Name | Modified | Size | Downloads / Week |
---|---|---|---|
Parent folder | |||
cutterff-1.3-src.tar.bz2 | 2024-07-13 | 253.4 kB | |
cutterff-1.3-win32-i386-gtk2.zip | 2024-07-13 | 9.5 MB | |
cutterff-1.3-linux-x86_64-gtk3.tar.bz2 | 2024-07-13 | 122.0 kB | |
cutterff-1.3-linux-x86_64-static-gtk3.tar.bz2 | 2024-07-13 | 4.7 MB | |
cutterff-1.3-linux-x86_64-gtk4.tar.bz2 | 2024-07-13 | 126.1 kB | |
cutterff-1.3-linux-x86_64-static-gtk4.tar.bz2 | 2024-07-13 | 4.7 MB | |
cutterff-1.3-linux-i386-gtk3.tar.bz2 | 2024-07-13 | 135.8 kB | |
cutterff-1.3-linux-i386-static-gtk3.tar.bz2 | 2024-07-13 | 4.5 MB | |
README.txt | 2024-07-13 | 3.0 kB | |
Totals: 9 Items | 24.1 MB | 1 |
README.txt ------- Note: ------- The linux versions need the libs from gtk-3.10, gtk-4.6 or higher versions of them, and if not statically linked, the libs from FFmpeg-7.0 The i386 are 32-bit, and x86_64 are 64-bit versions The 'cutterff-1.3-linux-*-static-gtk?.tar.bz2' versions have the libraries of FFmpeg-7.0 and the encoders 'libx264' and 'libx265' statically linked. The supported hardware accelerators are 'vaapi' and 'vdpau', and the shared libraries 'libva', 'libva-drm' and 'libvdpau' must be installed. The win32 version has the libs from gtk-2.24 statically linked, and the FFmpeg-7.0 libaries and the encoders 'libx264' and 'libx265' as 'dll' included. The supported hardware accelerators are 'd3d11va2' and 'dxva2'. Tested with Windows-7 (32-bit) and Windows-11 (64-bit) ----------- ChangeLog ----------- version 1.3 (Jul 13 2024) ----------- * gui.c + in 'cff_gui_file_selection_new_gettext' the filechooser gets a combo-box for selecting a hwaccel. A tooltip shows available decoders + in 'cff_gui_file_save' a tooltip shows the available (but disabled) encoders * ffmpeg.c + works with FFmpeg version 7.0 + memory leak in 'cff_ffmpeg_program_open', 'vframe.tmppkt' only unrefed, not freed. Now this is okay, see below + in 'cff_ffmpeg_file_open' the temporary packet and frames were allocated once, avoids allocating and freeing them in other functions + added 'vframe.has_keyframe', which is set in 'cff_ffmpeg_get_frame_dts'. If no keyframe was found, I-Frames was treated as keyframes. + if in 'cff_ffmpeg_get_frame_dts' a frame with no (unknown) picture type will return an error + in function 'cff_ffmpeg_get_frame' entering 'drain_mode' and storing the packet data (dts+pos) in 'video->vframe.pktlist' have been changed. Also rewind and searching again if the first packet after seeking has not at least a distance of seven frames to the next frame was removed + function 'cff_ffmpeg_copy_softcut_position_index' was removed in CutterFF-1.1, now it's back + decoding 'vp9' with multi-threading wasn't working. Setting the frame-type to 'slice' works. + encoding 'mpeg2' fails due invalid timebase. Fixed! + flushing the encoder in 'cff_ffmpeg_copy_softcut' was removed + removed 'kCFFStreamFlagBsfInit', copying the parameters from bsf to codec already done in 'cff_ffmpeg_copy_open' + in 'cff_ffmpeg_get_bitstream_filter_names' the filter is only shown if the stream has a codec for it + saving a cut list adds 'LastFrame', this is the dts of the last video frame. Needed if loading with a different decoder to calculate the offset + the function 'cff_ffmpeg_get_encoder_names' returns a negative value if encoding should be disabled * gprop.c + added item 'kPropStrFileHWAccel' /* EOF */