UHD BD to BD?

[zoidberg]

Generally with CRF encodes it's preferable to run some tests to 'tune' the rate factor, this is when an intermediate lossless file is desirable rather than a lengthy, involved script. A lot of times a CRF encode will turn out lower than expected in size because it can really reduce the bitrate when necessary, as opposed to 2-pass where the bitrate will generally 'loiter' around the target bitrate. But when you've really got to hit the filesize accurately then 2-pass (or n-pass) is the way to go.

Tom, you bring up some interesting points about x264, is it worth starting a new thread?

[deleted user]

Yea, by higher I meant lower. Also, the majority of the encodes out there use crf so seems like a good compromise. Also, the other annoying thing with 2 pass in ripbot is, you can't preview the first pass (like you can in Simple x264 Envoder) as it doesn't write the video pass to hard drive. At least the x264 settings above are better than the standard ripbot ones which are not bd compliant.

What's the point of previewing the first pass though? First pass always uses lower settings by default, so whatever you get as preview is not how it will actually look. That's why x264 by default doesn't write out the first pass render to a file.

Generally with CRF encodes it's preferable to run some tests to 'tune' the rate factor, this is when an intermediate lossless file is desirable rather than a lengthy, involved script. A lot of times a CRF encode will turn out lower than expected in size because it can really reduce the bitrate when necessary, as opposed to 2-pass where the bitrate will generally 'loiter' around the target bitrate. But when you've really got to hit the filesize accurately then 2-pass (or n-pass) is the way to go.

Tom, you bring up some interesting points about x264, is it worth starting a new thread?

True I guess. 2-pass runs the theoretical risk of using more bitrate than is needed for an acceptable quality, but this will also increase the quality. It's not like CRF has any quality benefit or loss when operating at the same bitrate as 2-pass. At least that is the common wisdom I've read on forums (doom9 etc), I don't know how accurately this reflects the programming of the software. I have no reason to doubt it, though.

In theory it would be worth starting a thread, but I have no final perfect solution nor do I see one at the horizon, so it might just not lead anywhere. Also, my encodes are typically 10-bit 4:4:4, while most people here seem to like BD-compliance, so the settings I came up with may just not be much use to anyone either, as they don't 1:1 translate to 8-bit x264 afaik.

[X5gb]

I like to preview my encode as it goes along just to make sure everything goes to plan, don't want to be encoding for 2 days to find out there is glitch or freeze like I had twice with Grease and with CRF you are seeing the final encode unlike 2 pass. Also in the case of Grease the encode with CRF20 is using a bitrate of on average 35000kbps which is higher than the 2 pass variable 30000kbps I chose initially so am more than happy the way the encode is turning out using CRF. Another 17 hrs to go, hopefully no freeze towards the end this time.

[DoomBot]

In theory it would be worth starting a thread, but I have no final perfect solution nor do I see one at the horizon, so it might just not lead anywhere. Also, my encodes are typically 10-bit 4:4:4, while most people here seem to like BD-compliance, so the settings I came up with may just not be much use to anyone either, as they don't 1:1 translate to 8-bit x264 afaik.

I wouldn't mind reading about 10-bit 4:4:4 encodes and maybe a script for it in another thread. For my personal stuff i don't care about BD compliant, i don't have a BD player anyways, everything plays from a computer. And if there's a drastic quality difference in doing it that way, i say why not.

[Beber]

I don't really care about Blu-ray compliant settings either, as long as my Zappiti media player plays it, I'm fine.

This is what I got from using the "slower" preset on MIB 2:

Encoding settings           : cabac=1 / ref=4 / deblock=1:-1:-1 / analyse=0x3:0x133 / me=umh / subme=9 / psy=1 / psy_rd=1.00:0.15 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=2 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-3 / threads=6 / lookahead_threads=1 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / bluray_compat=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=2 / b_bias=0 / direct=3 / weightb=1 / open_gop=0 / weightp=2 / keyint=240 / keyint_min=24 / scenecut=40 / intra_refresh=0 / rc_lookahead=60 / rc=2pass / mbtree=1 / bitrate=15000 / ratetol=1.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / cplxblur=20.0 / qblur=0.5 / vbv_maxrate=62500 / vbv_bufsize=62500 / nal_hrd=vbr / filler=0 / ip_ratio=1.40 / aq=1:1.00

Actual Blu-ray:


UHD -> 1080p BT.709 using RipBot264 (hue -5, sat 1, bright -2, contr 0.9):

[Beber]

Here are some tests with Groundhog Day. I'm shocked to realize now how bad the original Blu-ray was. I tried to approximate something in the vicinity of both the Japanese mastered in 4K Blu-ray and the MadVR conversion:

Original Blu-ray / Japanese Mastered in 4K Blu-ray encode / UHD MadVR / UHD -> RipBot264 (hue -5, sat 1, bright -2, contr 0.8)

[deleted user]

I actually like the Blu Ray colors more here

[Chewtobacca]

Do you mean the JP BD? Yes, I quite like how it looks.

[Beber]

I actually like the Blu Ray colors more here

The original Blu-ray? Really? Even how the lighting looks under the snow? Even in the bar? Don't you find it too red in the payphone shot and on the black bartender? I also think the B&B corridor looks more cosy with nice, warm colors.

[X5gb]

On Groundhog Day, definitely prefer the 4k colours over the original bluray.