This is an old revision of the document!


60 fps

Some good links:

As a general video snob, I always prefer to have the best quality available, and by principle having to re-encode anything makes life frustrating for me. However, encode we must, and so we make the best of it. Taking original DVD source video from 23.97, 24, 25, 29.97 or variable frame rate to 60 fps seems a little weird, and so it is. In doing any kinds of tests, the ultimate benchmark is measuring how the user experience rates between two or more sets of examples. Despite that, I still enjoy the intellectual challenge and discovery of new options.

Having a source be variable framerate has always bothered me, and caused problems in the past when I was first doing encoding from one format to another (aka ripping DVDs). Just in general, trying to make sense of things, it didn't help that the source wasn't consistent. More especially, it made life miserable trying to encode a video using standard tools (ffmpeg) instead of popular tools (HandBrake). Just as I'm a purist when it comes to original source, I'm also driven by wanting to use as small a tool and set of options as possible – in that case, ffmpeg. Since I couldn't do that, it only added to the frustration.

It basically spiraled me down into a confusing mess of trying to hit target bitrates, and basically causing all kinds of audio sync issues from day one. It was (and still is) very frustrating. I went years banging my head against a wall, and finally actually tried using HandBrake, and I've been using it since then for all my DVDs.

So, the first reason I like encoding my video to 60 fps is that it standardizes the frame rate. No more variable frame rate, and I always know it's going to be the same across the board.

On that same principle, I do the same with my audio encodes as well. I prefer using a constant bitrate for my rips. All my MP3s I'll encode at 320 kbps. Can I tell the difference between that and variable framerate with default settings? The answer is, sometimes.

Fundamentally, we re-encode our audio or video primarily to save space. The second reason for video is for filtering, such as removing interlacing. But storage is the main reason. Since the cost of storage consistently decreases, the ideal for storing audio and video in its primary formats becomes an option.

There's still reasons items will have to be re-encoded though, and in most cases re-muxed as well. Mostly to add metadata when remuxing, or fixing broken containers in one form or another. I'm not *too* upset about those, but again, if the playback software can work around the quirks, I'd prefer original source still.

The second reason for moving video to 60 fps is that it's becoming the standard. Blu-rays have it in some cases, and as ridiculous as it is, I also encode my DVDs to other specs that Blu-rays have, such as using the 4.1 H.264 level. Again, more consistency.

A third reason is that the encodes are somehow smaller with 60 fps. I have no idea why, as I don't completely understand the encoding process, but I'll happily take the 5% decrease as a bonus since I don't have *that* much storage still.

The fourth reason is that it just gives me warm fuzzies. It's nice to know that I've got things so squared away

The final reason is the argument in general that the TV is playing back at 60hz anyway, and that the video will be smoother at 60 fps for that reason.

Now for the cons, of which there are good ones as well.

For someone who is a purist and determined to get the best quality as possible with no modification to the material that is not necessary, changing the framerate flies completely in the face of that goal. Converting video to another framerate is obvious modification.

Next is doing basic QA. If there's a problem with the encoded video, how do you know that converting the framerate isn't the cause of the problems? If there was A/V sync issues, the FPS is probably the first culprit in every case that's going to be called into question.

Warm fuzzies are just that – emotional reasoning that ignores logic in favor of believing something is better, just because it is! That's some serious hand-waving there, and I'm being harsh on purpose … and to its favor, placebos actually do work. It logically doesn't make sense, though, as its own reason.

Most of it is going to be creating duplicate frames anyway. You'd get just as much “quality” improvement when it comes to wanting to standardize on a fixed framerate. However, I think if you're going to standardize, there'd be more problems creating duplicate frames every 4 to 2 frames in general, so going to 60 makes sense here – you might as well give it much more room.

So, there's some good reasons to go ahead and encode it in 60 fps, but probably the best of all is a good generic “it won't hurt” summary argument. Don't do it with the expectation to improve quality of the source, but because it can improve playback of the video .. possibly. That's good enough. :)


Navigation