+ Reply to Thread
Results 931 to 960 of 2222
-
-
Please someone help on topic for newbie beginners.
Code:ffmpeg -i "test2.avs" -b:v 23552k -maxrate 23553k -bufsize 600k -r 23.976 -pix_fmt yuv420p -f yuv4mpegpipe -y "test2.y4m"
-
@rhaz:
YUV4MPEG2 is a simple file format designed to hold uncompressed frames of YCbCr video formatted as YCbCr 4:2:0, YCbCr 4:2:2 or YCbCr 4:4:4 data for the purpose of encoding, likely to MPEG-2. The part "YUV" in its name just derives from the fact that the color space YCbCr (used for color encoding in digital media) is often falsely mixed up with the color space YUV (used in analog PAL based applications, including analog TV and video tapes).
-> you tell ffmpeg to create uncompressed yuv 4:2:0 output, therefore bitrate settings will be ignored.
@tnti: this thread has 936 posts, telling someone to read the precious replies without at least giving some directions isn't really any help at all -
Selur thanks for reply. So you mean that this is a correct way of what I did and it's normal to get terabyte size .y4m files from 2 hour long bluray sources? If so that's OK, I can live with that.
So pretty much this is how it works? I do cropping, resizing and etc in .avs then I convert it to .y4m and then I encode it with x265? And then how do I mux it to .mkv?
@tnti yeah I am reading it, I am still at page #7! -
So you mean that this is a correct way of what I did and it's normal to get terabyte size .y4m files from 2 hour long bluray sources?
So pretty much this is how it works? I do cropping, resizing and etc in .avs then I convert it to .y4m and then I encode it with x265?
And then how do I mux it to .mkv? -
Thanks everyone for fast help. So I am trying this avs2yuv now, but I get errors so I do something wrong. I put both x265.exe and avs2yuv in the same dir and this is the line I use:
Code:avs2yuv "test.avs" -raw -o -|x265 --fps 23.976 --input-res 1920x1080 --bitrate 2215 --preset slow --output "test.hevc" --max-merge=1 --no-tskip --tu-inter-depth=2 --tu-intra-depth=1 --wpp
That very same .avs works fine with that previous method I wrote above, but I'd really like to skip those monster size .y4m files. -
Example
avs2yuv "%~dpn1.avs" -o -|X265 --y4m --crf 24 --rd 0 --max-merge 3 --tu-intra-depth 1 --tu-inter-depth 1 --output "%~dpn1_video.hvc" -
WPP is enabled by default
You do not need a separate set -
you didn't specify the input, in your case:
Code:--input -
-
With ^ example I get:
Code:x265 [warning]: preset or tune unrecognized test.avs: 1280x690, 24017/1001 fps, 1455 frames Output error: wrote only 1110380 of 1324800 bytes
-
x265 [warning]: preset or tune unrecognized
-> think you could have understood that yourself -
Yes. I kind of made it work, it starts encoding, creates ~100KB size .hevc, but then x265.exe crashes (I am on i3 with 8GB memory). Line:
Code:avs2yuv "test.avs" -raw -o -|x265 --fps 23.976 --input - --input-res 1920x1080 --bitrate 2215 --preset slow --output "test.hevc" --max-merge=1 --no-tskip --tu-inter-depth=2 --tu-intra-depth=1
-
I know, I guess the nerdy fuckers needed some purpose to stay at work. One of the key creators of JPEG used Lena as a test image for AIC (advanced image coding) on Usenet back in 2007 before I knew who Lena was. Despite being in high school and he being a MD graduate, I actually was able to offer suggestions that he found useful. I demanded he send us nudes for our contribution, not expecting to actually get anything and he... gave us the full pic of Lena.
I did think El Heggunte's sample was a video though, which it wasn't. Does x265 compress still images with better quality than JPEG?
Originally Posted by poisondeathray -
try another x265 version and/or throw out non essential stuff from you x265 call,...
Just a reminder:
x265 is in a pre-alpha state so it's not really a surprise that it sometimes produces non compliant output or is broken in another way. -
Yeah but I was wondering maybe there's something wrong with my command line. And I am using latest version, I just started today and compiled my first x265. So I am sure it's latest version.
Update: I replaced --bitrate 2215 --preset slow with --crf 24 and it works now, it encodes by frame now, not percentage, and boy oh boy it's mega slow and that would take ages to finish anything at all. Maybe I should stick to that method 1 with .y4m monsters. -
Does x265 compress still images with better quality than JPEG?
So I am sure it's latest version. -
-
@Mephesto it didn't crash when encoding not via avs2yuv, but straight from .y4m.
Update: So while in process, I did some calcs and ~1h 40minute long Bluray will take up to 150GB+ .y4m (: If this is the feature of latest encoding technology - I don't like it ;P
Update: Yup, 186GB.Last edited by rhaz; 14th Nov 2013 at 12:53.
-
As a less-risky workaround... first, mux to MP4 with MP4Box, then,
ffmpeg -i input.mp4 -vcodec copy -acodec copy output.mkv
(many thanks to vhelp)
Originally Posted by Mephesto -
well, it boils down to the decoder. and, ffmpeg is you basic video skinner. look at it as a factory of codecs and custom skins. as long as it is updated to the latest codec, it will skin your codec in any container of choice, .ts m2ts vob mpg mpeg avi mkv and so on. to skin hevc encoded videos, you need the experimental ffmpeg v2.1 build. see post # 893 for info/download link.
-
Just a note regarding decoding speed:
A Phenom-II X4 945 3 GHz is sufficient to decode FullHD 1080p HEVC fluidly with MPC-HC 1.7.0.208 (1.7.0.154 only with reduced complexity).
Osmo4 (GPAC 0.5.1-DEC r4904) is visibly slower, and – according to the rather stable 50% CPU utilization – only using 2/4 cores.
__
BTW, the quality of CRF 24 with AQ is amazing!
Code:--crf 24 -w -i 240 --bframe-bias 20 --rdpenalty 1 --rd 2 --aq-mode 1
@ vhelp:
Maybe "to skin" is not the best matching term regarding multiplexing material into a container; and due to the "filter" behaviour of ffmpeg (writing output files sequentially, not seeking back to the header when finished, because someone might use it in a pipe), the probability is high that media files created my ffmpeg may be incomplete because some values which would have to be written in the beginning of the output file are not known before its end. -
i would like to compare the two modes, --q and --crf to see the differences, but i am still unable to make this happen. i tried the latest (experimental) lavfilters 0.59.1 and also the latest Lentoid (V2.0.1.17, Release Date: 2013/11/6) decoder but neigther open in graphstudio nor my sw dshow player nor avs script (l-smash LSMASHSource.dll) to virtualdub.
i've asked b4 but gotta ask again... other than the reliable but timeconsuming and converting back to the huge file method {tappdecoder -raw video.hm10} is there any new advances in timelining either raw hm10 or encoded/muxed to mp4 videos inside virtualdub for reviewing/analizing these videos ? -
@ Selur: even a DirectShowSource() script works fine, as I showed in the post featuring Sharon Stone
vhelp's XP setup is broken, period. IIRC, he uses a video capture device which is incompatible with Service Pack 3, to begin with.
Similar Threads
-
help - how to compile latest "nightly" ffmpeg for win32 (XP) with mingw
By hydra3333 in forum ProgrammingReplies: 32Last Post: 20th May 2017, 00:33 -
x265 vs x264
By deadrats in forum Video ConversionReplies: 71Last Post: 10th Jan 2016, 06:14 -
ffdcaenc (an upgrade to dcaenc)
By El Heggunte in forum AudioReplies: 22Last Post: 9th Dec 2014, 06:09 -
MulticoreWare Annouces x265/HEVC Mission Statement
By enim in forum Latest Video NewsReplies: 4Last Post: 9th Aug 2013, 22:09 -
New PC Build(s)
By thedeificone in forum ComputerReplies: 6Last Post: 25th May 2010, 16:57