Jump to content




Photo

Advice about "halo" needed


  • Please log in to reply
6 replies to this topic

#1 David Cunningham

David Cunningham
  • Basic Members
  • PipPipPipPip
  • 978 posts
  • Cinematographer

Posted 12 December 2014 - 09:57 AM

Hi All,

 

Here is some Super16 E100D footage shot with a Kern Switar 25mm AR lens on a Super16 converted Eclair ACL.

 

There is noticeable "halo" around anything bright white.  Strong examples at 1:24 and 1:44.

 

My question is, do you think this is caused by:

 

1.) The scan - note it's around the perfs too...  although that may be a different halo

 

2.) Chromatic Aberrations in the lens - I originally thought this was it, but then realized I usually at least had it stopped down to f8.  So, I figure that is unlikely.

 

3.) Lack of anti-halation layer on E100D (leaning towards this).

 

 

Also at 1:24 notice that you cannot read any of the text on the bouncy-house.  This is a 4K scan.  You cannot read it even on the original.  Do you think this is:

 

1.) Lack of resolution of the E100D film

 

2.) Lack of resolution/mtf of the lens

 

3.) Imperfect focus of the lens (I don't think this is it as it's a 25mm lens, about 25 ft away, at about f8).

 

Thanks for your input!

 

 

Dave

 


  • 0




#2 David Cunningham

David Cunningham
  • Basic Members
  • PipPipPipPip
  • 978 posts
  • Cinematographer

Posted 15 December 2014 - 09:12 PM

So, FYI that I answered a bunch of my own questions by re-watching this footage a few times on my trusty Kodak Pageant projector...

 

The hallos are not there.  So, it's something in my video.  Working with Premeire and the original Prores 4K 4444 files now to try and get rid of it.  But, the good thing is:

 

1.) It's not on the film

 

2.) It's not caused by the film (no anti-halation, etc)

 

3.) It's not the Switar lens.

 

Also, the text in question at 1:24... I can read when projected.

 

So... just not high-enough res a scan, I guess.


  • 0

#3 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 16 December 2014 - 12:16 AM

If it's a 4K scan of Super-16 and the grain is in focus, the image isn't going to get any sharper -- detail in film exists on grains. Super-16 does not contain 4K worth of information so a 4K scan should be sufficient. Have you looked at the individual 4K frames as opposed to playing it back as video? I assume you aren't watching this in 4K motion on a 4K monitor.
  • 0

#4 Dirk DeJonghe

Dirk DeJonghe
  • Basic Members
  • PipPipPipPip
  • 543 posts
  • Industry Rep
  • Kortrijk,Belgium

Posted 16 December 2014 - 06:07 AM

If you scan reversal film you have a large dynamic range to cope with; Pushing the whites up will cause flare in the system (a bit in the lens, a bit in the prism, some in the sensor). For example the perforations have the highest possible white level, no film density at all. The scan will be a lot easier from negative film since everything is more compressed by the film gamma of 0.55.

 

Slightly overexposing the reversal stock may also help.


  • 0

#5 Chris Burke

Chris Burke
  • Basic Members
  • PipPipPipPip
  • 1517 posts
  • Boston, MA

Posted 16 December 2014 - 07:21 AM

If you scan reversal film you have a large dynamic range to cope with; Pushing the whites up will cause flare in the system (a bit in the lens, a bit in the prism, some in the sensor). For example the perforations have the highest possible white level, no film density at all. The scan will be a lot easier from negative film since everything is more compressed by the film gamma of 0.55.

 

Slightly overexposing the reversal stock may also help.

Slightly over?? Huh. This goes against my rule of thumb to expose reversal slightly under,  but you learn something every day. Is this a better method specifically for scanning?  shadows do get crunched pretty quick with reversal don't they? Is this to protect the shadows or bring detail into them? And can modern scanners handle all that highlight that overblown reversal might give them?


  • 0

#6 Dirk DeJonghe

Dirk DeJonghe
  • Basic Members
  • PipPipPipPip
  • 543 posts
  • Industry Rep
  • Kortrijk,Belgium

Posted 17 December 2014 - 05:37 AM

You need to adjust the light on telecine/scanner to accommodate shadow detail. If you need to give much light the highlights may suffer or halo may appear (sensor overload, lens or prism flare).
What really matters is the density difference between the brightest and darkest parts of the image.
Low contrast reversal 7252 is much easier to scan. Negatives or IP are low contrast by design.
You want rich shadow details with sufficient detail in the highlights at the same time.
  • 0

#7 David Cunningham

David Cunningham
  • Basic Members
  • PipPipPipPip
  • 978 posts
  • Cinematographer

Posted 17 December 2014 - 06:23 AM

I was actually able to use RGB curves in Premeire pro to time back most of the halo. Looking much better now.
  • 0


CineLab

Abel Cine

Glidecam

Rig Wheels Passport

Zylight

Ritter Battery

Technodolly

The Slider

Visual Products

Willys Widgets

rebotnix Technologies

CineTape

Paralinx LLC

Tai Audio

Aerial Filmworks

Pro 8mm

Broadcast Solutions Inc

Broadcast Solutions Inc

Aerial Filmworks

CineTape

Rig Wheels Passport

Willys Widgets

Paralinx LLC

The Slider

Ritter Battery

Abel Cine

CineLab

Tai Audio

Zylight

Technodolly

Visual Products

rebotnix Technologies

Pro 8mm

Glidecam