3 ms·
Framerate hacking aside, the real question is why there was motion blur directly applied to the motion. As Malki's comparison shows, it's not at all clear why t
by KirinDave 8y ago
Framerate hacking aside, the real question is why there was motion blur directly applied to the motion. As Malki's comparison shows, it's not at all clear why that's there. No default GIF conversion would do it, it's not present in the original source video even under the zoom factor.
So even if these subtle framerate shifts are simply artifacts of very bad encoding software (which is weird... the free tools are much better than this), the fact that there are significant changes to the pixels of the motion is much harder to explain away.
Unsurprisingly, the ars article DOES mention this but the Buzzfeed and Vice articles, which represent less technically savvy sources, don't.
- makomk 8y agoWhatever software was used to create a proper video from the GIF seems to have done some kind of framerate doubling by inserting interpolated frames. It alternates between a clear frame and one with odd motion blur throughout. Also, I reckon the weird frame duplication is an interaction between the software used to create the GIF and the software used to turn it into an edited video later. More specifically: - The GIF has a non-standard framerate of 14.56 FPS. This means that it's probably playing about 3% slower than the original video, which isn't perceptible and isn't particularly important by itself. - The video has a more standard framerate of 29.9 FPS. The software seems to mostly handle this by doubling the framerate with interpolated frames, as described above. However, this doesn't quite fix the problem - roughly every 40 frames it'll be a frame short. - Sure enough, there's another frame duplication about 40 frames after the first.
- KirinDave 8y agoI find this analysis unobjectionable, but it does not explain the motion blurs.