Jump to content
Bassem

Strange "box of light" in photo from X-Pro 2?

Recommended Posts

Bassem

Hi everyone,

Shooting a wedding yesterday and noticed this image on the back of my screen.  Do you see you the square box of light on the bride's dress? At first I thought maybe it was just a processing error on the camera only effecting the JPEG's but after bringing the raw file into LR realized it has effected that file as well.  No other images contain the same "light box", Maybe it was caused by a guest's flash but wondering how it only appeared in this unique shape.  Has anyone every seen or experienced this? 

 

XP012479-export.jpg

Share this post


Link to post
Share on other sites
Owtafocus

Can't explain it from here and my XPro2 has not done that (yet).  Looks too linear to be a light shaft thru a window.  Are other frames from this sequence the same?  Were there alien ships hovering above?

Share this post


Link to post
Share on other sites
raagamuffin

This is interesting, are you thinking its a camera fault?

To my amateur eyes, it looks like a projector was turned on, I know some of the consumer Sony video cameras have a projector mode (but the person would have to be fairly close for such a small light box).

Of course, it may be something else entirely. 

Share this post


Link to post
Share on other sites
Bassem

Thanks for your replies.  No this was the only frame from the sequence, actually this is the first time this has ever happened.  Not too sure what you mean by video camera "projector mode" but this seems like it's in camera.  Curious if anyone else has had something similar happen?

Share this post


Link to post
Share on other sites
artuk

I don't think its a camera fault.

look at the elderly male guest behind the brides dress where it is lighter - you can see him through the gauze of the dress and he is normal.

its something lighting her, and my guess is it's flash. Were you using electronic shutter?

when the data is read off the sensor it is done row by row, you can can get strange effects with non continuous light sources. If you were using electronic shutter then the sensor read out could have caught a fraction of someone else's flash on only some rows of pixels as the sensor is read out row by row

  • Like 2

Share this post


Link to post
Share on other sites
veejaycee
5 hours ago, artuk said:

I don't think its a camera fault.

look at the elderly male guest behind the brides dress where it is lighter - you can see him through the gauze of the dress and he is normal.

its something lighting her, and my guess is it's flash. Were you using electronic shutter?

when the data is read off the sensor it is done row by row, you can can get strange effects with non continuous light sources. If you were using electronic shutter then the sensor read out could have caught a fraction of someone else's flash on only some rows of pixels as the sensor is read out row by row

I can't think of a better explanation - I couldn't even think of this one. I'd say good catch artuk.

It is plain it is not a shaft of light from some aperture in the building because the light edges are linearly parallel.

Share this post


Link to post
Share on other sites
Bassem
6 hours ago, artuk said:

I don't think its a camera fault.

look at the elderly male guest behind the brides dress where it is lighter - you can see him through the gauze of the dress and he is normal.

its something lighting her, and my guess is it's flash. Were you using electronic shutter?

when the data is read off the sensor it is done row by row, you can can get strange effects with non continuous light sources. If you were using electronic shutter then the sensor read out could have caught a fraction of someone else's flash on only some rows of pixels as the sensor is read out row by row

Thank you Artuk for the explanation, yes I was using electronic shutter.  If someone did fire their flash at that moment how did it create a "box of light"? I would have expected the light to be in a form of a horizontal row across the whole image, am I wrong?  Thank you again for your reply!

Share this post


Link to post
Share on other sites
artuk
33 minutes ago, Bassem said:

Thank you Artuk for the explanation, yes I was using electronic shutter.  If someone did fire their flash at that moment how did it create a "box of light"? I would have expected the light to be in a form of a horizontal row across the whole image, am I wrong?  Thank you again for your reply!

I am guessing here but I *think* the sensor read out is along the width of the sensor (so read out from successive horizontal lines in your photo). Therefore the flash was present as the row was read at the top of the light patch and no longer present below the last row at the bottom of the light patch. I think the portion of flash captured came from behind the bride and was blocked by her - hence nothing on the left of this picture (*on those rows*). There may have been flash illumination to the right of the frame in other areas, but these were not captured by the sensor read out from the upper and lower parts of the frame.

it I important to remember that fully electronic shutter only offers "fast read out" of the data, row by row. Electronic "global" shutters read the entire sensor in one go, but Sony has yet to offer such architecture on any aps-c (or, I think, full frame) commercially available sensor used in stills cameras (nor has Canon, Samsung, Kodak etc). Therefore when using fully electronic (silent) shutter or electronic front curtain shutter, you must always be aware of its characteristics with flash, non continuous lighting (fluorescent, I think sodium etc) where the fast flickering of light can cause strange effects. The same is also true of motion, which can cause very strange effects and results in the "jelllo effect" (pardon my americanese) when shooting video or moving objects that causes straight lines to bend (e.g. Propellers on planes become bent).

overall its often much safer to use fully mechanical shutter. On Sony cameras I believe there is also a restriction on fully electronic shutter when using flash for the reasons outlined here (I *think*, I would need to check as EFCS is certainly allowed), and I assume Fuji would implement a similar restriction.

Edited by artuk
typos
  • Like 2

Share this post


Link to post
Share on other sites
Bassem

Hi Artuk that makes complete sense.  Culling through the wedding now and I found another photo with a similar situation with the guest firing off his flash.  Thank you again!  

 

Share this post


Link to post
Share on other sites
Christopher

Yes, definitely a guest flash causing that issue. It happens to me from time to time. 

Share this post


Link to post
Share on other sites
dem

This "flash" seems to be rather slow.

A flash pulse would normally last for about 1/5000 s and would cause a bright line on the image of just a few pixels wide.

Here the "box of light" is taking about a fifth of the image height. The electronic shutter on the XP2 takes about 1/22 s to scan the whole image, so the bright light was on for about 1/100s - that is about 50 times longer than a typical off-camera flash.

Maybe phones that use LEDs for lighting have a flash pulse that is that long, but I do not think this was a camera flash.

I think 1/100 s is too slow for "Autofocus Assist light" that is also often responsible for ruining other people's photos.

Share this post


Link to post
Share on other sites
veejaycee
3 hours ago, dem said:

This "flash" seems to be rather slow.

A flash pulse would normally last for about 1/5000 s and would cause a bright line on the image of just a few pixels wide.

Here the "box of light" is taking about a fifth of the image height. The electronic shutter on the XP2 takes about 1/22 s to scan the whole image, so the bright light was on for about 1/100s - that is about 50 times longer than a typical off-camera flash.

Maybe phones that use LEDs for lighting have a flash pulse that is that long, but I do not think this was a camera flash.

I think 1/100 s is too slow for "Autofocus Assist light" that is also often responsible for ruining other people's photos.

Since you quote 1/100 sec and LEDs as an example perhaps this experiment using those same parameters and showing results from pulsing (@ 1/100th) LEDs over a complete frame using electronic shutter on X100T may throw some light (excuse the pun) on the matter. It seems to me that as Bassem's shot received only a single pulse of (poss LED) flash that Artuk's explanation is likely to be correct.

It's worth a read anyway.

http://janssico.com/2014/12/x100ts-electronic-shutter-speed-analysis/

Perhaps the OP could post the exif for the image.

Share this post


Link to post
Share on other sites
raagamuffin

I have never used the electronic shutter on my camera (knowingly anyway). Are there other advantages using it besides the quiet click and use in very bright conditions? 

Share this post


Link to post
Share on other sites
artuk
5 hours ago, raagamuffin said:

I have never used the electronic shutter on my camera (knowingly anyway). Are there other advantages using it besides the quiet click and use in very bright conditions? 

You've summed up the main benefits:

- silent

- no cycle time required to start the mechanical shutter (close the shutter, clear down the live view data from the sensor then re-open it to start: the effect is generally sharpened shutter response

- no vibration

- in Fuji cameras, allows access to shutter speeds greater than 1/4000s of the mechanical shutter

Share this post


Link to post
Share on other sites
dem
On 19/07/2016 at 5:22 PM, veejaycee said:

Since you quote 1/100 sec and LEDs as an example perhaps this experiment using those same parameters...

Haha. Small world. This is the same link I posted on this forum about a year ago.

Surely it is some kind of "flash" used by one of the guests, but similar to the OP, I was surprised by how wide the light box was.

Well, there seem to be "mystery ebay" flashes with flash duration of about 1/100 sec even at 1/32 power setting:

http://gock.net/2012/01/flash-durations-small-strobes/

Share this post


Link to post
Share on other sites
veejaycee

The duration of flash from a reduced power output setting of 1/32 is actually shorter than at full power.

The "light box" is thus created by a combination of those pixels that were actuated by the electronic shutter coinciding with the ultra short period of the flash duration. Am I wrong in this thinking?

Share this post


Link to post
Share on other sites
artuk
On ‎19‎/‎07‎/‎2016 at 2:16 PM, dem said:

This "flash" seems to be rather slow.

A flash pulse would normally last for about 1/5000 s and would cause a bright line on the image of just a few pixels wide.

Here the "box of light" is taking about a fifth of the image height. The electronic shutter on the XP2 takes about 1/22 s to scan the whole image, so the bright light was on for about 1/100s - that is about 50 times longer than a typical off-camera flash.

Maybe phones that use LEDs for lighting have a flash pulse that is that long, but I do not think this was a camera flash.

I think 1/100 s is too slow for "Autofocus Assist light" that is also often responsible for ruining other people's photos.

the effect is a function of the sensor readout speed, you cant really judge the duration of the flash as you don't know the sensor readout speed. The flash happens to be "on" when those rows were read off the sensor, which is nothing to do with shutter speed.

edit: I dint think the read out speed is 1/22s as that implies you can't read more than 22 frames a second, which is rather slow for video or even the evf.

Share this post


Link to post
Share on other sites
dem
17 hours ago, artuk said:

I dint think the read out speed is 1/22s as that implies you can't read more than 22 frames a second, which is rather slow for video or even the evf.

Well, the ES read out time for the X-T1 and the X100T is 1/15 sec (compare to 1/14 sec for Sony alpha 7 Mark II) and that did not stop any of these cameras from being able to shoot video.

One does not need an oscilloscope to check the ES read out time on their camera. Just take a photo at a high shutter speed under fluorescent light and count the bands. With the X-T1 you will get 6.9 bands in Europe (50 Hz) and 8.3 bands in the US (60 Hz), which means the ES read out time is 6.9/100 = 8.3/120 = 1/14.5 sec.

I do not have access to the X-Pro2, so was relying on other people reporting about 4.5 bands under 50 Hz lighting. That is about 1/22 sec ES read out time.

Share this post


Link to post
Share on other sites
artuk

I'm not going to get into some stupid argument about this, but if the sensor really did take 1/22s to read out the data, the continuous shooting could not support 10/11 frames per second, as the absolute limit would be about 4.5 fps

so whatever those measurement's are, there are clearly other factors at play which makes me question their validity.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recommended Discussions



×
×
  • Create New...