Hacker News Reader

Comment by rebeccaskinner | Hacker News Reader

I do a lot of video compression for hobby projects, and I stick with h264 for the most part because h265 encoding requires far too much extra compute relative to the space savings. I can spend an hour compressing a file down to 1gb with h264, or I can spend 12 hours compressing the same file to 850mb with h265. Depending on the use-case, I might still need the h264 version anyway since it's far more widely supported by clients. If I had a data center worth of compute to throw at encoding, or I were running a streaming service where the extra 150mb per video started to add up, then I'd definitely on-board with h265 but it's really hard to justify for a lot of practical use-cases.

View on HN

> I stick with h264 for the most part because h265 encoding requires far too much extra compute relative to the space savings. I can spend an hour compressing a file down to 1gb with h264, or I can spend 12 hours compressing the same file to 850mb with h265.

FWIW, as someone who is a complete layman on this topic but has spent much of the past 6 months converting age-old videos to h265...

i have a pi5 where i queue up old videos into a RAM disk, it converts them to h265, and moves them back to the machine which uploaded them. It can take up to 2 days to compress any given video (depending mostly on the input format - old AVIs tend to compress in a few hours), but the space savings is _much_ higher than what you're reporting. The average size of my h265-encodes is anywhere from 1/3rd to 1/2 of the original, and 1/4th of the original is not terribly uncommon. That said: the input formats vary wildly, with possibly only a minority fraction being h264.

i've tried the same thing using a pi4 but it needs roughly 4-5x as long as the pi5. An 8-core intel-based laptop can do it in some 25% of the time of the pi5 but pulls a lot more electricity while doing so. So my otherwise idle pi5 gets the assignment.

My Firefly collection (14 tv episodes plus the movie) dropped from 8gb to 2.1gb. My Bob Ross collection (141 videos) went from 38gb to 7.4gb. 128 Futurama episodes/movies was cut from roughly 50gb to 21gb.

i was once asked why i didn't choose AV1 and the answer is simple: when i googled along the lines of "how to get the best video compression with ffmpeg" the most common answer was h265. Whether that's correct or not, i can't say, but it works for me and i've freed up the better part of a terabyte of space with it.

Interesting, your numbers look to be quite a bit better than the 30% improvement I've heard is the standard rule of thumb for how much improvement to expect with h.265, and on average for me it's been closer to 20%. I think h.265 starts to show much more meaningful improvements as you start using lower bitrates, and the specific type of media probably matters a bit as well.

I do most of my encoding on an i9-13900k. I've also noticed that h.265 seems to be quite a bit more likely to trigger the smt related crashes in ffmpeg, which makes it a pain when I've queued up a lot of things to encode over night.

What about when you’re recording (say your iPhone / android / standalone camera) - are you choosing h264 or h265 (or something else like an intra-frame only codec for easier editing).

Most of what I do is working with pre-recorded video, but for projects where I'm doing recording I tend to use h.264 since I don't have a lot of devices that support h.265 anyway, and picking one format with broad compatibility is generally more important to me than the efficiency gains I'd get using h.265 when I could.

Is GPU accelerated encoding not an option?

I haven't actually done the comparison myself, but the common refrain is that the quality of the video suffers when you use GPU accelerated encoding. That's not a tradeoff I want to make, so I've just stuck with CPU encoding.

Wait, why does GPU encoding affect the output quality? Is that common in transcoding stuff?

>why does GPU encoding affect the output quality

Every single encoder, regardless or hardware or software will output different output quality. Video Encoding is a lossy encoding, not lossless like ZIP or RAR. That is the same with Audio that is why people test different audio encoder. It seems more people knows this about Audio than Video. At least on HN.

GPU encoding quality has always been lower quality than Software. Primary because they trade off quality for speed. It is good enough once you hit certain bitrate, but at low bitrate where absolutely encoding efficiency is required Hardware Encoding just dont compare to software. Even with Software you could have multiple encoder with different results. It is not like H.266 / HEVC only has x266 as encoder, example Netflix uses BEAMR. And Broadcasting TV station uses other encoder that better suits their needs.

And it is one reason why I dislike all these AV1 discussion, whenever people said it is slow they said use SVT-AV1, well yes SVT-AV1 is faster but dont produce the best AV1 quality encode. So what is the point. It is like every time these discussions about AV1 AOM supporters will just move the goal post.

Do you blanketly refuse to encode with h.264 or h.265? Because they're always worse than the best AV1 encode too.

If you only use h.266 and you're willing to wait extreme amounts of time to encode, then that's valid, but understand that you're an outlier. Most people don't have that much time to spend on encoding.

You don't need to find the best possible encoding. SVT-AV1 can encode as fast as x264, keeping the same visual quality to human eyes while reducing bit rate by 50%.

If you want to retain visual quality, you always have an option to use higher bit rate.

If we assume that to be true (VMAF and SSIM aren’t the whole picture), just keep in mind that’s only true at particular speeds, bitrates, and types of content.

What I should say is, please show me an AV1 encoder that can beat x264 in quality at a source-transparent encode of a 4K or 1080p Blu-ray film, if given ~3GB per hour of content and the encode had to run in at least 3x the runtime. I’d start using it! It may be there as of recently, it’s been a year or two since I looked.

This is a usecase where AV1 encoders and even x265 have had trouble.

Why does GPU need to trade off quality for speed? GPU is faster than CPU, that is the whole point of GPU encoding, isnt it?

Does it mean just that nVidia and AMD engineers done a poor job when implementing those encoders?

No. Decoding is a job mostly done by specialized hardware - the shader units are used sometimes, before a fully fixed function implementation is ready. Encoding in particular doesn’t map well to GPUs. They can do it, using varying degrees of fixed function and shader cores, and it’s nice to isolate that load from the CPU, but they implement fewer of the analysis, prediction, and psychovisual optimization tricks that x264 and x265 use, and less of the optional features of the format. They often can beat software at specific, fast speeds with lower power consumption, but the trade-off is being inflexible and unuseful for making encodes that are transparent to the source.

I mean the thing with something like SVT-AV1 is, even if it doesn’t give you the best efficiency encode, does it do a more efficient encode than your alternatives in a reasonable timeframe.

GPU video encoding is pretty much always optimised for real-time encoding, meaning that it can't run certain optimisations as it would increase the time to encode.

Compare x264 veryfast and veryslow presets. There is a quality difference at the same bitrate.

Additionally, GPU encoders don't have as many psychovisual options as CPU encoders as they would need to be included in the hardware and adding extra options to CPU encoders is much faster, easier and cheaper.

You could build a non-realtime GPU encoder, but there is not much point.

There two types of GPU acceleration: acceleration by a hardware codec bolted on a GPU, and actual GPU computational acceleration. the latter is not widely used but probably is not n gonna have any difference in terms of quality, and provides only modest acceleration.

Reply on news.ycombinator.com