May 2019 - Cin - Cinelerra-GG Mailing Lists

Latest GIT checkin + ubuntu16 without-vaapi build
by Phyllis Smith 16 May '19

16 May '19

The latest GIT checkin contains many changes to draw refreshes in the Compositor and Viewer windows. A deadlock was recently discovered and it was determined that Mutex-s were being used when Condition-s should have been instead. This is a very good cleanup and even though we did quite a lot of testing, we could have missed something so if Cinelerra hangs unexpectedly, please do a Ctrl-C and get a dmp file to send to us for analysis. IgorB: a ubuntu16 build without libva is available at the following URL. Please let us know if this totally resolves your ./cin start up problem OR if libvdpau has to be removed also. It includes the above changes too which needs to be used more by users so it is better tested. gg/phyllis https://www.cinelerra-gg.org/download/testing/cinelerra-5.1-ub16-without-va…

4 17

New pkgs and tars for this month's release are ready
by Phyllis Smith 15 May '19

15 May '19

Latest builds are now available at: https://www.cinelerra-gg.org/download/tars/ https://www.cinelerra-gg.org/download/pkgs/ Description of GPU usage and the new Tracer plugin are temporarily at: https://www.cinelerra-gg.org/download/GPU_potential_speedup.pdf https://www.cinelerra-gg.org/download/Tracer_plugin.pdf Cin will NOT START without the system libva-dev (or libva) AND libvdpau-dev (or libvdpau) libraries installed. You will see an error message at startup like the following if missing: error while loading shared libraries: libva.so.1 cannot ... Release notes are at: https://www.cinelerra-gg.org/download/releasenotes.pdf or as shown in text below. -------------------------------------------------------------------------------------------------------------------------------------------- Cinelerra-GG (Version Infinity) Release Notes for 04/01/2019-04/30/2019 for builds *VIP* - if you have an error starting Cin, look at the error message in the window from where you started Cin. It might look something like this: error while loading shared libraries: libva.so.1 cannot … You will have to install your Operating System’s libva-dev and libvdpau-dev (or libva and libvdpau). *VIP* - if when using OpenGL, you encounter a crash occurring in the OpenGL (Nvidia?) code, you will have to use a workaround and set the environment variable: CIN_FINISH=0 until a better fix is made. For more details, please refer to MantisBT issue #201 on the cinelerra-gg.org website. 1. *Updated libvpx* from 1.7.0 to 1.8.0. *Upgraded LV2* library from version 14 to lv2-1.16.0. 2. *New native Tracer plugin* added for still images, but works with keyframes if moves. This plugin makes it easy to outline, that is “Trace”, an object with only a few points, kind of like a mask. There is a pdf file with the description of usage. 3. *Potential decode speedup with GPU* usage for some computers/some formats via vaapi and vdpau with the use of an environment variable CIN_HW_DEV=vdpau(or vaapi). Computers with Nvidia, Nouveau, Amdgpu drivers use vdpau and computers with Broadcom, Intel HD graphics, Radeon and other graphics device drivers use vaapi. Please see the pdf file for details. 4. *15 native plugins have been enhanced* by a user contributor to include a “Clear” button for individual sliders making it easy to reset just 1 instead of all. In addition a “Default” button for some of these plugins was added to revert to the plugin’s default settings. Plugins affected include: brightness, colorbalance, huesaturation, linearblur, polar, radialblur, rgbshift, sharpen, shiftinterlace, unsharp, wave, whirl, yuv, yuvshift, and zoomblur. Thank you very much to the contributor. 5. *Usability Improvements* A yellow colored “outline” has been added to the Camera view in the compositor in order to more easily view location. Some minor code improvements have been maded. You can now nest and unnest multiple clips instead of just 1 at a time. Added shortcut shift-d to delete the first track; change "create DVD" to alt-d instead. An Apply button has been added to the “Attach Effect” option menu so that multiple plugins can be added without having to reopen the window for each plugin. FPS and Codec have been added in the comment field in the Resources window for easy viewing. Enhanced ffmpeg error reporting to include file name that is causing a problem. In the Proxy menu, when the user checks Scaler on or off, that setting is retained for ease of use. For aesthetic purposes, changed the program code to draw a more symmetrical pulldown arrow. The Alt+LMB in the Sketcher has been replaced with Alt+Ctrl+LMB due to O/S grabbing first. 6. *Bugs/Issues fixed* Bug when you play a selection on the timeline and it stops, and when you play again, you have to hit space bar twice has been fixed. Updated shortcuts.html file. Rework update_lv2 for jitter bug + override LV2_PATH in order to avoid hangs for new usage An ffmpeg segv was occurring with PAL100/75 bars when not RGB; fix added to only not crash. Shuttle driver (not USB_DIRECT) for the new Linux kernel 5.x was broken, but now fixed. When using ffmpeg to render, a select file name to render to was not working if there was not at least one period or one slash. This has been corrected. A fix was added so that the camera and projector in the Compositor are less jittery. A patch was added to ffmpeg code to avoid a SEGV when using YUV format for certain plugins. Some non-standard characters, like /, that were not working right in the Titler have been corrected. A booby was caught in the Title plugin and has been fixed. In masking, a double lock problem in the cwindow and tool window is alleviated. Fader slider update for video is now correctly being updated just like the audios in the patchbay. The histogram plugin “histogram plot” checkbox when unchecked was still updating when Automatic is checked too. This has now been turned off. Ran Valgrind and found some memory leaks that were repaired. Added the diamond icon for Bezier versus the default square icon for Linear as used for some autos. A case for the Aging plugin was crashing on an edge case that was incorrectly fielded. The Sketcher and CriKey plugins had a grab problem but now the boundary test fixes this. Serious canvas height bug has been fixed – it was erroneously using “y” as the height. Some code changes involving the leaker test were improved – this just helps to find memory leaks. Build scripts have been modified to compensate for new GPU usage and to include shuttle. Clips folder thumbnails were popping/playing when in the “no play” mode has been fixed. CIN_FINISH variable added for OpenGL video driver (GL-ew) as a workaround for segv. Two tweaks in Mixer reload to solve a rare case that caused a crash. Unset In/Out pointer on “replace current project”. Lots of miscellaneous code cleanup. Fix to stop render engine races; Dialog joins fix. Proxy menu fix when scaler is enabled for the size to be correctly calculated. Removed the ffmpeg F_interlace plugin because it was problematic. Use F_tinterlace instead. Added Tracer info one line description. Fixed too small boxes on Sketcher and Motion51 plugins. In order to enable LV2 plugins for ubuntu14, had to add LANG=C99 to the build script.

5 13

Fwd: I tried to isolate x86 part of thirdparty mjpegtools from configure.ac
by Andrew Randrianasulu 14 May '19

14 May '19

No, better attempt (with code duplication): diff --git a/cinelerra-5.1/configure.ac b/cinelerra-5.1/configure.ac index 90048a2..5948fba 100644 --- a/cinelerra-5.1/configure.ac +++ b/cinelerra-5.1/configure.ac @@ -321,10 +321,17 @@ PKG_3RD([libvorbis],[auto], lib/.libs/libvorbisfile.a ], [ include . ]) +## arch dep tests +ARCH=[`uname -m`] +I86=[`expr "x$ARCH" : 'xi[346]86.*'`] +X86=[`expr "x$ARCH" : 'x..._64*'`] + +if test "x$I86$X86" != "x00" ; then + PKG_3RD([mjpegtools],[yes], [mjpegtools-2.1.0], - [ utils/mmxsse/.libs/libmmxsse.a \ - utils/.libs/libmjpegutils.a \ + [ utils/.libs/libmjpegutils.a \ + utils/mmxsse/.libs/libmmxsse.a \ lavtools/.libs/liblavfile.a \ lavtools/.libs/liblavjpeg.a \ mplex/.libs/libmplex2.a \ @@ -332,6 +339,21 @@ PKG_3RD([mjpegtools],[yes], mpeg2enc/.libs/libmpeg2encpp.a ], [ . lavtools utils ]) +else + + +PKG_3RD([mjpegtools],[yes], + [mjpegtools-2.1.0], + [ utils/.libs/libmjpegutils.a \ + lavtools/.libs/liblavfile.a \ + lavtools/.libs/liblavjpeg.a \ + mplex/.libs/libmplex2.a \ + yuvfilters/.libs/libyuvfilters.a \ + mpeg2enc/.libs/libmpeg2encpp.a ], + [ . lavtools utils ]) + +fi + PKG_3RD([openexr],[auto], [openexr-2.2.1], [ IlmImf/.libs/libIlmImf.a \ ------------------ there also was unrelated hunk about forcing libvpx (locally). survived ./autogen.sh + configure, global_config looks sane (in place where it talks about mjpegtools) ---------- Пересланное сообщение ---------- Тема: I tried to isolate x86 part of thirdparty mjpegtools from configure.ac Дата: Вторник 14 мая 2019 Отправитель: Andrew Randrianasulu <randrianasulu(a)gmail.com> Получатель: "Cinelerra.GG" <cin(a)lists.cinelerra-gg.org> ..but not sure if I did it correctly? ---------- ## arch dep tests ARCH=[`uname -m`] I86=[`expr "x$ARCH" : 'xi[346]86.*'`] X86=[`expr "x$ARCH" : 'x..._64*'`] PKG_3RD([mjpegtools],[yes], [mjpegtools-2.1.0], [ if test "x$I86$X86" != "x00" ; then utils/mmxsse/.libs/libmmxsse.a \ fi utils/.libs/libmjpegutils.a \ lavtools/.libs/liblavfile.a \ lavtools/.libs/liblavjpeg.a \ mplex/.libs/libmplex2.a \ yuvfilters/.libs/libyuvfilters.a \ mpeg2enc/.libs/libmpeg2encpp.a ], [ . lavtools utils ]) based on x86 asm test below .... -------------------------------------------------------

1 0

I tried to isolate x86 part of thirdparty mjpegtools from configure.ac
by Andrew Randrianasulu 14 May '19

14 May '19

..but not sure if I did it correctly? ---------- ## arch dep tests ARCH=[`uname -m`] I86=[`expr "x$ARCH" : 'xi[346]86.*'`] X86=[`expr "x$ARCH" : 'x..._64*'`] PKG_3RD([mjpegtools],[yes], [mjpegtools-2.1.0], [ if test "x$I86$X86" != "x00" ; then utils/mmxsse/.libs/libmmxsse.a \ fi utils/.libs/libmjpegutils.a \ lavtools/.libs/liblavfile.a \ lavtools/.libs/liblavjpeg.a \ mplex/.libs/libmplex2.a \ yuvfilters/.libs/libyuvfilters.a \ mpeg2enc/.libs/libmpeg2encpp.a ], [ . lavtools utils ]) based on x86 asm test below ....

1 0

FFmpeg and opencl (in case anyone with OpenCL hw will want to experiment)
by Andrew Randrianasulu 13 May '19

13 May '19

Hello again, while I successfully compiled Cin-GG with opencl support via ffmpeg - I have no hardware to test it (in theory I can install CPU-only OpenCL implementation, but then it will be sloooow ...) Still, I found some mystery I want to share: ENHANCED VIDEO PROCESSING WITH FFMPEG AND OPENCL Kelvin Phan – Massey University. 2016 Unfortunately it seems main page where this file was hosted is down at the moment, and archive.org has no copy of it, too ... so, Google cache was used: http://webcache.googleusercontent.com/search?q=cache:H-ISEoUS43UJ:https://k… whole 45 pages of it a bit too big for email text (body), so I just copy conclusion at very end: ======== 1. Introduction: In the old implementation (OpenCL version 1.0 to 1.2), the host and the device couldn’t share the same address space. To transfer memory data between host and device, buffers need to be created. One buffer for transferring data from host to device and another buffer for transferring data back from device to host. An offset needs to be passed to and from devices for accessing a location within a buffer such as accessing a region within an image. A host memory pointer cannot be used on the device. In OpenCL version 2.0, communication between host and device does not need to have a copied buffer. Shared Virtual Memory (SVM), host and device may use the same virtual address space. This address space can be effectively used to share virtual pointers created in this space. The data structures which are based on those memory pointers can be shared between host and device. [...] • OS: Ubuntu 14.04 (64 bits) • CPU: AMD Phenom(tm) II X4 955 Processor × 4 • GPU: AMD Radeon (TM) R7 360 Series [...] On the other hand, the same kind of result was expected to happen with un-sharp filter, but they was not. It is true that the filter with OpenCL enable gave better performing comparing without OpenCL enable, but the performing was not even nearly two times speed up as we see with de-shake filter. The possible reason for this is that the program took a lot more time to transfer the data between host buffers and device buffers, and there are a lot of transferring which needs to do. ============== hm, so, it was not as big win as author hoped ..yet, those filters appeared in mainline ffmpeg: http://ffmpeg.org/pipermail/ffmpeg-devel/2017-November/219828.html [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters ----------------- Changes since the last time this was posted: * Add unsharp filter (to replace existing unsharp). * Remove old experimental API. * Miscellaneous fixes. Now also tested with AMD OpenCL on Windows (DXVA2 mapping works nicely, D3D11 does not because it wants the Intel extension for NV12 support). Thanks, - Mark Silly example using everything (for i965 VAAPI + Beignet): ./ffmpeg_g -y -init_hw_device vaapi=va:/dev/dri/renderD128 -init_hw_device opencl=ocl at va -hwaccel vaapi -hwaccel_device va -hwaccel_output_format vaapi -i in.mp4 -f image2 -r 1 -i overlays/%d.png -an -filter_hw_device ocl -filter_complex '[1:v]format=yuva420p,hwupload[x2]; [0:v]scale_vaapi=1280:720:yuv420p,hwmap[x1]; [x1][x2]overlay_opencl=0:0,program_opencl=test.cl:rotate_image,unsharp_opencl=lx=17:ly=17:la=5,hwmap=derive_device=vaapi:reverse=1,scale_vaapi=1280:720:nv12' -c:v h264_vaapi -frames:v 1000 out.mp4 test.cl: __kernel void rotate_image(__write_only image2d_t dst, __read_only image2d_t src, unsigned int index) { const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE | CLK_FILTER_LINEAR); float angle = (float)index / 100; float2 dst_dim = convert_float2(get_image_dim(dst)); float2 src_dim = convert_float2(get_image_dim(src)); float2 dst_cen = dst_dim / 2; float2 src_cen = src_dim / 2; int2 dst_loc = (int2)(get_global_id(0), get_global_id(1)); float2 dst_pos = convert_float2(dst_loc) - dst_cen; float2 src_pos = { cos(angle) * dst_pos.x - sin(angle) * dst_pos.y, sin(angle) * dst_pos.x + cos(angle) * dst_pos.y }; src_pos = src_pos * src_dim / dst_dim; float2 src_loc = src_pos + src_cen; if (src_loc.x < 0 || src_loc.y < 0 || src_loc.x > src_dim.x || src_loc.y > src_dim.y) write_imagef(dst, dst_loc, 0.5); else write_imagef(dst, dst_loc, read_imagef(src, sampler, src_loc)); } ----------- and another patch was added in 2018, for tonemapping: https://patchwork.ffmpeg.org/patch/9032/ [FFmpeg-devel] lavfi: add opencl tonemap filter. -------------- This filter does HDR(HDR10/HLG) to SDR conversion with tone-mapping. An example command to use this filter with vaapi codecs: FFMPEG -init_hw_device vaapi=va:/dev/dri/renderD128 -init_hw_device \ opencl=ocl@va -hwaccel vaapi -hwaccel_device va -hwaccel_output_format \ vaapi -i INPUT -filter_hw_device ocl -filter_complex \ '[0:v]hwmap,tonemap_opencl=t=bt2020:tonemap=linear:format=p010[x1]; \ [x1]hwmap=derive_device=vaapi:reverse=1' -c:v hevc_vaapi -profile 2 OUTPUT ------------------- so, their command-line examples quite complex, and not sure if libavcodec can construct such pipeline automagically in case it was called by external code (cinelerra-GG in our case)......

1 0

Cin-GG git fails to fetch ffmpeg.git if requested
by Andrew Randrianasulu 13 May '19

13 May '19

Hello, I was about to try --with-ffmpeg-git configure option, but found it not working .... ------------ root@slax:/home/guest/botva/src/cinelerra-git/cin-5# TMP=/dev/shm/tmp setarch i686 ./cinelerra-goodguy.SlackBuild Клонирование в «cinelerra-goodguy-20190512»… remote: Counting objects: 10381, done. remote: Compressing objects: 100% (7640/7640), done. remote: Total 10381 (delta 2823), reused 9565 (delta 2585) Получение объектов: 100% (10381/10381), 127.52 MiB | 889.00 KiB/s, готово. Определение изменений: 100% (2823/2823), готово. Проверка патча cinelerra-5.1/libzmpeg3/Makefile… Часть #1 успешно применена на 77 (со сдвигом в 2 строки). Проверка патча cinelerra-5.1/mpeg2enc/Makefile… Часть #1 успешно применена на 54 (со сдвигом в 2 строки). Проверка патча cinelerra-5.1/mplexlo/Makefile… Часть #1 успешно применена на 11 (со сдвигом в 2 строки). Патч cinelerra-5.1/libzmpeg3/Makefile применен без ошибок. Патч cinelerra-5.1/mpeg2enc/Makefile применен без ошибок. Патч cinelerra-5.1/mplexlo/Makefile применен без ошибок. Проверка патча cinelerra-5.1/cinelerra/Makefile… Часть #1 успешно применена на 416 (со сдвигом в 33 строки). Часть #2 успешно применена на 491 (со сдвигом в 33 строки). Патч cinelerra-5.1/cinelerra/Makefile применен без ошибок. Проверка патча cinelerra-5.1/db/utils/Makefile… Часть #1 успешно применена на 7 (со сдвигом в 2 строки). Часть #2 успешно применена на 52 (со сдвигом в 6 строк). Проверка патча cinelerra-5.1/db/utils/xtv.C… Патч cinelerra-5.1/db/utils/Makefile применен без ошибок. Патч cinelerra-5.1/db/utils/xtv.C применен без ошибок. patching file cinelerra/main.C Hunk #1 succeeded at 147 with fuzz 2 (offset 30 lines). Проверка патча cinelerra-5.1/configure.ac… Часть #1 успешно применена на 321 (со сдвигом в 56 строк). Часть #2 успешно применена на 369 (со сдвигом в 56 строк). Патч cinelerra-5.1/configure.ac применен без ошибок. patching file configure.ac patching file configure.ac + rm -f global_config configure Makefile Makefile.in + rm -f aclocal.m4 depcomp compile install-sh ltmain.sh + rm -f config.log config.guess config.h config.h.in config.sub config.status missing + rm -rf autom4te.cache m4 + '[' '' = clean ']' + mkdir m4 + autoreconf --install /usr/share/aclocal/imlib.m4:9: warning: underquoted definition of AM_PATH_IMLIB /usr/share/aclocal/imlib.m4:9: run info Automake 'Extending aclocal' /usr/share/aclocal/imlib.m4:9: or see http://www.gnu.org/software/automake/manual/automake.html#Extending-aclocal configure.ac:6: installing './compile' configure.ac:4: installing './install-sh' configure.ac:4: installing './missing' configure: WARNING: unrecognized options: --with-opus checking for a BSD-compatible install... /usr/bin/ginstall -c checking whether build environment is sane... yes checking for a thread-safe mkdir -p... /usr/bin/mkdir -p checking for gawk... gawk checking whether make sets $(MAKE)... yes checking whether make supports nested variables... yes checking for style of include used by make... GNU checking for gcc... ccache gcc checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executables... checking whether we are cross compiling... no checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether ccache gcc accepts -g... yes checking for ccache gcc option to accept ISO C89... none needed checking whether ccache gcc understands -c and -o together... yes checking dependency style of ccache gcc... none checking dependency style of ccache gcc... none checking for g++... g++ checking whether we are using the GNU C++ compiler... yes checking whether g++ accepts -g... yes checking dependency style of g++... none checking for gtk_init in -lm... yes ls: cannot access '/usr/lib64': No such file or directory checking for nasm... yes checking nasm x264 compatible... yes checking for yasm... yes checking for objcopy... yes checking for pactl... no checking whether X_HAVE_UTF8_STRING is declared... yes checking for XOpenDisplay in -lX11... yes checking for X11/X.h... yes checking X11 headers... yes checking for X11/keysym.h... yes checking X11 headers... yes checking for XShmQueryExtension in -lXext... yes checking for X11/Xlib.h... yes checking Xlib XShm extention... yes checking for XineramaQueryExtension in -lXinerama... yes checking for X11/extensions/Xinerama.h... yes checking Xinerama headers... yes checking for XvQueryExtension in -lXv... yes checking for X11/Xlib.h... (cached) yes checking Xlib Xv extention... yes checking for XFixesQueryVersion in -lXfixes... yes checking for BZ2_bzDecompress in -lbz2... yes checking for FcInit in -lfontconfig... yes checking for FT_Init_FreeType in -lfreetype... yes checking for lzma_version_number in -llzma... yes checking for png_read_png in -lpng... yes checking for pthread_create in -lpthread... yes checking for gzopen in -lz... yes checking for fftw_execute in -lfftw3... yes checking for fftw3.h... yes checking fftw headers... yes checking for uuid/uuid.h... yes checking uuid headers... yes checking for linux/cdrom.h... yes checking linux cdrom headers... yes checking for sys/ioctl.h... yes checking linux sys... yes checking for FLAC__stream_decoder_new in -lFLAC... yes checking for FLAC/stream_decoder.h... yes checking FLAC headers... yes checking for lame_init in -lmp3lame... yes checking for lame/lame.h... yes checking lame headers... yes checking for jpeg_start_decompress in -ljpeg... yes checking for stdio.h... yes checking jpeg headers... yes checking for ogg_stream_init in -logg... yes checking for ogg/ogg.h... yes checking ogg headers... yes checking for opj_version in -lopenjp2... yes checking for openjpeg.h... yes checking openjpeg headers... yes checking for sf_open in -lsndfile... yes checking for sndfile.h... yes checking sndfile headers... yes checking for ImfOpenInputFile in -lIlmImf... no checking for theora_info_init in -ltheora... yes checking for theora/theoraenc.h... yes checking threora headers... yes checking for vorbis_encode_init in -lvorbisenc... yes checking for vorbis/vorbisenc.h... yes checking vorbis encoders headers... yes checking for ov_open in -lvorbisfile... yes checking for vorbis/vorbisfile.h... yes checking vorbis file headers... yes checking for vpx/vpx_decoder.h... yes checking vpx headers... yes checking for mjpegtools/mjpeg_types.h... yes checking mjpegtools headers... yes checking for TIFFOpen in -ltiff... yes checking for tiff.h... yes checking tiff headers... yes checking for twolame_encode_buffer_float32_interleaved in -ltwolame... yes checking for twolame.h... yes checking twolame headers... yes checking for x264_encoder_encode in -lx264... yes checking for stdint.h... yes checking x264 headers... yes checking for x265_encoder_encode in -lx265... yes checking for x265.h... yes checking x265 headers... yes checking for opus/opus_multistream.h... yes checking libopus headers... yes checking for aom_codec_version in -laom... no checking for WebPGetEncoderVersion in -lwebp... yes checking for webp/encode.h... yes checking libwebp headers... yes checking for a52.h... no checking a52 headers... no checking for encore.h... no checking encore headers... no checking for gif_lib.h... yes checking gif lib headers... yes checking for jbg_dec_init in -ljbig... no checking for vdp_device_create_x11 in -lvdpau... yes checking for vaInitialize in -lva... yes checking for va/va_x11.h... yes checking va x11 headers... yes checking for vaGetDisplay in -lva-x11... yes checking for va/va_drm.h... yes checking va drm headers... yes checking for vaGetDisplayDRM in -lva-drm... yes checking for glUseProgram in -lGL... yes checking for gluOrtho2D in -lGLU... yes checking for GL/gl.h... yes checking opengl headers... yes checking for sys/soundcard.h... yes checking oss headers... yes checking for XftInit in -lXft... yes checking for X11/Xlib.h... (cached) yes checking Xft/freetype headers... yes checking for snd_pcm_open in -lasound... yes checking for alsa/asoundlib.h... yes checking asound headers... yes checking for avc1394_init_target in -lavc1394... yes checking for libavc1394/avc1394.h... yes checking libavc1394 headers... yes checking for rom1394_get_bus_id in -lrom1394... yes checking for libavc1394/rom1394.h... yes checking librom1394 headers... yes checking for iec61883_mpeg2_recv_init in -liec61883... yes checking for libiec61883/iec61883.h... yes checking libiec61883 headers... yes checking for raw1394_iso_recv_init in -lraw1394... yes checking for libraw1394/raw1394.h... yes checking libraw1394 headers... yes checking for dv_init in -ldv... yes checking for libdv/dv.h... yes checking libdv headers... yes checking for linux/dvb/dmx.h... yes checking dvb device headers... yes checking for linux/kernel.h... yes checking v4l2 system headers... yes checking for X11/Xlib.h... (cached) yes checking XF86VM headers... yes checking for esd.h... yes checking esound headers... yes checking for audiofile.h... yes checking audiofile headers... yes checking for X11/keysymdef.h... yes checking x11 keysym defs... yes checking for libusb_init in -lusb-1.0... yes checking for libusb-1.0/libusb.h... yes checking libusb headers... yes checking for lilv_world_new in -llilv-0... no checking for sord_world_new in -lsord-0... no checking for serd_reader_new in -lserd-0... no checking for sratom_new in -lsratom-0... no checking for lilv/lilv.h... no checking lilv headers... no checking for suil_instance_new in -lsuil-0... no checking for dlopen in -ldl... yes checking for numa_alloc in -lnuma... no === want NUMA Failed. checking for openexr available... yes Reason Package ------ ------- forced a52dec forced djbfft disabled encore disabled audiofile disabled esound forced ffmpeg forced giflib forced ilmbase needed ladspa forced lame forced libraw1394 forced libogg forced libtheora forced libuuid forced libvorbis forced mjpegtools needed libaom needed openexr forced openjpeg forced twolame forced x264 forced x265 forced libvpx forced opus needed lv2 needed lilv needed sratom needed serd needed sord needed suil shared -ltheoraenc shared -ltheoradec shared -logg shared -lvorbis shared -lvorbisfile shared -logg shared -lfftw3 shared -lFLAC shared -lavc1394 shared -lrom1394 shared -liec61883 shared -ldv shared -ljpeg shared -lsndfile shared -lwebp shared -ltiff system -lX11 system -lXext system -lXinerama system -lXv system -lXfixes system -lbz2 system -lfontconfig system -lfreetype system -llzma system -lpng system -lpthread system -lz system -lvdpau system -lva system -lva-x11 system -lva-drm system -lGL system -lGLU system -lXft system -lasound system -lusb-1.0 system -ldl using: with-gl using: with-xft using: with-xxf86vm using: with-oss using: with-alsa using: with-firewire using: with-dv using: with-dvb using: with-video4linux2 using: without-esound using: without-pactl using: with-openexr using: with-lv2 using: with-commercial using: with-libzmpeg using: with-shuttle using: with-shuttle_usb using: with-jobs = 4 using: exec-name = cin using: with-cinlib = /usr/lib/cin using: with-cindat = /usr/share/cin using: with-config-dir = $$HOME/.bcast5 using: with-browser = firefox using: with-plugin-dir = /usr/lib/cin/plugins using: with-ladspa-dir = /usr/lib/cin/ladspa using: with-opencv = no using: with-git-ffmpeg = yes using: with-noelision = auto using: with-booby = yes using: with-libzmpeg = yes using: with-commerical = yes using: with-vaapi = yes using: with-vdpau = yes using: thirdparty build = yes using: single-user = no using: static-build = no using: ladspa-build = yes checking that generated files are newer than configure... done configure: creating ./config.status config.status: creating Makefile config.status: executing depfiles commands configure: WARNING: unrecognized options: --with-opus mkdir -p bin bin/applications bin/pixmaps bin/lv2 make -j4 all-recursive make[1]: Вход в каталог `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1' Making all in thirdparty make[2]: Вход в каталог `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty' make[2]: предупреждение: в суб-Makefile принудительно задан -jN; режим сервера задач запрещён mkdir /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/../thirdparty/build tar -xJf src/a52dec-0.7.4.tar.xz tar -xJf src/djbfft-0.76.tar.xz git clone yes ffmpeg.git tar -xJf src/twolame-0.3.13.tar.xz fatal: репозиторий «yes» не существует make[2]: *** [/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/../thirdparty/build/ffmpeg.source] Ошибка 128 make[2]: *** Ожидание завершения заданий... patch -d djbfft* -p1 < src/djbfft-0.76.patch1 patching file error.h touch /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/../thirdparty/build/djbfft.source patch -d a52dec* -p1 < src/a52dec-0.7.4.patch1 patching file liba52/imdct.c touch /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/../thirdparty/build/a52dec.source touch /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/../thirdparty/build/twolame.source make[2]: Выход из каталога `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty' make[1]: *** [all-recursive] Ошибка 1 make[1]: Выход из каталога `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1' make: *** [all] Ошибка 2 ------------------- as you can see it choked at "git clone yes ffmpeg.git" ... may be removing 'yes' will make it work, until finish or at least next err... Time to patch. PS: reason why I tried this was interesting commit in ffmpeg, removing nvenc from non-free: https://git.ffmpeg.org/gitweb/ffmpeg.git/commit/c8b50ff8f9b3c102765e8b8902d… ===== LICENSE: Remove nvenc from the non-free section. The headers are auto-detected by current FFmpeg. ===== Also, I found this text, explaining various options you can pass to hw encoders: https://gist.github.com/Brainiarc7/4b49f463a08377530df6cecb8171306a FFmpeg and libav's playbook: Advanced encoding options with hardware-based acceleration, NVIDIA's NVENC and Intel's VAAPI-based encoder. ----------------- ffmpeg -h encoder=h264_vaapi Encoder h264_vaapi [H.264/AVC (VAAPI)]: General capabilities: delay Threading capabilities: none Supported pixel formats: vaapi_vld h264_vaapi AVOptions: -qp <int> E..V.... Constant QP (for P-frames; scaled by qfactor/qoffset for I/B) (from 0 to 52) (default 20) -quality <int> E..V.... Set encode quality (trades off against speed, higher is faster) (from 0 to 8) (default 0) -low_power <int> E..V.... Use low-power encoding mode (experimental: only supported on some platforms, does not support all features) (from 0 to 1) (default 0) ----------------- Also, I wonder if some ffmpeg filters, based on va-api (deinterlace, scale, brightness ...) or opencl will work for at least Intel/AMD ... But then Cinelerra probably should understand what kind of hw (GPU) based filters attached to track, and not call download from GPU until first sw effect appear ... (a bit like it should work with OpenGL today).

2 2

Error if I forgot to create ffmpeg.git symlink in thirdparty folder
by Andrew Randrianasulu 13 May '19

13 May '19

make[2]: Вход в каталог `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo' make[2]: предупреждение: в суб-Makefile принудительно задан -jN; режим сервера задач запрещён g++ -o i686/mplexlo i686/mplex.o g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavutil/libavutil.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavcodec/libavcodec.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libpostproc/libpostproc.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavdevice/libavdevice.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavformat/libavformat.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libswscale/libswscale.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavfilter/libavfilter.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libswresample/libswresample.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavutil/libavutil.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavcodec/libavcodec.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libpostproc/libpostproc.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavdevice/libavdevice.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavformat/libavformat.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libswscale/libswscale.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libavfilter/libavfilter.a: Нет такого файла или каталога g++: ошибка: /dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo/../thirdparty/ffmpeg.git/libswresample/libswresample.a: Нет такого файла или каталога make[2]: *** [all] Ошибка 1 make[2]: Выход из каталога `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/mplexlo' make[1]: *** [all-recursive] Ошибка 1 make[1]: Выход из каталога `/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1' make: *** [all] Ошибка 2 ------------ so, sorry, it was NOT mpeg2enc, but mplexlo! ONLY if you enable building with external ffmeg.git!

1 0

Build with external ffmpeg fixed, thanks!
by Andrew Randrianasulu 13 May '19

13 May '19

by https://git.cinelerra-gg.org/git/?p=goodguy/cinelerra.git;a=commit;h=413642… while I still have some troubles, may be because I run configure and after it initially failed I just rerun make in cinelerra source directory I was forced to make ffmpeg.git symlink in thirdparty folder, because otherwise mpeg2enc part of cinelerra failed to link correctly. But if I made this symlinc before cloning/patching ffmpeg.git - patching will fail! Right now diff between my thirdparty/Makefile and official one looks like this: guest@slax:/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/ffmpeg$ diff -u /dev/shm/Makefile ../Makefile --- /dev/shm/Makefile 2019-05-13 04:23:58.274067188 +0300 +++ ../Makefile 2019-05-13 04:19:25.410065898 +0300 @@ -46,7 +46,8 @@ ext=$(lastword $(subst ., ,$(1))) tarball=$(if $($(1).git),$(1).git,$(lastword $(sort $(wildcard src/$(1)*.tar.*)))) unpack=$(call unpack_$(call ext,$(1)),$(1)) -unpack_git=git clone $($(1)) $(1) +#unpack_git=git clone $($(1)) $(1) +unpack_git=git clone --depth 1 https://git.ffmpeg.org/$(1) unpack_gz=tar -xzf $(1) unpack_bz2=tar -xjf $(1) unpack_xz=tar -xJf $(1) @@ -109,7 +110,7 @@ esound.ldflags=" -lm -lstdc++" fftw.cfg_params= --disable-fortran --enable-shared=no ffmpeg.cfg_params= \ - --enable-pthreads --enable-gpl --disable-ffplay \ + --enable-pthreads --enable-gpl --enable-opencl --enable-libdav1d --enable-frei0r --enable-libsnappy --enable-libvidstab --disable-ffplay \ $(if $(WANT_VAAPI),--enable-vaapi,--disable-vaapi) \ $(if $(WANT_VDPAU),--enable-vdpau,--disable-vdpau) \ $(call if_pkg,twolame,--enable-libtwolame) \ @@ -152,7 +153,7 @@ $(call ld_path,libvpx) \ $(call ld_path,x264) \ $(call ld_path,x265) \ - -Wl,--end-group -lm -lstdc++ -pthread \ + -Wl,--end-group -lm -lstdc++ -pthread -ldav1d -lvidstab \ $(EXTRA_LIBS)" $(FFMPEG_EXTRA_CFG) \ #cmake_config=echo "exec cmake \$$$$@ $(1)" > ./configure; chmod +x ./configure; ------- as you can see, I enabled libdav1d (0.2.1) installed locally, enabled opencl, frei0r (local), libsnappy (local, for hap encoding), and libvidstab (local). Liblensfun requred enabling (L)GPL v3 in ffmpeg config, so I omitted it for now. Full "no" list for ffmpeg's configure looks like this: guest@slax:/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty/ffmpeg$ ./configure --help | grep no --disable-logging do not log configure debug information not part of the dynamic linker search path and binaries will be under GPL [no] --enable-version3 upgrade (L)GPL to version 3 [no] --enable-nonfree allow use of nonfree code, the resulting libs and binaries will be unredistributable [no] --disable-static do not build static libraries [no] --enable-shared build shared libraries [no] --disable-autodetect disable automatically detected external libraries [no] --disable-programs do not build command line programs --disable-doc do not build documentation --disable-htmlpages do not build HTML documentation pages --disable-manpages do not build man documentation pages --disable-podpages do not build POD documentation pages --disable-txtpages do not build text documentation pages --enable-avresample enable libavresample build (deprecated) [no] --disable-network disable network support [no] will become enabled, if all their other dependencies are met and they are not Also note that the following help text describes the purpose of the libraries themselves, not all their features will necessarily be usable by FFmpeg. --enable-avisynth enable reading of AviSynth script files [no] --enable-chromaprint enable audio fingerprinting with chromaprint [no] --enable-frei0r enable frei0r video filtering [no] if openssl, librtmp or gmp is not used [no] if openssl or librtmp is not used [no] if openssl, libtls or mbedtls is not used [no] --enable-jni enable JNI support [no] --enable-ladspa enable LADSPA audio filtering [no] --enable-libaom enable AV1 video encoding/decoding via libaom [no] --enable-libaribb24 enable ARIB text and caption decoding via libaribb24 [no] needed for subtitles and ass filter [no] --enable-libbluray enable BluRay reading using libbluray [no] --enable-libbs2b enable bs2b DSP library [no] --enable-libcaca enable textual display using libcaca [no] --enable-libcelt enable CELT decoding via libcelt [no] --enable-libcdio enable audio CD grabbing with libcdio [no] --enable-libcodec2 enable codec2 en/decoding using libcodec2 [no] --enable-libdav1d enable AV1 decoding via libdav1d [no] --enable-libdavs2 enable AVS2 decoding via libdavs2 [no] and libraw1394 [no] --enable-libfdk-aac enable AAC de/encoding via libfdk-aac [no] --enable-libflite enable flite (voice synthesis) support via libflite [no] --enable-libfontconfig enable libfontconfig, useful for drawtext filter [no] --enable-libfreetype enable libfreetype, needed for drawtext filter [no] --enable-libfribidi enable libfribidi, improves drawtext filter [no] --enable-libgme enable Game Music Emu via libgme [no] --enable-libgsm enable GSM de/encoding via libgsm [no] --enable-libiec61883 enable iec61883 via libiec61883 [no] --enable-libilbc enable iLBC de/encoding via libilbc [no] --enable-libjack enable JACK audio sound server [no] --enable-libklvanc enable Kernel Labs VANC processing [no] --enable-libkvazaar enable HEVC encoding via libkvazaar [no] --enable-liblensfun enable lensfun lens correction [no] --enable-libmodplug enable ModPlug via libmodplug [no] --enable-libmp3lame enable MP3 encoding via libmp3lame [no] --enable-libopencore-amrnb enable AMR-NB de/encoding via libopencore-amrnb [no] --enable-libopencore-amrwb enable AMR-WB decoding via libopencore-amrwb [no] --enable-libopencv enable video filtering via libopencv [no] --enable-libopenh264 enable H.264 encoding via OpenH264 [no] --enable-libopenjpeg enable JPEG 2000 de/encoding via OpenJPEG [no] --enable-libopenmpt enable decoding tracked files via libopenmpt [no] --enable-libopus enable Opus de/encoding via libopus [no] --enable-libpulse enable Pulseaudio input via libpulse [no] --enable-librsvg enable SVG rasterization via librsvg [no] --enable-librubberband enable rubberband needed for rubberband filter [no] --enable-librtmp enable RTMP[E] support via librtmp [no] --enable-libshine enable fixed-point MP3 encoding via libshine [no] --enable-libsmbclient enable Samba protocol via libsmbclient [no] --enable-libsnappy enable Snappy compression, needed for hap encoding [no] --enable-libsoxr enable Include libsoxr resampling [no] --enable-libspeex enable Speex de/encoding via libspeex [no] --enable-libsrt enable Haivision SRT protocol via libsrt [no] --enable-libssh enable SFTP protocol via libssh [no] for DNN based filters like sr [no] --enable-libtesseract enable Tesseract, needed for ocr filter [no] --enable-libtheora enable Theora encoding via libtheora [no] if openssl, gnutls or mbedtls is not used [no] --enable-libtwolame enable MP2 encoding via libtwolame [no] --enable-libv4l2 enable libv4l2/v4l-utils [no] --enable-libvidstab enable video stabilization using vid.stab [no] --enable-libvmaf enable vmaf filter via libvmaf [no] --enable-libvo-amrwbenc enable AMR-WB encoding via libvo-amrwbenc [no] native implementation exists [no] --enable-libvpx enable VP8 and VP9 de/encoding via libvpx [no] --enable-libwavpack enable wavpack encoding via libwavpack [no] --enable-libwebp enable WebP encoding via libwebp [no] --enable-libx264 enable H.264 encoding via x264 [no] --enable-libx265 enable HEVC encoding via x265 [no] --enable-libxavs enable AVS encoding via xavs [no] --enable-libxavs2 enable AVS2 encoding via xavs2 [no] native MPEG-4/Xvid encoder exists [no] for dash demuxing support [no] --enable-libzimg enable z.lib, needed for zscale filter [no] --enable-libzmq enable message passing via libzmq [no] --enable-libzvbi enable teletext support via libzvbi [no] --enable-lv2 enable LV2 audio filtering [no] --enable-decklink enable Blackmagic DeckLink I/O support [no] if openssl, gnutls or libtls is not used [no] --enable-mediacodec enable Android MediaCodec support [no] --enable-libmysofa enable libmysofa, needed for sofalizer filter [no] --enable-openal enable OpenAL 1.1 capture support [no] --enable-opencl enable OpenCL processing [no] --enable-opengl enable OpenGL rendering [no] if gnutls, libtls or mbedtls is not used [no] Windows if openssl and gnutls are not used [autodetect] on OSX if openssl and gnutls are not used [autodetect] --enable-vapoursynth enable VapourSynth demuxer [no] --enable-cuda-nvcc enable Nvidia CUDA compiler [no] --enable-libdrm enable DRM code (Linux) [no] --enable-libmfx enable Intel MediaSDK (AKA Quick Sync Video) code via libmfx [no] --enable-libnpp enable Nvidia Performance Primitives-based code [no] --enable-mmal enable Broadcom Multi-Media Abstraction Layer (Raspberry Pi) via MMAL [no] --enable-omx enable OpenMAX IL code [no] --enable-omx-rpi enable OpenMAX IL code for Raspberry Pi [no] --enable-rkmpp enable Rockchip Media Process Platform code [no] --disable-amd3dnow disable 3DNow! optimizations --disable-amd3dnowext disable 3DNow! extended optimizations Cannot be combined with --target-exec --samples=PATH location of test samples for FATE, if not set use --disable-valgrind-backtrace do not print a backtrace under Valgrind --ignore-tests=TESTS comma-separated list (without "fate-" prefix in the name) of tests whose result is ignored ------------ so, plenty of things one can enable .....

1 0

Cinelerra-gg git + ffmpeg.git - still fails
by Andrew Randrianasulu 12 May '19

12 May '19

..at patching stage: root@slax:/dev/shm/tmp/cinelerra-goodguy-20190512/cinelerra-5.1/thirdparty# make git clone https://git.ffmpeg.org/ffmpeg.git Клонирование в «ffmpeg»… remote: Counting objects: 566891, done. remote: Compressing objects: 100% (121426/121426), done. remote: Total 566891 (delta 453622), reused 555771 (delta 444308) Получение объектов: 100% (566891/566891), 134.17 MiB | 1.78 MiB/s, готово. Определение изменений: 100% (453622/453622), готово. patch -d ffmpeg* -p1 < src/ffmpeg.git.patch1 patching file libavformat/bluray.c patch -d ffmpeg* -p1 < src/ffmpeg.git.patch2 patching file libavformat/mpegtsenc.c Hunk #1 FAILED at 56. Hunk #6 FAILED at 719. Hunk #7 succeeded at 742 (offset 8 lines). Hunk #8 succeeded at 767 (offset 8 lines). Hunk #9 succeeded at 776 (offset 8 lines). Hunk #10 succeeded at 831 (offset 8 lines). Hunk #11 succeeded at 918 (offset 8 lines). Hunk #12 succeeded at 956 (offset 8 lines). Hunk #13 succeeded at 1011 (offset 8 lines). Hunk #14 succeeded at 1058 (offset 8 lines). Hunk #15 succeeded at 1076 (offset 8 lines). Hunk #16 succeeded at 1145 (offset 8 lines). Hunk #17 succeeded at 1154 (offset 8 lines). Hunk #18 succeeded at 1190 (offset 8 lines). Hunk #19 succeeded at 1205 (offset 8 lines). Hunk #20 succeeded at 1213 (offset 8 lines). Hunk #21 succeeded at 1317 (offset 8 lines). Hunk #22 succeeded at 1494 (offset 8 lines). Hunk #23 succeeded at 1514 (offset 8 lines). Hunk #24 succeeded at 1710 (offset 8 lines). Hunk #25 succeeded at 1879 (offset 6 lines). Hunk #26 succeeded at 1928 (offset 6 lines). 2 out of 26 hunks FAILED -- saving rejects to file libavformat/mpegtsenc.c.rej ------------- reject is just those lines: --- libavformat/mpegtsenc.c 2018-04-20 04:02:57.000000000 -0600 +++ libavformat/mpegtsenc.c 2018-04-24 10:27:57.193689213 -0600 @@ -56,9 +56,8 @@ int sid; /* service ID */ char *name; char *provider_name; - int pcr_pid; - int pcr_packet_count; - int pcr_packet_period; + int64_t pcr, pcr_packet_timer, pcr_packet_period; + int pcr_sid, pcr_pid; AVProgram *program; } MpegTSService; @@ -719,6 +716,7 @@ service->pmt.pid = ts->pmt_start_pid + ts->nb_services; service->sid = sid; service->pcr_pid = 0x1fff; + service->pcr_sid = 0x1fff; service->provider_name = av_strdup(provider_name); service->name = av_strdup(name); if (!service->provider_name || !service->name) but I don't understand ffmpeg enough for reapplying them or even understand what they were doing! (probably blu-ray stuff?) Note, i hacked thirdparty/Makefile so it git clones ffmpeg only, not other projects: unpack_git=git clone https://git.ffmpeg.org/$(1) instead of unpack_git=git clone $($(1)) $(1)

1 0

LIMITED Hardware Acceleration using the GPU Encoding/Rendering checked in
by Phyllis Smith 09 May '19

09 May '19

VERY LIMITED hardware acceleration for rendering has been checked into GIT and a description is included below that will eventually make it into the manual. Limitations are described there and why, but here is a one line summary. *If you have an Intel computer with Intel graphics board and render to mp4, you can take advantage of this.* We have no hardware to test Intel computer with Nvidia graphics board but are assuming that it will not work. An AMD computer with the Nvidia graphics board installed does not work (not implemented). If you can prove me wrong, I would really like to hear that! The following URL, although not generic for all operating systems, does provide a Comparison tables for ascertaining whether or not the VA-API will work for your specific hardware. * Do not be fooled by the GE Force listed in the Decoding section *-- it does not appear in the Encoding section. The table is about 2/3 of the way down the article. https://wiki.archlinux.org/index.php/Hardware_video_acceleration @Ubuntu16 users: there is a test build at: https://www.cinelerra-gg.org/download/testing/cinelerra-5.1-ub16-x86_64-sta… @Pierre: if someone lets us know that it works on an Intel computer with Nvidia graphics, gg will make a Mint18 build but otherwise prepare to be disappointed. :) @Andrea: we did look a little bit at OpenCL but got the impression that the overhead may be too much and that it it seems to only help with filters. @Andrew: thanks for your last email on this subject as it provided the useful vdpauinfo amongst other things. ----------------------------------------------------------- *Rendering with Hardware Acceleration / Encoding using VA-API* Encoding using hardware acceleration of your graphics board GPU is included in CinelerraGG but it is of limited availability and may be exclusive to only a particular set of hardware graphics boards and only with certain ffmpeg formats. The encoding is done via vaapi (libva installed), which is known to work with Intel HD graphics board and some others. Usage of encoding as developed by Nvidia is not included with CinGG because ffmpeg has to be compiled with the *non-free *option and that goes against GPL. In addition the Nvidia encoding, referred to as *nvenc, *requires proprietary libraries to be installed. To use hardware acceleration for rendering (that is, encoding) you do not have to set a preference or an environment variable, as used by decoding. To use this feature you must use an ffmpeg render options file which specifies a vaapi codec, such as h264_vaapi. You must also include the line in the file to trigger the hardware probe: cin_hw_dev=vaapi There are currently 4 options files available in the Render menu already set up for you that you see when you select the Video wrench and use the down arrow on the first line in the menu. These are: h264_vaapi.mp4 # known to work on Intel computer with Intel Broadwell graphics driver mpeg2_vaapi.mp4 # known to work on Intel computer with Intel Broadwell graphics driver mjpeg_vaapi.mp4 # error message of “open failed with mjpeg_vaapi...” on above computer hevc_vaapi.mp4 # error message of “open failed with hevc_vaapi...” on above computer Other option files can be added as needed for your specific hardware if it is known to work for you, such as VP8 and VP9. The contents of the included cinelerra’s ffmpeg/video/h264_vaapi.mp4 file: mp4 h264_vaapi cin_hw_dev=vaapi profile=high According to an online wiki, hardware encoders usually create output of lower quality than some software encoders like x264, but are much faster and use less CPU. Keep this in mind as you might want to set a higher bitrate to get output of similar visual quality. Results of a particular test case performed on a Intel, 4-core computer, with Broadwell Graphics using an mp4 input video/audio file with dimensions of 1440x1080 / 29.97fps is shown next (note, tutorial.mp4). This may very well be a “best case” scenario! But clearly, at least on this computer with only 4 cores, the hardware acceleration seems to be advantageous. A comparison of the 2 output files using ydiff (as described in Appendix C in the manual) shows no obvious defects. *CPU usage* *Render Time* *Output File Size* *Options File* Without Hardware Acceleration 388% 100 secs. 36,862,542 h264.mp4 With vaapi Harware Acceleration 150% 19 secs. 74,522,736 h264_vaapi.mp4

2 2