Where are the ghosts!?

Hmm.... wait - actually, there are "ghosts" somewhere here too.



They aren't everywhere, but after checking this photo for the third time I think I found some - take a look at the middle-right section of the image - that street with orange lights - each one of them has a dot below. Not that everyone serious would pixel-peep photo like that but it is there.
 
Some great shots and beautiful colors :)
 
They aren't everywhere, but after checking this photo for the third time I think I found some - take a look at the middle-right section of the image - that street with orange lights - each one of them has a dot below. Not that everyone serious would pixel-peep photo like that but it is there.
I saw those too - but if the cause is translucent mirror ghosting why aren't they on other lights, some much brighter?

If these are streetlights could it be reflections from wet pavement beneath? (I'll leave it up to someone else to check if these reflections/ghosts are at 11 pixels.)

Whatever they are, reflections or ghosts, you have to hunt for them and nobody who wasn't trying to find them would notice.
  • C
 
I saw those too - but if the cause is translucent mirror ghosting why aren't they on other lights, some much brighter?

If these are streetlights could it be reflections from wet pavement beneath? (I'll leave it up to someone else to check if these reflections/ghosts are at 11 pixels.)
Well, on this photo the range between center of flash and the reflection is 12 pixels mostly, sometimes 11px. So it's roughly agreed with the theory.

They are underneath each of 18-arm stars - I guess that those are the brightest points... though on the photo it's clearly visible that non-stars objects are also very very bright, point light sources... no idea, maybe they are below the critical point ?

Also - if these are really ghosts, than there should be a gradation - on such photo with such a huge variation of light sources some of them should have less bright points than the other - all kinds of sizes and brightness should appear, as well as multi-reflected ghosts (two? three? four?). Something like that does not happen - even brightest sources, which have ghosts nearly melting with them don't have any secondary ghosts.

By the way: 18-arm stars ?! WTF? Guy had some filter on the lens? I guess he did, cause there's huge flare in right-bottom corner (roughly the size of building, orange one)
Whatever they are, reflections or ghosts, you have to hunt for them and nobody who wasn't trying to find them would notice.
Yep, true. I couldn't find them at all, looking on the image for the first time.
 
18 arm stars are a result of 9 blade aperture...

Ghosting only occurs when the point light source is so bright that it has saturated the sensor and the ghost is magnitudes less bright. In order for a second ghost to occur, the first one would have to be extremely bright, which I don't see.

It appears it is more sensitive to the red side of the spectrum. Neutrals and blue points can get them, but aren't nearly as prone.
I saw those too - but if the cause is translucent mirror ghosting why aren't they on other lights, some much brighter?

If these are streetlights could it be reflections from wet pavement beneath? (I'll leave it up to someone else to check if these reflections/ghosts are at 11 pixels.)
Well, on this photo the range between center of flash and the reflection is 12 pixels mostly, sometimes 11px. So it's roughly agreed with the theory.

They are underneath each of 18-arm stars - I guess that those are the brightest points... though on the photo it's clearly visible that non-stars objects are also very very bright, point light sources... no idea, maybe they are below the critical point ?

Also - if these are really ghosts, than there should be a gradation - on such photo with such a huge variation of light sources some of them should have less bright points than the other - all kinds of sizes and brightness should appear, as well as multi-reflected ghosts (two? three? four?). Something like that does not happen - even brightest sources, which have ghosts nearly melting with them don't have any secondary ghosts.

By the way: 18-arm stars ?! WTF? Guy had some filter on the lens? I guess he did, cause there's huge flare in right-bottom corner (roughly the size of building, orange one)
Whatever they are, reflections or ghosts, you have to hunt for them and nobody who wasn't trying to find them would notice.
Yep, true. I couldn't find them at all, looking on the image for the first time.
--
Rick Krejci
http://www.ricksastro.com
 
Once again, when you look closely at this night shot you can see flare, blooming, etc. (extend radially)

You can also see light reflections on buildings, streets, house lights spilling on the window sills, billboard lights illuminating the buildings underneath them, streetlights and their reflection, car lights, etc.

Don't confuse ghosting with natural occurring phenomas as well as lens + sensor artifacts.

Where are the ghosts that were visible in the DPR bridge shot?
 
There are some small light points under larger ones, that might is "ghosts".
But anyway, just wait some days and Dpreview.com will publish
their updated review, also including something on this ghost issue.

I must say the focus tracking in both video and 10 fps has impressed me.
Will be nice so hear some user experience with different memory cards
about write speed/time.
 
Look at the giant starbursts off all the street lights etc.

Why would the ghosts be bothersome but these huge starbursts not even be mentioned?

Seems people know you get them so jsut ignore them - or even worse - like the starbursts! OMG no! Starbursts are not real but an abberation of the lens/aperture! That is a false image! :-)

Really the ghosting issue is so tiny that it will only effect a very very few people - and they should not use an A55/33.

Just like those worthless D3x's that weigh too much to carry around some time so are fatally flawed and should not be bought by anyone. To heavy. :-)
--
tom power
 
I would suppose that you repeat the shot holding the cam in different orientation: landscape, portrait left, portrait right. If the dots are real ghosts caused by the mirror they should appear on different locations: below, left right. If this occurs then we know that they are really caused by the pellicle mirror.
 
I can't do it as long as I have to shoot with A700 !
 
Hi there!
I am new to this forum.
My eye is on the A55 and I've been doing some research before taking the plunge.

But I got my concerns on this 'ghosting' phenomena from this night shot photo from the OP.

But looking further into the picture I saw what I think is another 'aberration' but not sure about this because my eyes are not yet trained to figure out what these are.

Could anybody please explain what this is in the circled portion of the same picture above?



Thanks!

--
c3s2
 
That was one scary ghost!
No that is not the ghost issue people are talking about.
I don't know what that is. Could be something on the lens or
maybe moving city lights? I don't know, but have not seen this before
so I don't believe it is a new issue with the camera.

Dpreview.com will soon publish an updated review of A55, and then
also write something about the "ghost issue". So just wait some
days and see. I believe they wrote it should be out early this week.
 
That is not a ghost. It could have been caused by the lens or something completely different.
 
Looks like a moving light display projected onto the side of that building in the full size image
 
If you look more closely this aberration extends further up and down through the dark areas.



c3s2
 

Keyboard shortcuts

Back
Top