[HEVC] x265.EXE: mingw builds

[Brazil]

x265 Builds for Windows Cross Compiled with MinGW by snowfag

FYI those builds don't work on Windows XP: "Not a valid Win32 application".

[LigH.de]

Not everyone here still cares about obsolete operating systems... I only do because it was just a simple flag for the build script.

[El Heggunte]

Marchand has become a "post whøre". Quite unfortunate indeed.

[Gravitator]

Who has the desire make rip out this sample > Park Joy (mega.co.nz).
Preferably bitrate 6800-7000kb/s, and your recipe for maximum quality.

[Selur]

Sounds like a waste of time:
a. presets&co are still changing every few weeks (in example aq-mode 1 is now the default)
-> so suggesting any base line settings will be troublesome
b. optimizing settings for a single source and bit rate combination seems like a waste of time

better try to figure out which psy-rd settings make sense for what bit rate, since:

The lower the bitrate, the lower the optimal psycho-visual settings. If the bitrate is too low for the psycho-visual settings, you will begin to see temporal artifacts (motion judder). This is caused when the encoder is forced to code skip blocks (no residual) in areas of difficult motion because it is the best option psycho-visually (they have great amounts of energy and no residual cost). One can lower psy-rd settings when judder is happening, and allow the encoder to use some blur in these areas of high motion.

source: https://bitbucket.org/multicoreware/x265/commits/9c3b478a60b27aa118b0eb71e8d495016a028...c/reST/cli.rst
doesn't really help without some general hints how to decide what psy-rd settings are too low what bit rate on what content.

Cu Selur

[Gravitator]

Very convenient to pick up the settings on the passive stage, but that's very sad when'll see what happens when applied to complex scenes (garbage).

[mchartier105]

Who has the desire make rip out this sample > Park Joy (mega.co.nz).
Preferably bitrate 6800-7000kb/s, and your recipe for maximum quality.

6500kbps, Placebo, and some tweaked settings

[LigH.de]

I will call everyone recommending preset placebo a troll. As well as everyone asking for maximum quality at limited size.

That said ... another weekly: x265 1.4+209-6ba7be7b1697

[Gravitator]

6500kbps, Placebo, and some tweaked settings

Please adhere to the bitrate 6800-7000kb/s (abr/2-pass).
I can not start this file in MPC player (requires codecs) Looked through the converter Hybrid
- In general, you have a good work out

[Selur]

"requires codecs" HEVC shouldn't require any codecs with MPC-HC, assuming one isn't using an old version

[LigH.de]

MPC-HC (v1.7.7, Dec 2014) is not the same as MPC (v6.4.9.1, Feb. 2010); more than 4 years ago, HEVC was not even published as draft standard yet (that happened 2 years later).

[El Heggunte]

There's something odd with the file "park_joy - test (p).encode.mkv",

LAV Splitter 0.63.0 refuses to connect to the File Source filter

The thing plays fine with MPlayer and ffplay though

EDIT: apparently it's something to do with this

EDIT 2: just confirming, the upgrade to the latest revision (57ca857ffbb5) fixed the problem.

[Gravitator]

MPC-HC (v1.7.7, Dec 2014) is not the same as MPC (v6.4.9.1, Feb. 2010); more than 4 years ago, HEVC was not even published as draft standard yet (that happened 2 years later).

Sorry It was a MPC-HC 1.7.7 through LAV v0.6.3

EDIT 2: just confirming, the upgrade to the latest revision (57ca857ffbb5) fixed the problem.

Thanks for the news

[El Heggunte]

build 1.4+217-78ae7996a1ce

(compiled with GCC 4.9.2, finally \o/ )

Changes since commit 775ebb4

chenm003      295d033   fix: output wrong WppEntryOffset when emulating start code at end of WPP row 
Steve Borho   42fb030   doc: improve documentation for --stats and multi-pass in general 
Steve Borho   3315d6c   ppa: minimize code foot-print of profiling events
Steve Borho   952a2a3   ppa: simplify interfaces, enforce coding style
Steve Borho   6cbd7d2   ppa: refine event names 
Steve Borho   78ae799   ppa: emit one event per CTU for more clarity,
                             disable frame threads events

[El Heggunte]

build 1.4+219-8d2f418829c8

Changes since commit 78ae799:

David T Yuen     7b816fd     asm: idct[8x8] sse2 12232 -> 3500 over c code,
                             3550 -> 3500 over intrinsic

????????????     8d2f418     fix 4:4:4 rd<=1

[LigH.de]

Belated Merry Christmas: x265 1.4+223-1bf769c6953d

[El Heggunte]

build 1.4+236-8379ff016b56

[LigH.de]

Looks like some "Evangelion"?

[El Heggunte]

Looks like some "Evangelion"?

¿Anta baka? EVAngelion happens in 2015!!

[LigH.de]

Ah, didn't know. Just like "Back to the Future" went to 2014, and we invented Hoverboards?

Well: Happy and healthy 2015!

[Gravitator]

Ask to evaluate my > ideas
My goal is to give the opportunity to earn money rights holders and various developers (interest).

Happy New Year!!!

[Gravitator]

Boring to watch a lack of competitors x265

[LigH.de]

Happy New Year: x265 1.4_244-f255e8d06423

[Brazil]

Happy New Year: x265 1.4_244-f255e8d06423

Thanks!
Frohes neues Jahr - Happy New Year to you too

[El Heggunte]

@all: Obey her, or suffer the consequences

[El Heggunte]

build 1.4+251-feebd0ecda69

Changes since commit f255e8d:

Divya Manivannan   79b9453   asm: psyCost_pp_32x32 in sse4: improve 110849c->13373c

Divya Manivannan   b117e00   asm: psyCost_pp_64x64 in sse4: improve 417824c->56347c

chenm003           935eb35   testbench: fix bug in generate weight input data

Gopu G             e3039bc   encoder: initialize analysis data to null

Praveen Tiwari     a70b516   saoCuOrgB0: asm code

Praveen Tiwari     0b6c83d   sao.cpp: fixed compiler warnings

Praveen Tiwari     feebd0e   sao.cpp: fixed shadow warnings

[LigH.de]

x265 1.4+304-50a2071500dc

CBR is no "tuning" anymore, but a stricter ABR rate control mode. Can be enforced now with separate option "--[no-]strict-cbr". Previously published option "--ratetol" is gone again.

[El Heggunte]

build 1.4+374-8c66a97cd442

Changes since commit 50a2071: too many to count 0_o

{{ https://bitbucket.org/multicoreware/x265/commits/all }}

[Selur]

does any one have experience with the '--cbqpoffs X' and the '--crqpoffs X' setting? (seems to be the same as qp step in x264)

[LigH.de]

And a few more, also fixing a Win32 compiler warning...

x265 1.4+380-9d26082ca7b3