Hello just making a poll, which one do you prefer? personally I prefer x265 but since the rarbg falldown i’ve seen that almost all 1080p rips are in x264, what do you think about that, and do you recommend any place to find more x265 content beside those in the megathread?
All my content is converted to CPU encoded x265. Size is MUCH smaller and quality better than GPU encoded x265. My preference is to get remux copies of the content and then encode it myself.
Let’s see if this works:
I don’t keep 4k content, I find that 2k encodes for the stuff I really want at high fidelity is enough with Nvidia upscaling (Nvidia shield). Plus surprisingly some of my 2k files are no larger than 1080p.
A lot of comments suggesting AV1 has better compatibility than h265. In my experience the opposite is true. H265 is supported by all of my devices including Plex on my smart TV without transcoding, whereas AV1 makes everything have a fit trying to play it. Am I doing something wrong?
Shit, I like HEVC in theory for the compression especially but it’s copyrighted bullshit or whatever.
I use Plex with lifetime pass on my QNAP NAS and it has to hardware transcode HEVC to a playable format because of said copyrighted bullshit.
It doesn’t affect me that much unless I’m trying to jump around on the media as it will need to load. The other thing is that you can have Plex save transcodes but that obviously gobbles up disk space.
tl;dr 264 = 👑
Because of this post, I reencode a BD rip I made using handbrake to see how small the output file would be. I used the 4k av1 fast profile, but changed the audio tract to passthrough. Holy crap, 44gb down to 1.5gb. what black magic is this?
AV1 is very efficient (around twice as good as h264), but a filesize that low was almost definitely because the default encoding settings were more conservative than the ones used to encode the blu-ray. The perceptual quality of that 1.5gb file will be noticeably lower than the 44gb one
I’ve recoded a bunch of x264 to AV1 and routinely gotten file sizes that are 10-15% of the original file size (a little more than 1/10th the original size)
What I’ve found is that source content often has a lot of key frames. By dropping key frames down to one per 300 or one per 150 frames (one per 10 or 5 seconds for 30fps) and at scene changes, you can save a LOT of space with no loss of quality. You do give up the ability to skip to an arbitrary point in the content, however. You may have to wait a few seconds for rendering to display if you scroll to an arbitrary point in the content.
If you’re just watching the content straight through, no issues. I set CRF to achieve 96 VMAF and I can’t tell any difference in quality between the content with that setup.
I had one corpus of content that I reduced from 1.3 TB down to 250 GB after conversion.
Unfortunately, only the most recent TVs have AV1 playback built in, and the current Fire sticks, Chromecast don’t have support for playback from a LAN source. I’m hoping the next crop of Chromecast and similar devices get full support, I’m assuming it’s just a matter of time until AV1 decoding is included in every hardware decoder since it’s royalyy-free.
