Sony 4K Handycam FDR-AX100 thread - Page 81 - AVS Forum
Forum Jump: 
 36Likes
Reply
 
Thread Tools
post #2401 of 2417 Old 08-20-2014, 02:42 PM
Newbie
 
Join Date: Aug 2014
Posts: 14
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Quote:
Originally Posted by hatchback View Post
This statement is false. Many modern TVs support frame rates that are even multiples of 24p so there is no 3:2 pulldown applied at any point in the playback chain. And for TVs/players apply 3:2 pulldown, it very obviously disrupts the motion cadence. 3:2 pulldown NEVER results in smoother motion.
OK.

But, it is interesting how smooth cinematographers get such smooth playback from 24 fps. I've always believed that mostly, it's because they're well versed in how to optimize shooting that way.

Hard as I try, shooting at either 1/48 or 1/60 shutter speed, at 24 and 30fps respectively, I've found it almost impossible to duplicate the smoothness of the end product of the masters who do it professionally.
kualakin is offline  
Sponsored Links
Advertisement
 
post #2402 of 2417 Old 08-20-2014, 06:31 PM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,783
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 26
1. slower shutter speed
2. shallower depth of field
3. following the subject keeping it almost static in relation to the frame
4. Combine 1, 2 and 3 in different variations
5. either very smooth pans or whip pans
Ungermann is offline  
post #2403 of 2417 Old 08-20-2014, 08:01 PM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 534
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 54
Quote:
Originally Posted by kualakin View Post
OK.

But, it is interesting how smooth cinematographers get such smooth playback from 24 fps. I've always believed that mostly, it's because they're well versed in how to optimize shooting that way.

Hard as I try, shooting at either 1/48 or 1/60 shutter speed, at 24 and 30fps respectively, I've found it almost impossible to duplicate the smoothness of the end product of the masters who do it professionally.
Many things can cause motion judder. The most common problem in my experience is the playback device. DVDs rarely stutter because they are encoded at a low bit rate with a codec that is very easy to decode. But if you are looking at the video on your computer, there are many possible bottlenecks, including slow storage, insufficient memory, slow cpu, poor playback software.
hatchback is offline  
post #2404 of 2417 Old 08-20-2014, 08:40 PM
Newbie
 
Join Date: Aug 2014
Posts: 14
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Quote:
Originally Posted by hatchback View Post
Many things can cause motion judder. The most common problem in my experience is the playback device. DVDs rarely stutter because they are encoded at a low bit rate with a codec that is very easy to decode. But if you are looking at the video on your computer, there are many possible bottlenecks, including slow storage, insufficient memory, slow cpu, poor playback software.
The "judder" I'm referring to is what is inherent with 24fps, and 30fps video, compared to 60fps footage.

Watching 30p straight out of the camera, as opposed to 60p straight out of the camera, in this case, the Sony AX100, with no frame interpolation on the monitor or TV employed, the smoothness of motion is demonstrably more profound with the 60p footage, while the 30p footage has an inherent "judder", especially noticeable with pans and zooms.

So, the "judder" I'm describing is not computer or software related, although I'm sure it could be exacerbated by the conditions you described if you were feeding your monitor the footage from a computer and its installed software ecosystem.
kualakin is offline  
post #2405 of 2417 Old 08-20-2014, 11:47 PM
Member
 
leamas's Avatar
 
Join Date: Dec 2006
Posts: 82
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 14
Quote:
Originally Posted by Ungermann View Post
1. slower shutter speed
2. shallower depth of field
3. following the subject keeping it almost static in relation to the frame
4. Combine 1, 2 and 3 in different variations
5. either very smooth pans or whip pans
I agree with all points but not 2. There are scenes filmed at wide, where subjects run by example, but you also see the buildings around. I think they don't have any limitation because the camera is moved by some equipment that does every pan at 1/24 or so. Of course the other aspects like manual focus and correct exposure settings by default.


And then, there is another theory, that the human eye cannot perceive more than 24 fps. I think this is not true. The human eye can be learned to do many things. A trained eye will figure out small differences while a regular eye won't. If this would be true, we will not need 60p cameras of course.
leamas is offline  
post #2406 of 2417 Old Yesterday, 08:25 AM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 534
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 54
Playback on the AX100 is not necessarily free of problems. XAVC S is a highly compressed codec that requires significant computer resources to decode. So the problem you're observing could be a limitation of the AX100 decoder.

Pans require a lot of information to encode because each frame contains new information that did not appear in the previous frame. So in a pan it is easy to exceed the maximum bit rate of the codec (eg 50Mbps), which can severely compromise the image quality.

Speaking to your original point, skilled cinematographers work within the limits of their equipment. They avoid fast pans when they know they won't look good in the delivery medium.

If you want to know exactly what the problem is, upload the native footage so we can all take a look at it and tell you what went wrong.
hatchback is offline  
post #2407 of 2417 Old Yesterday, 08:39 AM
Newbie
 
Join Date: Aug 2014
Posts: 14
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Quote:
Originally Posted by hatchback View Post
Playback on the AX100 is not necessarily free of problems. XAVC S is a highly compressed codec that requires significant computer resources to decode. So the problem you're observing could be a limitation of the AX100 decoder.

Pans require a lot of information to encode because each frame contains new information that did not appear in the previous frame. So in a pan it is easy to exceed the maximum bit rate of the codec (eg 50Mbps), which can severely compromise the image quality.

Speaking to your original point, skilled cinematographers work within the limits of their equipment. They avoid fast pans when they know they won't look good in the delivery medium.

If you want to know exactly what the problem is, upload the native footage so we can all take a look at it and tell you what went wrong.
No computer necessary.

If you plug the AX100 directly in to a TV, such as mine, a 4K LG 55" UB9500, (HDMI 2.0 compliant), it's easy to see the demonstrable difference between 30p and 60p footage.

The AX100 is limited to 30P at 4K resolution.

Panning and zooming at that frame rate is inherently, irrespective of the codec or the camera itself, more "juddery" that 1080 60P.

So, for now, the trade off for 4K resolution, at least at the consumer level....and as far as I know, there are no 4K60P cameras available at a consumer level price point....is less "silky smooth" footage, albeit with much higher detail and resolution.

This is clearly apparent, whether you plug the camera into a monitor directly, or play the files the camera created with a computer hooked up to a monitor.

This "judder" can be mitigated with frame interpolation, which the LG 4K has and does quite well, however, as one of the previous posters noted, this comes at a cost of a "shimmering" effect that frame interpolation causes with dense colors, like grass and foliage.

Nothing can completely make 30P footage look as "liquid smooth" as 60P footage, especially when you pan.

So, the only solution natively, is to shoot at 60P if you want virtually no judder, and wait, like I am, for an affordable 4K60P "prosumer" type of camcorder or camera, which I think we'll see quite soon.

Until then, I can live with the 30P "judder" at 4K, because there are workarounds that can make it quite enjoyable to watch.
kualakin is offline  
post #2408 of 2417 Old Yesterday, 08:48 AM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,783
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 26
Quote:
Originally Posted by leamas View Post
I agree with all points but not 2. There are scenes filmed at wide, where subjects run by example, but you also see the buildings around. I think they don't have any limitation because the camera is moved by some equipment
That. See #3 in my list.
Quote:
Originally Posted by leamas View Post
And then, there is another theory, that the human eye cannot perceive more than 24 fps.
They chose 24 fps simply because lower rate was completely unrealistic, watch early silent movies.
Ungermann is offline  
post #2409 of 2417 Old Yesterday, 09:27 AM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 534
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 54
Quote:
Originally Posted by kualakin View Post
No computer necessary.

If you plug the AX100 directly in to a TV, such as mine, a 4K LG 55" UB9500, (HDMI 2.0 compliant), it's easy to see the demonstrable difference between 30p and 60p footage.

The AX100 is limited to 30P at 4K resolution.

Panning and zooming at that frame rate is inherently, irrespective of the codec or the camera itself, more "juddery" that 1080 60P.

So, for now, the trade off for 4K resolution, at least at the consumer level....and as far as I know, there are no 4K60P cameras available at a consumer level price point....is less "silky smooth" footage, albeit with much higher detail and resolution.

This is clearly apparent, whether you plug the camera into a monitor directly, or play the files the camera created with a computer hooked up to a monitor.

This "judder" can be mitigated with frame interpolation, which the LG 4K has and does quite well, however, as one of the previous posters noted, this comes at a cost of a "shimmering" effect that frame interpolation causes with dense colors, like grass and foliage.

Nothing can completely make 30P footage look as "liquid smooth" as 60P footage, especially when you pan.

So, the only solution natively, is to shoot at 60P if you want virtually no judder, and wait, like I am, for an affordable 4K60P "prosumer" type of camcorder or camera, which I think we'll see quite soon.

Until then, I can live with the 30P "judder" at 4K, because there are workarounds that can make it quite enjoyable to watch.
I don't think you understand the playback issue. If you did, you would know that it's easier to decode 1080p60 than 2160p30.

You don't need 60 fps to get smooth motion. You can get smooth motion with 24 fps.

Instead of arguing generalities, you should post your original footage. Then we can tell you exactly why you have a motion problem.
hatchback is offline  
post #2410 of 2417 Old Yesterday, 09:36 AM
Newbie
 
Join Date: Aug 2014
Posts: 14
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Quote:
Originally Posted by hatchback View Post
I don't think you understand the playback issue. If you did, you would know that it's easier to decode 1080p60 than 2160p30.

You don't need 60 fps to get smooth motion. You can get smooth motion with 24 fps.

Instead of arguing generalities, you should post your original footage. Then we can tell you exactly why you have a motion problem.
I'm keenly aware that there are filming techniques that can be employed to make 24P and 30P footage look less "juddery", my only point is, that 60P, if only because there are more frames per second producing the image, makes employing those filming techniques less essential in producing a less "juddery" image, if that's the look you're after.

There are many who like the "film like" judder that 24P and 30P produce, and there are times when I prefer it also.

There is no argument here. The fact is, 24P and 30P are organically more judder prone, than 60P. It's just math and physics.

It has nothing to do with technique, computers, or software.
kualakin is offline  
post #2411 of 2417 Old Yesterday, 10:24 AM
AVS Special Member
 
Tom Roper's Avatar
 
Join Date: Apr 2002
Location: Denver, Colorado
Posts: 3,745
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Ungermann View Post
1. slower shutter speed
2. shallower depth of field
3. following the subject keeping it almost static in relation to the frame
4. Combine 1, 2 and 3 in different variations
5. either very smooth pans or whip pans
^+1
Tom Roper is offline  
post #2412 of 2417 Old Yesterday, 02:08 PM
Senior Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 401
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 33 Post(s)
Liked: 14
One factor may be the speed at which the image is read from the sensor. At 2k it takes less time than at 4k and thus less judder. The pros may simply be using cameras that read the images faster plus working within the specs when panning etc.

Gene
Eugene157 is offline  
post #2413 of 2417 Old Yesterday, 03:26 PM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,783
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 26
Quote:
Originally Posted by Eugene157 View Post
One factor may be the speed at which the image is read from the sensor. At 2k it takes less time than at 4k and thus less judder. The pros may simply be using cameras that read the images faster plus working within the specs when panning etc.
Nope. This would affect the severity of rolling shutter artefacts but not the frame-by-frame judder.
Ungermann is offline  
post #2414 of 2417 Old Yesterday, 05:33 PM
Advanced Member
 
hatchback's Avatar
 
Join Date: Jan 2004
Location: NJ
Posts: 534
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 54
Quote:
Originally Posted by kualakin View Post
There is no argument here. The fact is, 24P and 30P are organically more judder prone, than 60P. It's just math and physics.

It has nothing to do with technique, computers, or software.
The flaw in your argument is to infer that the problems in your footage are attributable to the difference between 30p and 60p. This is an invalid inference. If you knew what you were doing, you would be able to film at 24p or 30p without creating any motion artifacts on your target playback device. The fact that you are unable to do so means means you have room for improvement. The fact that you are unwilling to post the problematic footage suggests that either you are a troll or you aren't willing to learn.
hatchback is offline  
post #2415 of 2417 Old Yesterday, 06:42 PM
Senior Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 401
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 33 Post(s)
Liked: 14
Thank you ungermann I will try to remember the difference. Makes sense as far as rolling shutter, looked up judder on Wiki.

  1. (television) Jerky playback caused by converting between frame rates; telecine judder




Ultimate rolling shutter, passing speed is 600KMH with AX 100, 5 coach magnetic levitation train passed in about 20 frames.

Judder not a problem here, I get smooth pb from AX100 material either from hdd or DVD


Gene

Last edited by Eugene157; Yesterday at 07:04 PM.
Eugene157 is offline  
post #2416 of 2417 Old Yesterday, 07:17 PM
Senior Member
 
Eugene157's Avatar
 
Join Date: Nov 2003
Location: Palm Springs area
Posts: 401
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 33 Post(s)
Liked: 14
How does one post a video here directly w/o YouTube or Vimeo etc.?

Last edited by Eugene157; Yesterday at 07:23 PM.
Eugene157 is offline  
post #2417 of 2417 Old Yesterday, 08:56 PM
Newbie
 
Join Date: Aug 2014
Posts: 14
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Quote:
Originally Posted by hatchback View Post
The flaw in your argument is to infer that the problems in your footage are attributable to the difference between 30p and 60p. This is an invalid inference. If you knew what you were doing, you would be able to film at 24p or 30p without creating any motion artifacts on your target playback device. The fact that you are unable to do so means means you have room for improvement. The fact that you are unwilling to post the problematic footage suggests that either you are a troll or you aren't willing to learn.
There's no flaw in my argument, because, I'm not making one.

And, there's no problem in my footage, and have not professed to having any, having stated previously that I shoot either 30 or 60P, depending on the look I'm aiming for, which I can find pleasing either way, depending on what I'm trying to record.

60P provides smoother video than 30P or 24P, by definition, because, it captures more frames per second.

That's inarguable.

I really don't need to post any clips, because, you can see for yourself, with a camera that shoots at 24 or 30 and 60fps, which you probably have, no?

Go outside and shoot a moderate pan across any horizon, from your right to left, 180 degrees, with some sort of objects, like a column or building in the background. Take from 6 to 10 seconds to do the pan. One at either 24fps or 30fps, and then one at 60fps. Pick any shutter speed and aperture you choose.

Which pan looks more liquid and smooth?

24 and 30P can be smooth, but, because of physics, 60P will always be more smooth.
kualakin is offline  
Reply Camcorders

User Tag List

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