Jump to content

Video shooting question


rafael_macia

Recommended Posts

Advertisement (gone after registration)

I was just shooting some video clips with a long (400 6.8) lens on my SL. The image on the monitor was normal shooting in manual. 

 But when shooting 4K  in .mov 16x9 format  Ultra HD, the monitor image changed.

Once I pressed the video button the image became washed out, and overexposed on the monitor. Changing the f Stop did not affect the washed out monitor image.

The instant I returned to FF the contrast returned.

What am I doing wrong ?

thanks

rafael

Edited by rafael_macia
Link to post
Share on other sites

Rafael, haven't you asked this question in some other threads?  Actually there appear to be several of us wondering about all the things that change when you press the LV button, and then the red video button to start recording. And that's only the start of it.  Irakly is lucky if he has an editor so that he just has to look at the footage, while someone else stitches it together and adds the music, voice-over, and effects.  Take a look at the commercial interviews with synchronized sound that Kirk Tuck is making an important part of his business, if you want to see the max that one person can get out of a mirrorless camera without a full movie crew.  He's visualsciencelab.blogspot.com.

 

To recap, first in order to see, in a still, roughly the framing that video will use, you set (on the Image page) sensor format to APS-C and Photo Aspect Ratio to 16:9.  This gives a multiplication factor of about 1.6.  My 28 mm lens is now a 45 mm-eff lens, a comfortable normal length.  

 

Next, on the second Image page, select the video format/resolution that you want.  Let's go all-out:  24 FPS, 4K gives the highest data rate, and should therefore lead to the richest image data.  Also decide whether you want MOV (for the Mac's quicktime player) or MP4 file wrapper format (for use on PCs).  Shooting in 4K also gives more pixels per frame.  You can make the frame smaller in post processing, and there are reasons that you might want to, like having smaller files to share.  

 

Under video settings, there is a choice L-Log ON or OFF.  OFF will be like shooting in JPG, the range from shadow to highlight captured roughly matches the display and the histogram on the camera in still mode will reflect it.  But in video, compressing shadows up, highlights down and desaturating color (what L-Log and similar profiles on other cameras do) allows greater than 8 bits of dynamic range to be recaptured in post processing.  Only problem -- it looks awful when immediately reviewed.  As several have learned, it is best to start with settings that for still photography would be 1 to 2 stops underexposed, so that the Log transform in pushing up the shadows does not lose highlight detail.  In an editor, such as Premiere (if you have the full CC subscription) or FinalCutProX or DaVinci Resolve, you will resaturate the colors, pull down the shadows and raise the highlights to taste, using the funny horizontally spread histograms that film people use as a guide, and things will look great again.  Or use a canned log profile, such as Sony3LogCinema to get a reasonable starting point.

 

The other problem area is control of shutter speed, aperture, ISO and exposure offset while shooting video.  In still mode, you can see what you are doing on the camera display.  In video, not so much.  First, the shutter speed (not shown) can't go below the frame rate, and film purists would say that it can't be much faster or you will have jerky, pixilated motion.  Changing anything while shooting video is hard, so it requires planning for the camera to pan or move.  In A mode if you set up for the darkest part of the shot, the shutter speed will increase as the scene gets lighter, which may be bad.  So in Manual mode, leave Auto ISO on and set the camera for low shutter speed, and high enough aperture number that the brightest parts of the scene will use ISO 50 or ISO 200 (depending on what your belief of the SL's base ISO is).  Then increased ISO will compensate for the darker parts.  Since there is no EXIF in video files, you will never know what the camera chose to do unless you read off the information from the top plate LCD out loud, so that it is recorded on the sound track.  That's obviously only a technique to use while doing experiments to find out how to get things the way you want them.

 

OK, now go out and find something appealing that moves, brace yourself so that the camera doesn't move too much, and shoot away.

 

scott

  • Like 2
Link to post
Share on other sites

I took my own advice this morning.  I was heading up to the art museum to see a small photo exhibit of war and family pictures by Micha Bar-Am, carrying a new Olympus M1.2.  I encountered streams of people coming from the various exhibits, cranked the mode dial around to the movie camera image, found a comfortable position and just shot for a while.  This unit has in body stabilization, optical in-lens stabilization, and they work together.  The result is impressive, especially after trying to get steady shots with the SL and manual lenses.

 

I posted a sequence at https://drive.google.com/file/d/0B_DOXbFeLaWiX3JzQXZoZXFuOUU/view?usp=sharing .

 

The E-M1.2 doesn't yet have a log codec, and I have no idea what my camera settings were, but people did interesting stuff in the clip I saved (like the family activities in the GC Station clip someone posted here a few days ago).  It reminded me that video is really about attractive and connected movement, interesting actions, and story-telling.  Color correction and dynamic range are secondary attributes.

 

scott

Link to post
Share on other sites

Rafael, haven't you asked this question in some other threads?  Actually there appear to be several of us wondering about all the things that change when you press the LV button, and then the red video button to start recording. And that's only the start of it.  Irakly is lucky if he has an editor so that he just has to look at the footage, while someone else stitches it together and adds the music, voice-over, and effects.  Take a look at the commercial interviews with synchronized sound that Kirk Tuck is making an important part of his business, if you want to see the max that one person can get out of a mirrorless camera without a full movie crew.  He's visualsciencelab.blogspot.com.

 

To recap, first in order to see, in a still, roughly the framing that video will use, you set (on the Image page) sensor format to APS-C and Photo Aspect Ratio to 16:9.  This gives a multiplication factor of about 1.6.  My 28 mm lens is now a 45 mm-eff lens, a comfortable normal length.  

 

Next, on the second Image page, select the video format/resolution that you want.  Let's go all-out:  24 FPS, 4K gives the highest data rate, and should therefore lead to the richest image data.  Also decide whether you want MOV (for the Mac's quicktime player) or MP4 file wrapper format (for use on PCs).  Shooting in 4K also gives more pixels per frame.  You can make the frame smaller in post processing, and there are reasons that you might want to, like having smaller files to share.  

 

Under video settings, there is a choice L-Log ON or OFF.  OFF will be like shooting in JPG, the range from shadow to highlight captured roughly matches the display and the histogram on the camera in still mode will reflect it.  But in video, compressing shadows up, highlights down and desaturating color (what L-Log and similar profiles on other cameras do) allows greater than 8 bits of dynamic range to be recaptured in post processing.  Only problem -- it looks awful when immediately reviewed.  As several have learned, it is best to start with settings that for still photography would be 1 to 2 stops underexposed, so that the Log transform in pushing up the shadows does not lose highlight detail.  In an editor, such as Premiere (if you have the full CC subscription) or FinalCutProX or DaVinci Resolve, you will resaturate the colors, pull down the shadows and raise the highlights to taste, using the funny horizontally spread histograms that film people use as a guide, and things will look great again.  Or use a canned log profile, such as Sony3LogCinema to get a reasonable starting point.

 

The other problem area is control of shutter speed, aperture, ISO and exposure offset while shooting video.  In still mode, you can see what you are doing on the camera display.  In video, not so much.  First, the shutter speed (not shown) can't go below the frame rate, and film purists would say that it can't be much faster or you will have jerky, pixilated motion.  Changing anything while shooting video is hard, so it requires planning for the camera to pan or move.  In A mode if you set up for the darkest part of the shot, the shutter speed will increase as the scene gets lighter, which may be bad.  So in Manual mode, leave Auto ISO on and set the camera for low shutter speed, and high enough aperture number that the brightest parts of the scene will use ISO 50 or ISO 200 (depending on what your belief of the SL's base ISO is).  Then increased ISO will compensate for the darker parts.  Since there is no EXIF in video files, you will never know what the camera chose to do unless you read off the information from the top plate LCD out loud, so that it is recorded on the sound track.  That's obviously only a technique to use while doing experiments to find out how to get things the way you want them.

 

OK, now go out and find something appealing that moves, brace yourself so that the camera doesn't move too much, and shoot away.

 

scott

 

No, in another thread I mentioned having problems with LOG into FCPX then DaVinci, finally Exporting.......similar questions bur not the same.

 

It was me who posted the Grand Central Terminal shot, 4K of the information booth.

Thank you so much for the informative and detailed message. Pos too that I have some evolved through, and now my problem of the moment is group applying LUTs and exporting finished work in DaVinci Resolve Studio. I think I was just going too fast and not waiting for the rendering part to finish.

Anyway having fun!

Edited by rafael_macia
Link to post
Share on other sites

I took my own advice this morning.  I was heading up to the art museum to see a small photo exhibit of war and family pictures by Micha Bar-Am, carrying a new Olympus M1.2.  I encountered streams of people coming from the various exhibits, cranked the mode dial around to the movie camera image, found a comfortable position and just shot for a while.  This unit has in body stabilization, optical in-lens stabilization, and they work together.  The result is impressive, especially after trying to get steady shots with the SL and manual lenses.

 

I posted a sequence at https://drive.google.com/file/d/0B_DOXbFeLaWiX3JzQXZoZXFuOUU/view?usp=sharing .

 

The E-M1.2 doesn't yet have a log codec, and I have no idea what my camera settings were, but people did interesting stuff in the clip I saved (like the family activities in the GC Station clip someone posted here a few days ago).  It reminded me that video is really about attractive and connected movement, interesting actions, and story-telling.  Color correction and dynamic range are secondary attributes.

 

scott

That clip is really funny - people walking past unconcernedly carrying huge DSLR telezoom systems, people with cell phones getting all excited and taking photographs. Kind of sums up the state of photography nowadays.

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.
Note: Your post will require moderator approval before it will be visible.

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.

×
×
  • Create New...