If you don’t know what I mean, I don’t know what to say. It didn’t look like this in the theatre and on DVD. It isn’t that it is poor quality. It has high definition. It just looks… weird. The result makes it seem like wasn’t filmed outdoors in natural light.
I’ve seen films on display TVs in stores look like this too, which is what made me wonder.
Was the original in 3D? It just looks to me the way most 3D movies look like when they are rendered in 2D. A lot of movement and blocking from front to back that is meant to leap out of the screen in 3D, but looks less imaginative, and more repetitive in 2D. Also, the constant camera circles, that would be bringing different characters “into the theater” get repetitive too. Since they don’t show a different angle, they sort of remind you of a video game when you have to keep circling around to choose which character is going to strike.
It’s at double the usual framerate, 60fps. That adds extra detail and smoothing. TVs interpolate by adding frames in between the actual ones, but sometimes they’re actually filmed that way.
Get used to it, it will be more common in the future.
Thanks for clearing that up!
(Moderators: This is not intended as threadshitting. I will try to be factual.)
What? It looks just like every other video game I’ve ever seen, which is the level to which recent movies have descended. This is a combination of green screen and CGI, along with what we used to call “excessive intervalometer use.” The best way to think of it is that you are seeing discontinuity in the shots. CGI characters morph into actual actors, sequences are speeded up or slowed down unnaturally, the backgrounds are strictly CGI, etc.
Could you be a little clearer? Do you believe that the original (theater version) looked much better?
Better is a matter of taste. But here is a version of the same clip looking more like typical film. https://www.youtube.com/watch?v=_Qq6dQwLh1s
Actually, if you open it in an editor, you can see that every second frame is a repeat of the previous one (a duplicate frame). In other words, it’s theoretically 59.96 FPS, but it’s effectively 29.98 FPS. The effect would be minimal. This is not the reason it looks “weird.”
Yes, it is a different encoding type.
I’m still not clear about exactly what the OP sees as weird. The first video clip is refreshing twice as quickly as the second. This may not correspond to his screen refresh rate. Maybe this is the issue? (For example, 85 Hz refresh rate for a 59.96 fps video?) As others have pointed out, your video display refresh rate (typical TV vs. computer monitor) can make a difference.
it seems to lack the “motion blur” that I’d expect to see from something shot on film.
I am not sure what editor you are using, but I just downloaded the 1080p version and it is definitely 60fps with no duplicated frames (the lower resolution versions were only 30fps). It is also easy to see that it was not shot at 60fps, but interpolated to that rate. There is much “motion morphing” and blending among the frames.
YMMV…but the several versions I downloaded (including the 1080p) are not 60 FPS, They are 59.96 FPS (much closer to exactly twice the NTSC standard). Since MPEG-4 (H.264, AVC, etc.) have both an independent presentation time and frame duration time, this may just be an average. Being shot at (nominal) 60 FPS and being interpolated to 60 FPS are pretty different.
But there are also some significant issues that come up in a frame-by-frame analysis. The first I-frame is followed by two duplicate frames (using most players), then you have a P-frame followed by a single duplicate frame. This depends on your editor or player. Run ffprobe and check for yourself. (Please keep in mind that many players treat an extended frame as a single frame followed by a duplicate.)
In any case, I still don’t understand exactly what the OP thinks is weird. Artificial looking? Odd pauses? Discontinuity? Any of these would be on the order of 0.033 to 0.015 seconds, not usually noticeable to the human eye.
You are funny dude. Yes, the clip is technically at 59.940060fps, 60fps is just a convenient shorthand. It is not just doubling frames to get that rate, those are unique frames.
It looks weird because it is being shown at over twice the normal film rate, with interpolated frames smoothing out the action. It is similar to the “soap opera effect” seen on many high frame rate TVs.
(sigh…) OK, if you say so. The frame rate is an average, since each frame can have a unique (to the microsecond) presentation time.
I’m not trying to be “funny.” I get paid a lot of money ($200/hour ) to interpret video files. You are not correct in your statements about interpolated frames.
I don’t know anything about the technical side, but I agree that this clip definitely looks strange, in a way similar to the ‘Soap Opera Effect’. If I had to describe it, the parts of the scene that are moving (vehicles and actors, mostly), seem to have a different ‘quality’ to the parts of the scene that are static (eg buildings). This feels like it’s triggering some part of my brain that brings to mind watching a movie with the cheesiest special effects, where the actors are actually being filmed, but the backgrounds are either just a fake backdrop, or on old fashioned green screen. This is distracting to the point that the scene is almost unwatchable.
Then provide one iota of evidence that you know what you are talking about, please. And, by all means, elaborate on any errors I may have made.
If you are not aware of the “soap opera effect” and perceptions that people have of high frame rates, given long discussions about that very topic on this board and others, you are missing something fundamental to the topic.
I’m not trying to start an argument about high framerates and whether they are worthwhile. I’m pointing out that if you download this specific clip (the “1080p” version to which you refer), you can tear it apart with ffprobe and see that the image presentation duration times (usually referred to as “packet duration”) range from 0.050 seconds to 0.017 seconds, making the effective framerate from 19.978 to 59.960 fps. (I know…it should be ips, not fps, but I started out saying “frames” and I’ll stick with that.) The average is 29.933. Most players will play it at 29.933 (or some framerate very close to that). Some players will play it at a much faster rate and insert duplicate frames to maintain the correct timing and sound synch. You can pull the individual duplicate frames and run an image differential on them to show that they are identical to preceding or following frames. It’s simple to do.
So this specific clip is not running at 59.940060 fps with 59.940060 separate, different images being presented each second. A duplicate frame is not an interpolated frame. Further, when I run it on a media player that shows me continuous stats, the framerate varies a great deal, but never gets anywhere near 59.940060, even for short action sequences.
There are exactly 7371 separate frames/images/pictures in this clip and it is 246 seconds long, for an overall average framerate of 29.963 fps.
Caveat: I just ran a very quick analysis of this clip and so somebody may be able to come back with a more in-depth analysis and set of numbers.
It is apparent that you are not downloading the proper clip or there is a problem with your downloading tool. The actual 1080p60 clip has 14,760 frames over 246 seconds.
Here is a comparison of the clip in question with the 24fps blu-ray that includes frame counts and timestamps. Play it at 1/4 speed–hopefully you can see the difference.
I’m more than happy to admit when I make an error. The clip I downloaded from the OP’s link is 1920x1080 and is approximately 99,355 KB. I assumed that this was the one to which you referred. Is there another clip available for downloading that I have missed? I do not see one specifically labelled “1080p60.” Or did your clip come from a different site?
Again, I’m not trying to make a case either for 60 fps video or against it, simply that the clip I’ve downloaded is not 60 fps. I’ll be pleased to try again with another clip.
The clip I used is from the very same video referenced in the OP. It looks like you need a better download tool.
For reference, here are the HD formats listed for this video using the “youtube-dl” utility:
[info] Available formats for udKE1ksKWDE:
format code extension resolution note
...
247 webm 1280x720 720p 1717k , vp9, 30fps, video only, 41.26MiB
136 mp4 1280x720 720p 2391k , avc1.4d401f, 30fps, video only, 52.06MiB
302 webm 1280x720 720p60 2910k , vp9, 60fps, video only, 69.39MiB
248 webm 1920x1080 1080p 3044k , vp9, 30fps, video only, 74.22MiB
298 mp4 1280x720 DASH video 3514k , avc1.4d4020, 60fps, video only
137 mp4 1920x1080 1080p 4379k , avc1.640028, 30fps, video only, 97.03MiB
303 webm 1920x1080 1080p60 4757k , vp9, 60fps, video only, 118.15MiB
299 mp4 1920x1080 DASH video 5872k , avc1.64002a, 60fps, video only
...
Yes, I do see it now. Thank you.
But I’m afraid that we’re hijacking the thread at this point. There’s still no indication that the framerate is what is causing the OP to think the original clip is “weird.” He specifically comments on the lighting, for example. In my original response, I referred to possible issues with the green screen/CGI process.
I’ll be happy to come back to the framerate issue if it seems like it is part of the answer. Otherwise, we should probably start a new thread if we want to discuss 60 FPS video.