16:51
