H265 To H265 vs H264 to H265

[V1rg0_]

Hi there, i'm quite a novice in the Encode side and i need your help.

I was wondering if Theres some quality Difference between Encode H265 to H265 vs H264 To H265 at the same Bitrate
Example:

Main H265 File 8000Kb/s 1080p
To (encode option on staxrip: Nvenc P4 Vbr 2200kb/s
Encode H265 file 2200kb/s 1080p

VS

Main H264 File 8000Kb/s 1080p
To (encode option on staxrip: Nvenc P4 Vbr 2200kb/s
Encode H265 file 2200kb/s 1080p

I know, that x264 can handle more information and it's why you Rip your bluray in this codec,
But I was wondering to Rip my bluray in x265 directly, And after i watch it i compress the media to save some space on my poor NAS

Oh and Thanks in Advance for Help

[Cornucopia]

No that is not the reason why x264 is used. When you RIP, the optimal format is easy. Ripping only copies the decrypted files from optical disc, so the resulting files on the computer would be identical in format to the original. Remuxed files would also be as good quality, though they may be more or less compatible depending on your playback chain.

X264 use implies you are not just ripping, but ripping & converting. What you convert to should also depend on the target system(s). X264 is very good, but so is x265, and the h265 format is often more efficient than h264. But that should not be the sole determiner of your choice of codec anyway.

Ripping & converting to get a particular size does no good if you don't change the bitrate. Since h265 is usually more efficient, that can mean one of 2 things. Either h265 has better quality at the same bitrate, or h265 has the same quality at a lower bitrate. But this varies greatly depending on the content.

My suggestion is, since it sounds like you want to convert either way, you choose your codec based on your system capabilities, then do some tests using CRF type encoding to see what the acceptable level is for you, and then encode your titles at that crf level. Note that the crf numbering systems are different between x264 and x265, so you cannot just assume that same level of quality in a different codec.


Scott

[ConsumerDV]

I know, that x264 can handle more information

H.264 is a well-known video compression standard. x264 is a free software library that implements H.264. Similarly for H.265 and x265.

It is generally considered that H.264/AVC is about twice more efficient than H.262/MPEG-2 Part 2, and that H.265/HEVC is about twice more efficient than H.264/AVC. These differences can be revealed at low bitrates; at high enough bitrates they are all pretty good, although there are edge cases: for example, some algorithms don't work well at very low bitrates, other algorithms are not available at high resolutions, etc.

I just googled up this awesome article that I haven't read fully yet, just skimmed over. Granted, it is six years old, but the relationships between the compression algorithms are still valid. His conclusions:

On the whole, for the average case, x265 showed bitrate of about 59% of that of x264 at the same CRF. ... The result seems to be consistent across x264 and x265, although (unexpectedly) the difficult case seemed to tolerate higher CRF. ... x265 isn’t very good about handling interlaced content, the mature x264 encoder was much more adept at handling interlaced content correctly.

[V1rg0_]

In respond To ConsumerDV:

if i understand what you have say. No matter what the codec is, at high bitrate there no such difference, Its at low bitrate that really matter.

if that the case is my exemple true?

x264 at 20 mb/s
Encode to
H265 2200 kb/s

Will have the same Encode result as:

vp9,mp4,x265,etc at 20 mb/s
Encode to
H265 2200 kb/s


You say:

It is generally considered that H.264/AVC is about twice more efficient than H.262/MPEG-2 Part 2, and that H.265/HEVC is about twice more efficient than H.264/AVC.

basically if a h265 file is at 1500 ks/s this is approximately the equivalent of an h264 at 3000 kb/s with the same main File, is that True?


In Respond To Cornucopia:

Yeah you right, In fact I always forgot when i Rip my bluray than i Encode it too.

Is CRF indicate the strength of the encode process? Is like CBR and VBR if that so why should i use CRF more than VBR.
VBR is not suppose to better, And adapt to the difficulty of the media?

[Cornucopia]

x264 at 20 mb/s
Encode to
H265 2200 kb/s

Will have the same Encode result as:

vp9,mp4,x265,etc at 20 mb/s
Encode to
H265 2200 kb/s

First, remove "mp4" from that list, as that isn't necessarily a codec (unless you are referring to mpeg4 part2, simple or advanced simple, like divx). More properly, it refers to the container.

Second, as I already mentioned, the answer is "IT DEPENDS". Primarily on the content, but also on the encoded settings.

In general, however, assuming all other factors are equal, h264-type and vp9-type encodes are roughly equal in quality/efficiency, with h265-type encodes being more efficient. (This has been mentioned multiple times).
Re-encoding again to h265, ESPECIALLY to such a much lower bitrate, would create artifacts that would somewhat cover up those original differences, if any.
Think of it this way: 62% is different from 50% which is different from 39%, and that's a big spread of difference. But if you re-encode to 10% of that, you now have 6.2% vs 5.0% vs 3.9%, which spread is hardly any difference.
Those numbers are made up and don't correspond to any particular codecs, so don't read anything into that. It was just an analogy.

As far as bitrate/quality. CRF is a form of quality-centric encoding rather than a bitrate-centric encoding. However, one can for simplicity sake consider cfr is equivalent to wildly varying single pass vbr, whereas normal vbr (whether single pass or multiple) is much more constrained in its swings. It has been demonstrated here before that at (nearly) identical bitrates, a cfr encode looks (nearly) identical in quality to a 1 pass vbr encode. 2pass or higher might be better still, but then you are trading off slight improvement (s) in quality with extra passes.


Scott