Seeking advice on videos from Samsung S8+ ? no HDR ?

[hydra3333]

Seeking advice on video files recorded by a Samsung S8+ - is there no HDR file as advertised ?

I just bought a "Samsung Galaxy S8 Plus", which in video recording mentions an auto-HDR mode and a HDR-on mode for some of the video resolutions (the phone actually tells me no HDR for above 1080p) ...

So, I took a sample video with each resolution/mode, and as the attached results show ... the results are all .mp4/h.264 files with no HDR?, zero, zilch, all 8-bit, every clip. At least that's what both mediainfo and ffprobe tell me.

I googled around and it is extremely difficult to find anything which mentions S8+ video recordings and codecs and formats let alone "HDR". Samsung's site says zilch about it.

1. I reckon I should have at least one 10-bit HDR clip because the phone's camera software it tells me I have set it to HDR mode - I don't know what to think, am I thinking something wrong and/or am I being gypped ?

2. Secondarily, the tools tell me the videos are all VFR - I don't yet know what that means for playing these clips on various devices, and if I need to convert them to a fixed framerate let alone how to to that conversion to a fixed framerate using ffmpeg.

Advice would be very much welcomed.

[Cornucopia]

All phones shoot vfr. It's a limitation of the technology (or at least a certain re-prioritization).


In stills/photo mode, there has for a while now been common use of the term HDR for scenes which do not technically qualify in their end result as hdr, but the technique has included hdr via a scene combination during an interim stage. What you get at the end is more accurately known by the technical term, "tone mapping". If you are familiar with gamma adjustments, or with dithering, this belongs in the same family of bit-depth reduction operators. Some cams/apps allow you to keep the original elements, a few even allow saving the interim true hdr files.

Seems Samsung chose to use the term for a video mode as well, though it is unclear what they are doing to arrive at some form of interim hdr like they truly can do with the multi-image combinations of stills. Maybe it's just Marketing.

Scott

[hydra3333]

Thanks for the info. I did some more g00gling and yes it is marketing, no HDR10 nor anything like 10bit etc.

A question remains whether I need to somehow convert these VFR videos to a fixed framerate (notionally 30fps I suppose, which is what the camera says it shoots in) in order to play .mp4 files on TVs or to a Chromecast Ultra etc. How would one go about that in, say, ffmpeg or nvencc?

[smrpix]

The only software I have found that successfully deals with variable frame rate and keeps sync is TMPGenc Video Mastering works. It is not free, but has even handled some pretty ugly skype footage.

FWIW, Adobe recommends handbrake and apps like Filmic Pro which allow you to shoot cfr in the first place. (I used it on my old iPhone, seemed to be ok.)

Have you tried just playing your files to the tv? Many players should be able to manage it these days.

[hydra3333]

Thanks, I will try them.

[Cornucopia]

To be clear, Filmic Pro can among many other cool things it can do can allow you to specify the intended frame rate and it can help limit the vfr variation swings to something that is more manageable, sync-wise. But it CANNOT force your footage to truly record in cfr, especially during lower light levels. No phone app can, iOS or Android because it is a limitation built into the (existing) OS, if not the hardware itself.
But there is good news on that front because Adobe announced that the newest version of Premiere Pro will support vfr source files, with provisions for choice of frame rate accuracy, smoothness, or audio sync. This might make moot the need for an interim Handbrake, etc cfr conversion step, and with it, the extra generational loss.

Scott

[hydra3333]

Nice !

So, in summary:

For the unwary buying a phone/camera based on the marketing ... you don't have to believe the marketing if you don't want to ... "HDR" in video recording on the S8/S8+ phone cameras is a marketing term for playing with colours somehow in phone software, not HDR10 10bit or anything like that.

ALL videos recorded by the S8/S8+ cameras are only h.264 8-bit progressive VFR .mp4s in various resolutions and colour spaces and profile levels as outlined in the spreadsheet attached above.

Now i need to go though the process of testing and as a result figuring out if I need to transcode/convert these VFR .mp4 videos from the s8+ for my uses (eg Chromecast Ultra and a Samsung 4k telly). I've asked in nvencc's githib issues list whether nvencc (i.e. gpu acceleration encoding into h.265) handles VFR, juist in case I choose to transcode the files in to HEVC for space purposes (I see results of circa 70fps with nvencc having gpu acceleration and only 6fps or so with x265 ... no contest really provided it works).

[hydra3333]

Easier to see the differences in s8/s8+ video camera .mp4 file characteristics in a .pdf (attached)

[hydra3333]

Well, an example for posterity below since g00gling didn't seem to help a lot. When I forget, I can some back here and discover it anew

Given the Samsung S8+ mobile phone doesn't seem to have huge variability in the VFR frame rate vs the nominal frame rate (usually 30 fps, ignoring the 60 or 240 framerate recording modes) something like this example seems to do the trick in transcoding with ffmpeg using NVENC hardware encoder (nvenc not to do the VFR, only as a speedy encoder; the opencl stuff is also just a hangover from use in other filter).

The "-vsync 1" setting is the one that does it, hence one can roll one's own transcode settings.

PS: the S8+ "HDR" video modes are not HDR10 or anything like that, it's marketed as HDR and is definitely only 8 bit, hence the 8-bit settings.
For UHD only, color space says bt2020nc (with and without "HDR" set) for the rest it says bt709 ... https://en.wikipedia.org/wiki/Rec._2020 "Rec. 2020, a standard for ultra high definition television (UHDTV) with standard dynamic range" which seems about right.
ffprobe says

pix_fmt=yuv420p
level=51
color_range=tv
color_space=bt2020nc
color_transfer=bt709
color_primaries=bt2020
chroma_location=left

mediainfo says

Resolution                               : 8
Resolution                               : 8 bits
Colorimetry                              : 4:2:0
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Chroma subsampling                       : 4:2:0
Bit depth                                : 8
Bit depth                                : 8 bits
Scan type                                : Progressive
Scan type                                : Progressive

Example:

REM -----------------------------------------------------------------------------------------------------------------
REM look for an output line like [Parsed_vfrdet_0 @ 000002c01d4ded80] VFR:0.824390 (169/36) min: 2991 max: 3003)
REM which apparently shows the number of (vfr/non-vfr) frames and the variability (min/max)
".\ffmpeg.exe" -hide_banner -v verbose -i ".\test-UHD-HDR-3840x2160.mp4" -map_metadata -1 -sws_flags lanczos+accurate_rnd+full_chroma_int+full_chroma_inp+print_info -filter:v vfrdet -f null -
[h264 @ 000002c01aebd4c0] Reinit context to 3840x2160, pix_fmt: yuv420p
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '.\test-UHD-HDR-3840x2160.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 0
    compatible_brands: isommp42
    creation_time   : 2018-03-01T09:22:16.000000Z
    location        : -34.7703+138.7038/
    location-eng    : -34.7703+138.7038/
    com.android.version: 7.0
  Duration: 00:00:06.86, start: 0.000000, bitrate: 48309 kb/s
    Stream #0:0(eng): Video: h264 (High), 1 reference frame (avc1 / 0x31637661), yuv420p(tv, bt2020nc/bt2020/bt709, left), 3840x2160, 48047 kb/s, SAR 1:1 DAR 16:9, 29.98 fps, 30 tbr, 90k tbn, 180k tbc (default)
    Metadata:
      rotate          : 90
      creation_time   : 2018-03-01T09:22:16.000000Z
      handler_name    : VideoHandle
    Side data:
      displaymatrix: rotation of -90.00 degrees
    Stream #0:1(eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 256 kb/s (default)
    Metadata:
      creation_time   : 2018-03-01T09:22:16.000000Z
      handler_name    : SoundHandle
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> wrapped_avframe (native))
  Stream #0:1 -> #0:1 (aac (native) -> pcm_s16le (native))
Press [q] to stop, [?] for help
[graph_1_in_0_1 @ 000002c01b52bcc0] tb:1/48000 samplefmt:fltp samplerate:48000 chlayout:0x3
[format_out_0_1 @ 000002c01b52c140] auto-inserting filter 'auto_resampler_0' between the filter 'Parsed_anull_0' and the filter 'format_out_0_1'
[auto_resampler_0 @ 000002c01b52c600] ch:2 chl:stereo fmt:fltp r:48000Hz -> ch:2 chl:stereo fmt:s16 r:48000Hz
[h264 @ 000002c01aeaaa00] Reinit context to 3840x2160, pix_fmt: yuv420p
[graph 0 input from stream 0:0 @ 000002c01d4dee80] w:3840 h:2160 pixfmt:yuv420p tb:1/90000 fr:30/1 sar:1/1 sws_param:flags=2
[transpose @ 000002c01af202c0] w:3840 h:2160 dir:1 -> w:2160 h:3840 rotation:clockwise vflip:0
Output #0, null, to 'pipe:':
  Metadata:
    encoder         : Lavf58.19.100
    Stream #0:0: Video: wrapped_avframe, 1 reference frame, yuv420p(left), 2160x3840 [SAR 1:1 DAR 9:16], q=2-31, 200 kb/s, 30 fps, 30 tbn, 30 tbc (default)
    Metadata:
      encoder         : Lavc58.33.101 wrapped_avframe
    Side data:
      displaymatrix: rotation of -0.00 degrees
    Stream #0:1: Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s (default)
    Metadata:
      encoder         : Lavc58.33.101 pcm_s16le
No more output streams to write to, finishing.:06.82 bitrate=N/A speed=0.614x
frame=  206 fps= 18 q=-0.0 Lsize=N/A time=00:00:06.86 bitrate=N/A speed=0.612x
video:108kB audio:1280kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
Input file #0 (.\test-UHD-HDR-3840x2160.mp4):
  Input stream #0:0 (video): 206 packets read (41207916 bytes); 206 frames decoded;
  Input stream #0:1 (audio): 320 packets read (218454 bytes); 320 frames decoded (327680 samples);
  Total: 526 packets (41426370 bytes) demuxed
Output file #0 (pipe:):
  Output stream #0:0 (video): 206 frames encoded; 206 packets muxed (110416 bytes);
  Output stream #0:1 (audio): 320 frames encoded (327680 samples); 320 packets muxed (1310720 bytes);
  Total: 526 packets (1421136 bytes) muxed
[Parsed_vfrdet_0 @ 000002c01d4ded80] VFR:0.824390 (169/36) min: 2991 max: 3003)
[AVIOContext @ 000002c01aeb4300] Statistics: 42053851 bytes read, 38 seeks
REM -----------------------------------------------------------------------------------------------------------------
REM Transcode to h.264 using the nvidia GPU hardware via the ffmpeg nvdec encoder
".\ffmpeg.exe" -hide_banner -v verbose -strict -1 -init_hw_device opencl=ocl:1.0 -filter_hw_device ocl -i ".\test-UHD-HDR-3840x2160.mp4" -vsync 1 -map_metadata -1 -sws_flags lanczos+accurate_rnd+full_chroma_int+full_chroma_inp+print_info -c:v h264_nvenc -pix_fmt nv12 -preset slow -profile:v high -level 5.1 -bf 2 -g 50 -refs 3 -rc:v vbr_hq -rc-lookahead:v 32 -cq 18 -qmin 14 -qmax 27 -coder 1 -pix_fmt yuv420p -c:a libfdk_aac -cutoff 18000 -ab 256k -ar 48000 -movflags +faststart -y ".\test-UHD-HDR-3840x2160.mp4.h264_nvenc.mp4"
Routing option strict to both codec and muxer layer
[AVHWDeviceContext @ 000001797924c4c0] 1.0: NVIDIA CUDA / GeForce GTX 1050 Ti
[AVHWDeviceContext @ 000001797924c4c0] DXVA2 to OpenCL mapping function found (clCreateFromDX9MediaSurfaceKHR).
[AVHWDeviceContext @ 000001797924c4c0] DXVA2 in OpenCL acquire function found (clEnqueueAcquireDX9MediaSurfacesKHR).
[AVHWDeviceContext @ 000001797924c4c0] DXVA2 in OpenCL release function found (clEnqueueReleaseDX9MediaSurfacesKHR).
[AVHWDeviceContext @ 000001797924c4c0] The cl_khr_d3d11_sharing extension is required for D3D11 to OpenCL mapping.
[AVHWDeviceContext @ 000001797924c4c0] D3D11 to OpenCL mapping not usable.
[h264 @ 0000017979261d40] Reinit context to 3840x2160, pix_fmt: yuv420p
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '.\test-UHD-HDR-3840x2160.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 0
    compatible_brands: isommp42
    creation_time   : 2018-03-01T09:22:16.000000Z
    location        : -34.7703+138.7038/
    location-eng    : -34.7703+138.7038/
    com.android.version: 7.0
  Duration: 00:00:06.86, start: 0.000000, bitrate: 48309 kb/s
    Stream #0:0(eng): Video: h264 (High), 1 reference frame (avc1 / 0x31637661), yuv420p(tv, bt2020nc/bt2020/bt709, left), 3840x2160, 48047 kb/s, SAR 1:1 DAR 16:9, 29.98 fps, 30 tbr, 90k tbn, 180k tbc (default)
    Metadata:
      rotate          : 90
      creation_time   : 2018-03-01T09:22:16.000000Z
      handler_name    : VideoHandle
    Side data:
      displaymatrix: rotation of -90.00 degrees
    Stream #0:1(eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 256 kb/s (default)
    Metadata:
      creation_time   : 2018-03-01T09:22:16.000000Z
      handler_name    : SoundHandle
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> h264 (h264_nvenc))
  Stream #0:1 -> #0:1 (aac (native) -> aac (libfdk_aac))
Press [q] to stop, [?] for help
[graph_1_in_0_1 @ 000001790a191d80] tb:1/48000 samplefmt:fltp samplerate:48000 chlayout:0x3
[format_out_0_1 @ 000001790a192140] auto-inserting filter 'auto_resampler_0' between the filter 'Parsed_anull_0' and the filter 'format_out_0_1'
[auto_resampler_0 @ 000001790a192640] ch:2 chl:stereo fmt:fltp r:48000Hz -> ch:2 chl:stereo fmt:s16 r:48000Hz
[h264 @ 00000179792c5640] Reinit context to 3840x2160, pix_fmt: yuv420p
[graph 0 input from stream 0:0 @ 00000179792d7b80] w:3840 h:2160 pixfmt:yuv420p tb:1/90000 fr:30/1 sar:1/1 sws_param:flags=2
[transpose @ 00000179792d7c80] w:3840 h:2160 dir:1 -> w:2160 h:3840 rotation:clockwise vflip:0
[h264_nvenc @ 000001790a180980] Loaded Nvenc version 8.1
[h264_nvenc @ 000001790a180980] Nvenc initialized successfully
[h264_nvenc @ 000001790a180980] 1 CUDA capable devices found
[h264_nvenc @ 000001790a180980] [ GPU #0 - < GeForce GTX 1050 Ti > has Compute SM 6.1 ]
[h264_nvenc @ 000001790a180980] supports NVENC
[h264_nvenc @ 000001790a180980] Lookahead enabled: depth 32, scenecut enabled, B-adapt enabled.
Output #0, mp4, to '.\test-UHD-HDR-3840x2160.mp4.h264_nvenc.mp4':
  Metadata:
    encoder         : Lavf58.19.100
    Stream #0:0: Video: h264 (h264_nvenc) (High), 1 reference frame (avc1 / 0x31637661), yuv420p(left), 2160x3840 [SAR 1:1 DAR 9:16], q=14-41, 2000 kb/s, 30 fps, 15360 tbn, 30 tbc (default)
    Metadata:
      encoder         : Lavc58.33.101 h264_nvenc
    Side data:
      cpb: bitrate max/min/avg: 0/0/2000000 buffer size: 4000000 vbv_delay: -1
      displaymatrix: rotation of -0.00 degrees
    Stream #0:1: Audio: aac (libfdk_aac) (mp4a / 0x6134706D), 48000 Hz, stereo, s16, delay 2048, 256 kb/s (default)
    Metadata:
      encoder         : Lavc58.33.101 libfdk_aac
No more output streams to write to, finishing.e=00:00:06.78 bitrate=1545.7kbits/s speed=1.44x
[mp4 @ 0000017979251280] Starting second pass: moving the moov atom to the beginning of the file
[AVIOContext @ 000001790a57eec0] Statistics: 2088956 bytes read, 0 seeks
frame=  206 fps= 37 q=40.0 Lsize=    2048kB time=00:00:06.82 bitrate=2457.8kbits/s speed=1.22x
video:1825kB audio:215kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.401933%
Input file #0 (.\test-UHD-HDR-3840x2160.mp4):
  Input stream #0:0 (video): 206 packets read (41207916 bytes); 206 frames decoded;
  Input stream #0:1 (audio): 320 packets read (218454 bytes); 320 frames decoded (327680 samples);
  Total: 526 packets (41426370 bytes) demuxed
Output file #0 (.\test-UHD-HDR-3840x2160.mp4.h264_nvenc.mp4):
  Output stream #0:0 (video): 206 frames encoded; 206 packets muxed (1868994 bytes);
  Output stream #0:1 (audio): 320 frames encoded (327680 samples); 322 packets muxed (219914 bytes);
  Total: 528 packets (2088908 bytes) muxed
[AVIOContext @ 00000179792c5500] Statistics: 4 seeks, 19 writeouts
[h264_nvenc @ 000001790a180980] Nvenc unloaded
[AVIOContext @ 0000017979257b80] Statistics: 42053851 bytes read, 38 seeks
REM -----------------------------------------------------------------------------------------------------------------
REM Transcode to hevc using the nvidia GPU hardware via the ffmpeg nvdec encoder
".\ffmpeg.exe" -hide_banner -v verbose -strict -1 -init_hw_device opencl=ocl:1.0 -filter_hw_device ocl -i ".\test-UHD-HDR-3840x2160.mp4" -vsync 1 -map_metadata -1 -sws_flags lanczos+accurate_rnd+full_chroma_int+full_chroma_inp+print_info -c:v hevc_nvenc -pix_fmt nv12 -preset slow -profile:v main -level 5.1 -rc vbr_hq -cq 22 -rc-lookahead 32 -spatial_aq 1 -pix_fmt yuv420p -c:a libfdk_aac -cutoff 18000 -ab 256k -ar 48000 -movflags +faststart -y ".\test-UHD-HDR-3840x2160.mp4.hevc_nvenc.mp4"
Routing option strict to both codec and muxer layer
[AVHWDeviceContext @ 0000013a413fc280] 1.0: NVIDIA CUDA / GeForce GTX 1050 Ti
[AVHWDeviceContext @ 0000013a413fc280] DXVA2 to OpenCL mapping function found (clCreateFromDX9MediaSurfaceKHR).
[AVHWDeviceContext @ 0000013a413fc280] DXVA2 in OpenCL acquire function found (clEnqueueAcquireDX9MediaSurfacesKHR).
[AVHWDeviceContext @ 0000013a413fc280] DXVA2 in OpenCL release function found (clEnqueueReleaseDX9MediaSurfacesKHR).
[AVHWDeviceContext @ 0000013a413fc280] The cl_khr_d3d11_sharing extension is required for D3D11 to OpenCL mapping.
[AVHWDeviceContext @ 0000013a413fc280] D3D11 to OpenCL mapping not usable.
[h264 @ 0000013a41411c00] Reinit context to 3840x2160, pix_fmt: yuv420p
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '.\test-UHD-HDR-3840x2160.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 0
    compatible_brands: isommp42
    creation_time   : 2018-03-01T09:22:16.000000Z
    location        : -34.7703+138.7038/
    location-eng    : -34.7703+138.7038/
    com.android.version: 7.0
  Duration: 00:00:06.86, start: 0.000000, bitrate: 48309 kb/s
    Stream #0:0(eng): Video: h264 (High), 1 reference frame (avc1 / 0x31637661), yuv420p(tv, bt2020nc/bt2020/bt709, left), 3840x2160, 48047 kb/s, SAR 1:1 DAR 16:9, 29.98 fps, 30 tbr, 90k tbn, 180k tbc (default)
    Metadata:
      rotate          : 90
      creation_time   : 2018-03-01T09:22:16.000000Z
      handler_name    : VideoHandle
    Side data:
      displaymatrix: rotation of -90.00 degrees
    Stream #0:1(eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 256 kb/s (default)
    Metadata:
      creation_time   : 2018-03-01T09:22:16.000000Z
      handler_name    : SoundHandle
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> hevc (hevc_nvenc))
  Stream #0:1 -> #0:1 (aac (native) -> aac (libfdk_aac))
Press [q] to stop, [?] for help
[graph_1_in_0_1 @ 0000013a523f28c0] tb:1/48000 samplefmt:fltp samplerate:48000 chlayout:0x3
[format_out_0_1 @ 0000013a523f2d40] auto-inserting filter 'auto_resampler_0' between the filter 'Parsed_anull_0' and the filter 'format_out_0_1'
[auto_resampler_0 @ 0000013a523f3180] ch:2 chl:stereo fmt:fltp r:48000Hz -> ch:2 chl:stereo fmt:s16 r:48000Hz
[h264 @ 0000013a41475ac0] Reinit context to 3840x2160, pix_fmt: yuv420p
[graph 0 input from stream 0:0 @ 0000013a41484440] w:3840 h:2160 pixfmt:yuv420p tb:1/90000 fr:30/1 sar:1/1 sws_param:flags=2
[transpose @ 0000013a573aed40] w:3840 h:2160 dir:1 -> w:2160 h:3840 rotation:clockwise vflip:0
[hevc_nvenc @ 0000013a52290500] Loaded Nvenc version 8.1
[hevc_nvenc @ 0000013a52290500] Nvenc initialized successfully
[hevc_nvenc @ 0000013a52290500] 1 CUDA capable devices found
[hevc_nvenc @ 0000013a52290500] [ GPU #0 - < GeForce GTX 1050 Ti > has Compute SM 6.1 ]
[hevc_nvenc @ 0000013a52290500] supports NVENC
[hevc_nvenc @ 0000013a52290500] AQ enabled.
[hevc_nvenc @ 0000013a52290500] Lookahead enabled: depth 32, scenecut enabled, B-adapt disabled.
Output #0, mp4, to '.\test-UHD-HDR-3840x2160.mp4.hevc_nvenc.mp4':
  Metadata:
    encoder         : Lavf58.19.100
    Stream #0:0: Video: hevc (hevc_nvenc) (Main), 1 reference frame (hev1 / 0x31766568), yuv420p(left), 2160x3840 [SAR 1:1 DAR 9:16], q=-1--1, 2000 kb/s, 30 fps, 15360 tbn, 30 tbc (default)
    Metadata:
      encoder         : Lavc58.33.101 hevc_nvenc
    Side data:
      cpb: bitrate max/min/avg: 0/0/2000000 buffer size: 4000000 vbv_delay: -1
      displaymatrix: rotation of -0.00 degrees
    Stream #0:1: Audio: aac (libfdk_aac) (mp4a / 0x6134706D), 48000 Hz, stereo, s16, delay 2048, 256 kb/s (default)
    Metadata:
      encoder         : Lavc58.33.101 libfdk_aac
No more output streams to write to, finishing.e=00:00:06.78 bitrate=1854.8kbits/s speed=   2x
[mp4 @ 0000013a41403040] Starting second pass: moving the moov atom to the beginning of the file
[AVIOContext @ 0000013a414cc3c0] Statistics: 2239622 bytes read, 0 seeks
frame=  206 fps= 52 q=39.0 Lsize=    2194kB time=00:00:06.83 bitrate=2630.5kbits/s speed=1.74x
video:1972kB audio:215kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.327740%
Input file #0 (.\test-UHD-HDR-3840x2160.mp4):
  Input stream #0:0 (video): 206 packets read (41207916 bytes); 206 frames decoded;
  Input stream #0:1 (audio): 320 packets read (218454 bytes); 320 frames decoded (327680 samples);
  Total: 526 packets (41426370 bytes) demuxed
Output file #0 (.\test-UHD-HDR-3840x2160.mp4.hevc_nvenc.mp4):
  Output stream #0:0 (video): 206 frames encoded; 206 packets muxed (2019664 bytes);
  Output stream #0:1 (audio): 320 frames encoded (327680 samples); 322 packets muxed (219914 bytes);
  Total: 528 packets (2239578 bytes) muxed
[AVIOContext @ 0000013a41475980] Statistics: 4 seeks, 21 writeouts
[hevc_nvenc @ 0000013a52290500] Nvenc unloaded
[AVIOContext @ 0000013a41408a40] Statistics: 42053851 bytes read, 38 seeks
REM -----------------------------------------------------------------------------------------------------------------
REM look for an output line like [Parsed_vfrdet_0 @ 000002c01d4ded80] VFR:0.824390 (169/36) min: 2991 max: 3003)
REM which apparently shows the number of (vfr/non-vfr) frames and the variability (min/max)
".\ffmpeg.exe" -hide_banner -v verbose -i ".\test-UHD-HDR-3840x2160.mp4.h264_nvenc.mp4" -map_metadata -1 -sws_flags lanczos+accurate_rnd+full_chroma_int+full_chroma_inp+print_info -filter:v vfrdet -f null -
[h264 @ 000001d5f2cfd140] Reinit context to 2160x3840, pix_fmt: yuv420p
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '.\test-UHD-HDR-3840x2160.mp4.h264_nvenc.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.19.100
  Duration: 00:00:06.87, start: 0.000000, bitrate: 2442 kb/s
    Stream #0:0(und): Video: h264 (High), 1 reference frame (avc1 / 0x31637661), yuv420p(left), 2160x3840 [SAR 1:1 DAR 9:16], 2177 kb/s, 30 fps, 30 tbr, 15360 tbn, 60 tbc (default)
    Metadata:
      handler_name    : VideoHandler
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 256 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> wrapped_avframe (native))
  Stream #0:1 -> #0:1 (aac (native) -> pcm_s16le (native))
Press [q] to stop, [?] for help
[h264 @ 000001d5f2d01200] Reinit context to 2160x3840, pix_fmt: yuv420p
[graph_1_in_0_1 @ 000001d5f32eee80] tb:1/48000 samplefmt:fltp samplerate:48000 chlayout:0x3
[format_out_0_1 @ 000001d5f3897ac0] auto-inserting filter 'auto_resampler_0' between the filter 'Parsed_anull_0' and the filter 'format_out_0_1'
[auto_resampler_0 @ 000001d5f3898f40] ch:2 chl:stereo fmt:fltp r:48000Hz -> ch:2 chl:stereo fmt:s16 r:48000Hz
[graph 0 input from stream 0:0 @ 000001d5f2d64b40] w:2160 h:3840 pixfmt:yuv420p tb:1/15360 fr:30/1 sar:1/1 sws_param:flags=2
Output #0, null, to 'pipe:':
  Metadata:
    encoder         : Lavf58.19.100
    Stream #0:0: Video: wrapped_avframe, 1 reference frame, yuv420p(left), 2160x3840 [SAR 1:1 DAR 9:16], q=2-31, 200 kb/s, 30 fps, 30 tbn, 30 tbc (default)
    Metadata:
      encoder         : Lavc58.33.101 wrapped_avframe
    Stream #0:1: Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s (default)
    Metadata:
      encoder         : Lavc58.33.101 pcm_s16le
No more output streams to write to, finishing.:06.46 bitrate=N/A speed=6.43x
frame=  206 fps=191 q=-0.0 Lsize=N/A time=00:00:06.86 bitrate=N/A speed=6.36x
video:108kB audio:1280kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
Input file #0 (.\test-UHD-HDR-3840x2160.mp4.h264_nvenc.mp4):
  Input stream #0:0 (video): 206 packets read (1868994 bytes); 206 frames decoded;
  Input stream #0:1 (audio): 322 packets read (219914 bytes); 320 frames decoded (327680 samples);
  Total: 528 packets (2088908 bytes) demuxed
Output file #0 (pipe:):
  Output stream #0:0 (video): 206 frames encoded; 206 packets muxed (110416 bytes);
  Output stream #0:1 (audio): 320 frames encoded (327680 samples); 320 packets muxed (1310720 bytes);
  Total: 526 packets (1421136 bytes) muxed
[Parsed_vfrdet_0 @ 000001d5f2d607c0] VFR:0.000000 (0/205)
[AVIOContext @ 000001d5f2cf42c0] Statistics: 2097304 bytes read, 0 seeks
REM -----------------------------------------------------------------------------------------------------------------
REM look for an output line like [Parsed_vfrdet_0 @ 000002c01d4ded80] VFR:0.824390 (169/36) min: 2991 max: 3003)
REM which apparently shows the number of (vfr/non-vfr) frames and the variability (min/max)
".\ffmpeg.exe" -hide_banner -v verbose -i ".\test-UHD-HDR-3840x2160.mp4.hevc_nvenc.mp4" -map_metadata -1 -sws_flags lanczos+accurate_rnd+full_chroma_int+full_chroma_inp+print_info -filter:v vfrdet -f null -
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '.\test-UHD-HDR-3840x2160.mp4.hevc_nvenc.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2mp41
    encoder         : Lavf58.19.100
  Duration: 00:00:06.87, start: 0.000000, bitrate: 2616 kb/s
    Stream #0:0(und): Video: hevc (Main), 1 reference frame (hev1 / 0x31766568), yuv420p(tv, progressive), 2160x3840 (2176x3840) [SAR 1:1 DAR 9:16], 2353 kb/s, 30 fps, 30 tbr, 15360 tbn, 30 tbc (default)
    Metadata:
      handler_name    : VideoHandler
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 256 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
Stream mapping:
  Stream #0:0 -> #0:0 (hevc (native) -> wrapped_avframe (native))
  Stream #0:1 -> #0:1 (aac (native) -> pcm_s16le (native))
Press [q] to stop, [?] for help
[graph_1_in_0_1 @ 000001917a70d600] tb:1/48000 samplefmt:fltp samplerate:48000 chlayout:0x3
[format_out_0_1 @ 000001917a70db00] auto-inserting filter 'auto_resampler_0' between the filter 'Parsed_anull_0' and the filter 'format_out_0_1'
[auto_resampler_0 @ 000001917a70df40] ch:2 chl:stereo fmt:fltp r:48000Hz -> ch:2 chl:stereo fmt:s16 r:48000Hz
[graph 0 input from stream 0:0 @ 000001917fd9edc0] w:2160 h:3840 pixfmt:yuv420p tb:1/15360 fr:30/1 sar:1/1 sws_param:flags=2
Output #0, null, to 'pipe:':
  Metadata:
    encoder         : Lavf58.19.100
    Stream #0:0: Video: wrapped_avframe, 1 reference frame, yuv420p, 2160x3840 [SAR 1:1 DAR 9:16], q=2-31, 200 kb/s, 30 fps, 30 tbn, 30 tbc (default)
    Metadata:
      encoder         : Lavc58.33.101 wrapped_avframe
    Stream #0:1: Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s (default)
    Metadata:
      encoder         : Lavc58.33.101 pcm_s16le
No more output streams to write to, finishing.:05.39 bitrate=N/A speed=5.39x
frame=  206 fps=162 q=-0.0 Lsize=N/A time=00:00:06.86 bitrate=N/A speed=5.39x
video:108kB audio:1280kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
Input file #0 (.\test-UHD-HDR-3840x2160.mp4.hevc_nvenc.mp4):
  Input stream #0:0 (video): 206 packets read (2019664 bytes); 206 frames decoded;
  Input stream #0:1 (audio): 322 packets read (219914 bytes); 320 frames decoded (327680 samples);
  Total: 528 packets (2239578 bytes) demuxed
Output file #0 (pipe:):
  Output stream #0:0 (video): 206 frames encoded; 206 packets muxed (110416 bytes);
  Output stream #0:1 (audio): 320 frames encoded (327680 samples); 320 packets muxed (1310720 bytes);
  Total: 526 packets (1421136 bytes) muxed
[Parsed_vfrdet_0 @ 000001917a73eec0] VFR:0.000000 (0/205)
[AVIOContext @ 000001917a1342c0] Statistics: 2246918 bytes read, 0 seeks
REM -----------------------------------------------------------------------------------------------------------------