Sony 4K Handycam FDR-AX100 thread - Page 67 - AVS | Home Theater Discussions And Reviews
Forum Jump: 
 81Likes
Reply
Thread Tools
post #1981 of 3018 Old 06-20-2014, 01:16 PM
AVS Addicted Member
 
Ken Ross's Avatar
 
Join Date: Nov 2000
Location: N.Y.
Posts: 25,900
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 2218 Post(s)
Liked: 2177
Dr. Deano, what the Gentleman above said.
Ken Ross is offline  
Sponsored Links
Advertisement
 
post #1982 of 3018 Old 06-20-2014, 01:44 PM
AVS Addicted Member
 
Ken Ross's Avatar
 
Join Date: Nov 2000
Location: N.Y.
Posts: 25,900
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 2218 Post(s)
Liked: 2177
Eugene, some really nice shots there. You did an excellent job of steadying the camera, for the most part. I know well how irritating it can be when you've carefully set up your shot, only to be jostled by the crowd.

Some of the night shots were very nice too. I only recently saw a few of those same night shots, toward the end of your video, taken by some other 4K camera (I don't recall which), but yours was more impressive in detail & color.

As for the interaction with 30p, the one shot (I admit I scanned the video since I didn't have the time to watch it through) where it showed was where you were in the small vehicle, shooting perpendicular to the direction of travel. A more oblique angle, toward the direction of travel, might have masked the 30p motion better, but I guess you probably wanted to get your shots close to the action.

One thing you might consider (just a suggestion) is to hold some of your shots a bit longer. There were a few scenes where the clip didn't appear to last more than a couple of seconds...kind of a 'bang bang'.

Looks like you had a great time.

On another note, when you look at the air there, you wonder how anyone lives past 40.
Ken Ross is offline  
post #1983 of 3018 Old 06-21-2014, 11:24 AM
Newbie
 
Join Date: Jun 2014
Posts: 1
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
Quote:
Originally Posted by ljaybronx View Post
How's this for a use of the AX100...I do videos of local opera and theater productions (small semi-pro NON-UNION and amateur/community). I can film with this camera (I have one, and have tested the theory), and if I film the production properly I can (knowing that NO ONE is viewing in 4K) crop sections from my video in a "multi-cam" format (I happen to use Premiere Pro on a PC) and to the single-cam repertory of pans and zooms (along with judicious use of the iris control in my previous camcorder of choice, the HDR-CX550V) add cuts and dissolves. Duplicating the clip and creating a "2-camera" sequence, I can make the switched-to camera whatever shot I want it to be. Avoiding certain kinds of movement at transition points (that would make things look weird that are coming from EXACTLY the same angle because they're part of the same shot) I can effectively re-block the show and emphasize what I think is important in the shot.

The result is AMAZING on a standard HD display. It is not intended for display on 4K!

I have a PC which is designed for HD editing. It has a fast (i7 4770K) processor and 16GB RAM. I also have a GeForce 770 video card with 4GB onboard (my earlier 470 had 1GB and was inadequate to display the trim editor - would go black).

Jay
Nice! I'd like to do similar but I can't figure out which 1080 preset in Premiere Pro. What sequence settings or which preset do you use? Thanks in advance!

And thanks to all you guys here for the extensive info on this camera.
fortuon is offline  
post #1984 of 3018 Old 06-21-2014, 11:56 AM
Member
 
DrDeano's Avatar
 
Join Date: Apr 2014
Posts: 15
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 14
Thanks guys!

Tested it out yesterday and -.7 seemed a little on the dark side for direct sunlight. Going to test -.5 next to see if that's any better.

I appreciate the responses!
DrDeano is offline  
post #1985 of 3018 Old 06-21-2014, 04:30 PM
Senior Member
 
chenderson2's Avatar
 
Join Date: Dec 2000
Location: Houston, TX, USA
Posts: 488
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 17
Quote:
Originally Posted by TomWheeler View Post
I, too, would like to compliment you on your Germany 4K videos. I watched them on my Asus 32 4K monitor at 60 Hz refresh rate, and I found them quite informative and enjoyable. Nicely done!

I believe that you are using Final Cut Pro X as your NLE. If so, you might want to take a good look at the various image stabilization options offered by FCP X (i.e. smoothing, inertial stabilization, tripod mode, etc.). For the kind of fairly gentle rocking motion that you have in your video clips, I believe that FCP X's image stabilization would go a long way to removing that entirely without noticeable artifacts. I find that relatively low frequency movements are removed extremely well in FCP X without artifacts. High frequency movements of the camera (such as when it is jarred on a tripod) are not removed well at all, and applying image stabilization to such clips produces a very unpleasant warping artifact.

Tom
I finally got around to trying what you suggested. Thanks. It seemed to work very well. However, I noticed that it causes the image to be magnified (zoomed in.) Does this cause a loss of resolution?
chenderson2 is offline  
post #1986 of 3018 Old 06-21-2014, 06:14 PM
Senior Member
 
TomWheeler's Avatar
 
Join Date: Aug 2004
Location: Raleigh, NC
Posts: 232
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 14
Quote:
Originally Posted by chenderson2 View Post
I finally got around to trying what you suggested. Thanks. It seemed to work very well. However, I noticed that it causes the image to be magnified (zoomed in.) Does this cause a loss of resolution?
Image stabilization always results in some zooming of the image and the loss of some of the borders of the image because this is necessary to eliminate the movement form the frame. So technically, the answer is "yes", but in practice, especially with a 4K image, I don't see the loss in resolution even when viewed on my Asus 4K monitor. Attempting to remove large amounts of motion in the clip could result in a visible loss of resolution. However, in that case other image artifacts, e.g. warping of the image. will also become quite visible and thus effectively rule out image stabilization for such cases.

One neat thing about image stabilization in FCP X is that for cases where there is only a modest amount of camera movement, it will offer you a mode called "tripod mode", and this will cause the clip to look exactly like it was made on a locked down tripod with no camera movement. This mode will only be offered for cases of modest camera movement in the clip. For cases of considerable camera movement such as in a full zoom hand-held situation, I find "inertia mode" the best choice since it leaves a bit of camera movement in the clip and produces little or no noticeable artifacts such as warping.

Tom
TomWheeler is online now  
post #1987 of 3018 Old 06-21-2014, 06:51 PM
Senior Member
 
chenderson2's Avatar
 
Join Date: Dec 2000
Location: Houston, TX, USA
Posts: 488
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 17
Yes, I tried the tripod mode and it did indeed look like it was taken while mounted on a tripod. Thanks again for suggesting it to me. I had tried such options as import and stabilize and on Youtube to let them stabilize it, and was not impressed. I am impressed with Final Cut's stabilization.
chenderson2 is offline  
post #1988 of 3018 Old 06-22-2014, 10:15 AM
Senior Member
 
TomWheeler's Avatar
 
Join Date: Aug 2004
Location: Raleigh, NC
Posts: 232
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 14
Quote:
Originally Posted by chenderson2 View Post
Yes, I tried the tripod mode and it did indeed look like it was taken while mounted on a tripod. Thanks again for suggesting it to me. I had tried such options as import and stabilize and on Youtube to let them stabilize it, and was not impressed. I am impressed with Final Cut's stabilization.
Adobe's Premiere Pro CC, which I am also running on my Mac Pro, also does a very good job of stabilizing AX100 hand-held footage. While the results in Premiere Pro and FCP X are about the same in terms of the quality of the stabilization and freedom from artifacts, Premiere Pro CC's Warp Stabilizer effect does not use the GPU's in analyzing the footage and is much much slower that FCP's image stabilization. On my late 2013 Mac Pro i can analyze and stabilize a 4K clip from the AX100 in slightly greater than real time so I don't hesitate to use FCP X's image stabilization to remove modest camera movement from a shot. In Premiere Pro using its Warp Stabilization I would be much more selective in choosing which clips to stabilize since it is a slow process.

I have not tried the image stabilization on AX100 footage in Sony's Vegas software or in Cyberlink's Power Director 12, although I do have both installed and running under Window's 8.1 on my Mac Pro. I suspect that both also do a reasonable job, and I'll check that out soon just to see how they compares in quality and speed with FCP X's image stabilization.

Tom
TomWheeler is online now  
post #1989 of 3018 Old 06-22-2014, 10:55 PM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 60 Post(s)
Liked: 57
Quote:
Originally Posted by chenderson2 View Post
I finally got around to trying what you suggested. Thanks. It seemed to work very well. However, I noticed that it causes the image to be magnified (zoomed in.) Does this cause a loss of resolution?
Active steadyshot mode reduces resolution and dynamic range vs. regular steadyshot mode. I did some tests which convinced me to never use active steadyshot mode on my AX100.
hatchback is offline  
post #1990 of 3018 Old 06-23-2014, 06:34 AM
AVS Addicted Member
 
Ken Ross's Avatar
 
Join Date: Nov 2000
Location: N.Y.
Posts: 25,900
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 2218 Post(s)
Liked: 2177
Hatch, I'll sometimes use it for the additional reach it gives rather than the additional IS.
Ken Ross is offline  
post #1991 of 3018 Old 06-23-2014, 11:56 AM
Advanced Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 585
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 160 Post(s)
Liked: 27
Hatchback

How can it reduce dynamic range. It is better to turn it off when panning for example but that is about it.

My hands are less than steady, so I use it always. Other than a change in zoom range and somewhat lower use of the sensor area I see no significant negative effect.

But I would turn it off when using a tripod.

My choice, when zooming to 8 or 12x is an unusable video or a slight loss in resolution that you need a resolution chart to detect.

Eugene

Last edited by Eugene157; 06-23-2014 at 12:23 PM.
Eugene157 is offline  
post #1992 of 3018 Old 06-23-2014, 12:09 PM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 60 Post(s)
Liked: 57
Active steadyshot reduces dynamic range and resolution by cropping the sensor. The loss is very noticeable when you're shooting a high detail scene with uneven lighting (like strong sunlight on tree branches). It's harder to notice when the scene is low detail or modest dynamic range. The extra reach of active steadyshot is nice, but the loss of image quality is unpleasant for me. I guess I made an aesthetic judgment that I'd rather have a wider shot with higher resolution and dynamic range than a tighter shot with lower resolution and dynamic range. After all, you can always crop in post if you really need to get closer.
hatchback is offline  
post #1993 of 3018 Old 06-23-2014, 01:24 PM
Advanced Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 585
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 160 Post(s)
Liked: 27
I do not accept the premise of the reduced dynamic range. It should be the same, regardless if the whole sensor is used or only a part of it. Dynamic range is established at the pixel level and is a function of the amount of light= voltage generated vs. total noise floor.

But who cares, I find the video outstanding regardless.

Gene
Eugene157 is offline  
post #1994 of 3018 Old 06-23-2014, 05:56 PM
Member
 
RobAC's Avatar
 
Join Date: Feb 2010
Posts: 128
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 20
Quote:
Originally Posted by bsprague View Post
I forgot to smile. Shot it with the "old" SDT750.

https://vimeo.com/98583397

Watch it soon! It will get deleted before anyone else sees it!

I was just about to say - Pics or it didn't happen!
But seriously that is a very cool idea.

(Btw- I am contemplating adding music and funny thought bubbles to your video - you better nuke it quick ! ... heh)
RobAC is offline  
post #1995 of 3018 Old 06-23-2014, 06:03 PM
Member
 
RobAC's Avatar
 
Join Date: Feb 2010
Posts: 128
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 20
Quote:
Originally Posted by Eugene157 View Post
Well here is what all of you have been waiting for!

The camera was in the auto mode except during the night shots in Shanghai, there the gain was reduced to 15Db. In fact the camera is set to 15Db Max.

The scene of the Chinese women doing their laundry in the river was at max, 12x tele and I used post stabilization in PD12.

The Terra Cotta soldiers are mostly screen grabs, there were signs "no tripods" but there was such a crowd that I was constantly being jostled. The camera was set at
manual focus, infinity, because there was not enough contrast in the soldiers so the focus was constantly hunting. Fortunately the subjects were holding still.

In some scenes the limited 30P shows but it is a small price to pay. In fact not a problem when considering the frame rates of commercial optical discs. Because of the
additional data rate I doubt seriously that that 60P will ever be available for consumers,either internet, optical media or TV. The only application for 60P would be movie theaters.

I feel that that the steadiness of the shots is adequate, I can live with that. Any feed back on that subject would be much appreciated. BTW I bought a monopole and felt it was not that much help
when used in the normal manner. Nor that much weight saving.

The excellent video demonstrating its use under the armpit is something I have to play with.

my 2 cents worth.

It goes w/o saying that I was very glad that I bought and was able to get the camera in time. Its PQ certainly exceeded my expectations!
It will be around for some time.

Eugene


https://www.youtube.com/watch?v=sD1l...ture=youtu..be

https://www.youtube.com/watch?v=sD1l...ture=youtu..be
Very nice Eugene.
I just did a quick scan through.

I kept rewinding and looking at that crazy taxi driver just motoring through traffic and the bicycle cab in the second vid. That was some smooth video in those scenes. I did not find them shaky at all considering they were all over the place. ($30k per square meter real estate cost? Holy crapsakes !)

Rob
RobAC is offline  
post #1996 of 3018 Old 06-23-2014, 07:07 PM
AVS Special Member
 
bsprague's Avatar
 
Join Date: Dec 2008
Location: On the Road
Posts: 3,265
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 190 Post(s)
Liked: 113
Quote:
Originally Posted by RobAC View Post
....But seriously that is a very cool idea.

(Btw- I am contemplating adding music and funny thought bubbles to your video - you better nuke it quick ! ... heh)
I fixed it so you can't add the bubbles.

Thanks. It is not only cool, but it actually works. There are shoulder rigs, gun stocks and stabilizers that are too geeky for a tourist. They attract too much attention.

I found the Sony VCT-MP1 Monopod that comes with a case that can be used as a pocket for the foot while the strap is around ones neck. It comes with a ballhead. Seemed to me the case was unnecessary but the ballhead would do a lot.

Depending on my mood and what I'm shooting the ballhead equipped monopod foot may be tucked under my arm, stuffed in a front pants pocket or tucked behind my belt buckle. It all dampens my personal brand of wiggles and jiggles.

Another inspiration came from a CheesyCam DIY Video Projects. He built a square "rig" with hand grips to hold a camcorder and accessories. In the demo videos, footage stability improved. It seemed the reason was that the two hand grips had distance between them. I tried it with the ballhead equipped monopod. Bend the ballhead to ninety degrees and hold the stick horizontal with your hands spread apart about 18 inches. Try to visualize the camera mounted to handlebars. Try some careful walking. The result is very much like having the camera on a rail slider. Between having the hands spread apart and the camera stabilization on, the clips can be quite smooth.
bsprague is online now  
post #1997 of 3018 Old 06-23-2014, 07:50 PM
Advanced Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 585
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 160 Post(s)
Liked: 27
Looks stabilization is one of the problem areas. It does not come close to the 5 axis correction of the Panasonic X900M and would be for me the only serious reason to abandon the AX100, assuming a 4K camera with at least equal PQ appears.

But I appreciate the different hints posted by forum members to improve this problem.

Thanks

Eugene

Edit: Just came across this, showing the performance of the AX100, WOW what a difference sunshine and a tripod can make.


Last edited by Eugene157; 06-23-2014 at 08:28 PM.
Eugene157 is offline  
post #1998 of 3018 Old 06-23-2014, 08:46 PM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 60 Post(s)
Liked: 57
Quote:
Originally Posted by Eugene157 View Post
I do not accept the premise of the reduced dynamic range. It should be the same, regardless if the whole sensor is used or only a part of it. Dynamic range is established at the pixel level and is a function of the amount of light= voltage generated vs. total noise floor.

But who cares, I find the video outstanding regardless.

Gene
Gene, dynamic range is established at the sensor level, based on the difference between the least light that an area of the sensor can capture and the most light that a different area of the sensor can capture in the same exposure. Cropping the sensor reduces the total light gathering ability of the sensor. It's why medium format digital cameras have such high dynamic range even though they are using outdated sensor technology. You can also see this effect by comparing the "screen" and "print" dynamic range scores for two different sensors on dxomark. According to the "screen" dynamic range scores, you can see that the Nikon D800 and D5300 have very similar per-pixel dynamic range scores. But according to the "print" dynamic range scores, you can see that the Nikon D800 has significantly higher dynamic range because it has the larger sensor.

http://www.dxomark.com/Cameras/Compa...D800___919_792

Anyway, you don't need to accept my handwaving argument. Try shooting a scene with high dynamic range using steadyshot active vs standard; you should be able to see the difference yourself.
hatchback is offline  
post #1999 of 3018 Old 06-23-2014, 11:01 PM
Advanced Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 585
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 160 Post(s)
Liked: 27
Hatchback,

You must use the same sensor, in this case the AX100.

If half the sensor is focused on a 100 white signal and the other half is black you can measure the resulting video signal and calculate the dynamic range, peak white amplitude vs black (noise).

Now if you reduce the area illuminated you will still get the SAME peak white amplitude and SAME black noise level.


The dynamic range is the difference, expressed in decibels between those two signal levels.

If it were not for optical bleed thru, you could take two adjacent pixels and have the same dynamic range by iluminating one but not the other.

Quote:"Cropping the sensor reduces the total light gathering ability of the sensor".

That would be the case for a solar panel were you want maximum TOTAL output.

But we want the PEAK output, and that is the same regardless if one or all pixels are used.

In printed pictures or video, either viewed by eye or measured, the dynamic range is the difference between max white and black level, (noise floor), regardless of their size or location in the picture or display. And that applies to
our eyes as well.

Gene










Gene

Last edited by Eugene157; 06-23-2014 at 11:40 PM.
Eugene157 is offline  
post #2000 of 3018 Old 06-23-2014, 11:05 PM
Member
 
ricksastro's Avatar
 
Join Date: Mar 2009
Location: Michigan
Posts: 63
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 11
Quote:
Originally Posted by hatchback View Post
Gene, dynamic range is established at the sensor level, based on the difference between the least light that an area of the sensor can capture and the most light that a different area of the sensor can capture in the same exposure. Cropping the sensor reduces the total light gathering ability of the sensor. It's why medium format digital cameras have such high dynamic range even though they are using outdated sensor technology. You can also see this effect by comparing the "screen" and "print" dynamic range scores for two different sensors on dxomark. According to the "screen" dynamic range scores, you can see that the Nikon D800 and D5300 have very similar per-pixel dynamic range scores. But according to the "print" dynamic range scores, you can see that the Nikon D800 has significantly higher dynamic range because it has the larger sensor.

http://www.dxomark.com/Cameras/Compa...D800___919_792

Anyway, you don't need to accept my handwaving argument. Try shooting a scene with high dynamic range using steadyshot active vs standard; you should be able to see the difference yourself.
It can be confusing, since the saturation level stays the same you could reasonably assume that the DR is the same. But since as you say you are gathering more light, the noise floor is lower, so that is where the gains in DR happen.

The extra crop of the Active IS magnifies the subject as well as the shadow noise, so you would have less leeway in pushing the shadows in post, thus less DR.
ricksastro is offline  
post #2001 of 3018 Old 06-23-2014, 11:12 PM
Member
 
ricksastro's Avatar
 
Join Date: Mar 2009
Location: Michigan
Posts: 63
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 11
Quote:
Originally Posted by Eugene157 View Post
You must use the same sensor, in this case the AX100.

If half the sensor is focused on a 100 white signal and the other half is black you can measure the resulting video signal and calculate the dynamic range, peak white amplitude vs black (noise).

Now if you reduce the area illuminated you will still get the SAME peak white amplitude and SAME black noise level.


The dynamic range is a result of those two signal levels.

If it were not for optical bleed thru, you could take two adjacent pixels and have the same dynamic range by iluminating one but not the other.

Gene
It not the black level you are concerned about, it the darkest usable levels (above the read noise/noise floor). In sampling a smaller portion of the sensor and then viewing it at the same size (ex 1080) as a video taken with the entire chip, you will be gathering more light and averaging more pixels, so you will have better noise performance in the shadows, thus more DR at the dark end of the spectrum.

You are correct that it will still clip the same at the light end.
ricksastro is offline  
post #2002 of 3018 Old 06-24-2014, 12:31 AM
Member
 
Fiat131's Avatar
 
Join Date: Feb 2011
Posts: 53
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 10
Question to more experienced members...

What causes this "unnatural" looking sky? Is there a way to control it?
ND filters? Exposure?

Is this an example of the blown up highlights (sky)?
I'm not trying to be picky or criticize either camera or the video itself, just trying to learn.
I have seen this issue pop up on SD, as well as HD cameras.

Screenshot in the attachment was taken from chenderson2 video (Germany Part 2)
Attached Thumbnails
Click image for larger version

Name:	Sky.png
Views:	33
Size:	867.1 KB
ID:	133273  

Last edited by Fiat131; 06-24-2014 at 08:42 PM.
Fiat131 is offline  
post #2003 of 3018 Old 06-24-2014, 02:44 AM
Senior Member
 
Philip_L's Avatar
 
Join Date: Dec 2010
Posts: 320
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 18
Hi

Quote:
Originally Posted by Fiat131 View Post
Question to more experience members...

What causes this "unnatural" looking sky? Is there a way to control it?
ND filters? Exposure?

Is this an example of the blown up highlights (sky)?
I'm not trying to be picky or criticize either camera or the video itself, just trying to learn.
I have seen this issue pop up on SD, as well as HD cameras.

Screenshot in the attachment was taken from chenderson2 video (Germany Part 2)
http://youtu.be/lW4lR53QFAw
Yes, the exposure has been set for the darker area of the pictures which means the sky has been blown out. Blue has gone up well over 235, red is at 235 or over and so is green, therefore you get white. Sometimes you get an odd unrealistic coloured sky if blue is blown but red and green is still < 235.

The other factor is superwhites, while in theory the maximum RGB value for video is 235,235,235, most camcorders capture values up to 255. Your sky might be there, for example Blue might be at 255, green at 235 and red at 235. Because most video players only expect values up to 235 for video, they discard everything from 236 and above then stretch 235 up to 255. If you can recover those superwhites you can bring that detail back into being visible. You would need to do this on a video editing package, sometimes just using the brightness will do this, but editors like Sony Vegas are lot of more flexible in this regards.

Regards

Phil
Philip_L is offline  
post #2004 of 3018 Old 06-24-2014, 03:17 AM
Member
 
Fiat131's Avatar
 
Join Date: Feb 2011
Posts: 53
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 10
Quote:
Originally Posted by Philip_L View Post
Hi



Yes, the exposure has been set for the darker area of the pictures which means the sky has been blown out. Blue has gone up well over 235, red is at 235 or over and so is green, therefore you get white. Sometimes you get an odd unrealistic coloured sky if blue is blown but red and green is still < 235.

The other factor is superwhites, while in theory the maximum RGB value for video is 235,235,235, most camcorders capture values up to 255. Your sky might be there, for example Blue might be at 255, green at 235 and red at 235. Because most video players only expect values up to 235 for video, they discard everything from 236 and above then stretch 235 up to 255. If you can recover those superwhites you can bring that detail back into being visible. You would need to do this on a video editing package, sometimes just using the brightness will do this, but editors like Sony Vegas are lot of more flexible in this regards.

Regards

Phil

Phil,
Thank you very much for the explanation.

Do you think it would help first (before hitting the record button) to point to the sky and slowly move the camera down towards building, while still showing the sky?
I'm thinking, if that would help, while the camera is in "Auto" mode.

Would ND filter do anything helpful in such situation?

Thank you,
Pete
Fiat131 is offline  
post #2005 of 3018 Old 06-24-2014, 03:39 AM
Senior Member
 
Philip_L's Avatar
 
Join Date: Dec 2010
Posts: 320
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 18
Hi

Quote:
Originally Posted by Fiat131 View Post
Phil,
Thank you very much for the explanation.

Do you think it would help first (before hitting the record button) to point to the sky and slowly move the camera down towards building, while still showing the sky?
I'm thinking, if that would help, while the camera is in "Auto" mode.

Would ND filter do anything helpful in such situation?

Thank you,
Pete
Possibly not, if you expose for the sky, then the darker areas would become under exposed and very dark or black. This is where dynamic range kicks in, the camera can only capture a limited extreme of dark or bright with the same exposure setting. Something has to go out of range when you have very bright and dark areas in the picture.

Remapping values above 235 gets a bit more dynamic range, but even then that may not always help.

The professional way is to use a graduated ND filter, this filter is shaded at the top and clear towards the bottom (similar to half tinted sunglasses), so the sky is darkened and bringing it into range to allow the whole frame to be captured without blowing the highlights. You often see this on some lower budget TV programs or news reports where they've added a graduated ND filter but then decide they need to pan. The shaded area becomes visible as the picture changes.

Sometimes you need a matte box to make these sorts of graduated filters work well, and they only work on shots where the frame is split cleanly across the horizontal so sky at the top and landscape at the bottom. It has to be a steady or a tripod shot as movement will make the filter become visible as the line between shaded and unshaded will not move with the image.

Other ways are to brighten the subject or the shadows using reflectors and/or lighting.

However these techniques don't lend themselves to our domestic use of camcorders, so it is case of avoiding such scenes, letting the highlights blow out, or dealing with in post.

An ordinary ND filter will not help as the whole frame is affected, the sky becomes less dark, but the shadows become darker so you still have the same difference between light and dark.

Another option that can help with blue sky at certain times of the day is a polarising filter, as this will darken the sky without darkening other parts of the image, helping to stop the sky being blown out.

Regards

Phil
Philip_L is offline  
post #2006 of 3018 Old 06-24-2014, 08:51 AM
Advanced Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 585
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 160 Post(s)
Liked: 27
Sony, at a Singapore trade show, unveiled a pro version of the FDR-AX100 that allows 10 bit 4:2:2 recording as well as XLR audio.
Delivery supposedly by the end of the year.

Gene
Eugene157 is offline  
post #2007 of 3018 Old 06-24-2014, 08:57 AM
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,472
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 260 Post(s)
Liked: 151
Quote:
Originally Posted by Eugene157 View Post
Sony, at a Singapore trade show, unveiled a pro version of the FDR-AX100 that allows 10 bit 4:2:2 recording as well as XLR audio.
Delivery supposedly by the end of the year.

Gene

Does it shoot 4K? Most people think not. Pro version of CX900 maybe?


http://www.slashcam.de/news/single/S...100-11462.html


Note there is no '4K' indication on the body shown. To shoot 4K 10-bit 4:2:2 would require substantially faster media than class-10 sd cards. That would likely require a very different media slot. Note that there is already a pro 4K Sony handheld camera, shooting 10-bit etc.
http://www.bhphotovideo.com/bnh/cont...G&Q=&A=details


Its has the smaller sensor, of course.
markr041 is online now  
post #2008 of 3018 Old 06-24-2014, 09:39 AM
AVS Addicted Member
 
Ken Ross's Avatar
 
Join Date: Nov 2000
Location: N.Y.
Posts: 25,900
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 2218 Post(s)
Liked: 2177
Quote:
Originally Posted by Fiat131 View Post
Question to more experience members...

What causes this "unnatural" looking sky? Is there a way to control it?
ND filters? Exposure?
Since we all look at videos with different monitors, many of them totally uncalibrated, it's difficult to see exactly what you're seeing.

With that said, if you scrub to that portion of the video, you can easily see how the sky changes color as he changes exposure. At the beginning of that clip, before he lowered the exposure, there was nothing odd looking about the sky on my monitor. In fact, I might have left the exposure right there. There also appeared to be a slight haze which can alter the color of both sky and objects. That happens in still photography as well. Once he reduced the exposure, the scene looked a bit 'off' in terms of color, not just the sky.

I don't believe this is a limitation of RGB values or superwhites at all. As I said, on my monitor the beginning of that clip looked fine. Only the person shooting can tell us if it was truly 'accurate'

Edit: Didn't see your screengrab, just the video. I thought you were referencing the castle shot.

So for that shot a better compromise could probably have been established with a lesser exposure. Not sure if the sky would have been totally natural given the stark contrast, but it could have been improved, I'd suspect. As with any shot, you have to determine what the subject of the clip is. If the sky is important to the scene, than either expose for that or seek a compromise. In short, you work with the dynamic range that the camera you're using offers.
Mattias Burling likes this.

Last edited by Ken Ross; 06-24-2014 at 09:49 AM.
Ken Ross is offline  
post #2009 of 3018 Old 06-24-2014, 09:59 AM
Advanced Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 585
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 160 Post(s)
Liked: 27
I wonder too.
That was from the same German website and their talk is yes. They refer to it as the AX100 not CX900. The 900 has the same 1" sensor.


Gene

Last edited by Eugene157; 06-24-2014 at 10:05 AM.
Eugene157 is offline  
post #2010 of 3018 Old 06-24-2014, 11:31 AM
Senior Member
 
Mattias Burling's Avatar
 
Join Date: Jun 2014
Posts: 466
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 180 Post(s)
Liked: 23
I would guess its possible to shoot 4K 4:2:2 to a SD class U3. I havent done the math but since regular class 10 can handle the high bitrate of the GH4 and the Raw from the bmpcc my guess is that its enough.
4:2:2 in 1080p can be as low as 50mbit so at four times the resolution....
Again, havent done the math just guessing.
Mattias Burling is online now  
Sponsored Links
Advertisement
 
Reply Camcorders

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off