FFMetrics -- Yet another program for quality metrics calculation

Kommentare

fifonik schrieb am 19.09.2020 um 00:40 Uhr

Version 0.8.2b published. Changes:

- Change:    mkv, mts and mxf were added on video file extensions list
- Change: Scan type (progressive|interlaced) provided in metadata is honoured
- Bugfix:    'Clear All' button is not clearing files list in plot
- Bugfix:    Files list in plot not updated instantly when file added/removed
- Bugfix: VMAF values/graphs not displayed correctly when system number decimal separator is non a dot
 

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

jason-b schrieb am 23.09.2020 um 21:52 Uhr

@fifonik I think I've figured out the issue. First off, it's not HDCAM, it's XDCAM as I stated in my previous post. Here's a link with a bit of info.

It seems to be the SMPTE 436M track containing captions that's the issue. The issues have been addressed for ffmpeg (I haven't verified this entirely yet) with some libavformat libraries that can be enabled at compile-time.

My work-around for the task of checking quality has been to use ffmpeg to simply copy the video track into another mxf container without mapping the 436M track and your tool works perfectly. Thanks for looking into this.

Here's an output of one of those files with 436M:

Input #0, mxf, from 'sample.mxf':
  Metadata:
    uid             : 6c301b00-6388-05c1-01dd-080046020130
    generation_uid  : 6c301b00-6388-05c2-01dd-080046020130
    company_name    : SONY
    product_name    : Opt
    product_version : 1.22
    product_uid     : 060e2b34-0401-0103-0e06-012002030300
    modification_date: 2020-01-15T15:57:00.000000Z
    material_package_umid: 0x060A2B340101010501010D43130000002F421A006388058008004602013001DD
    timecode        : 00:57:06;38
  Duration: 00:33:53.97, start: 0.000000, bitrate: 60741 kb/s
    Stream #0:0: Video: mpeg2video (4:2:2), yuv422p(tv, bt709, progressive), 1280x720 [SAR 1:1 DAR 16:9], 50000 kb/s, 59.94 fps, 59.94 tbr, 59.94 tbn, 119.88 tbc
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:1: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:2: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:3: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:4: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:5: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:6: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:7: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:8: Audio: pcm_s24le, 48000 Hz, 1 channels, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:9: Data: none
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
      data_type       : vbi_vanc_smpte_436M
Unsupported codec with id 0 for input stream 9

fifonik schrieb am 23.09.2020 um 22:42 Uhr

Thanks for the media info provided.

It is possible that this is what causing the issue: "Stream #0:9: Data: none"

Will check my code when got home.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

fifonik schrieb am 24.09.2020 um 10:58 Uhr

Could you provide full output of the following command with your file:

ffmpeg -hide_banner -i sample.mxf -vframes 1 -t 1 -f null -

Thanks.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

jason-b schrieb am 24.09.2020 um 18:06 Uhr

@fifonik Yes, I think that libavformat.mxfdec might address the 436M track that isn't otherwise properly read, but I've yet to have time to validate. Here's the output you requested:

Input #0, mxf, from 'sample.mxf':
  Metadata:
    uid             : 6c301b00-6388-05c1-01dd-080046020130
    generation_uid  : 6c301b00-6388-05c2-01dd-080046020130
    company_name    : SONY
    product_name    : Opt
    product_version : 1.22
    product_uid     : 060e2b34-0401-0103-0e06-012002030300
    modification_date: 2020-01-15T15:57:00.000000Z
    material_package_umid: 0x060A2B340101010501010D43130000002F421A006388058008004602013001DD
    timecode        : 00:57:06;38
  Duration: 00:33:53.97, start: 0.000000, bitrate: 60741 kb/s
    Stream #0:0: Video: mpeg2video (4:2:2), yuv422p(tv, bt709, progressive), 1280x720 [SAR 1:1 DAR 16:9], 50000 kb/s, 59.94 fps, 59.94 tbr, 59.94 tbn, 119.88 tbc
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:1: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:2: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:3: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:4: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:5: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:6: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:7: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:8: Audio: pcm_s24le, 48000 Hz, mono, s32 (24 bit), 1152 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
    Stream #0:9: Data: none
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
      data_type       : vbi_vanc_smpte_436M
Stream mapping:
  Stream #0:0 -> #0:0 (mpeg2video (native) -> wrapped_avframe (native))
  Stream #0:1 -> #0:1 (pcm_s24le (native) -> pcm_s16le (native))
Press [q] to stop, [?] for help
Output #0, null, to 'pipe:':
  Metadata:
    uid             : 6c301b00-6388-05c1-01dd-080046020130
    generation_uid  : 6c301b00-6388-05c2-01dd-080046020130
    company_name    : SONY
    product_name    : Opt
    product_version : 1.22
    product_uid     : 060e2b34-0401-0103-0e06-012002030300
    modification_date: 2020-01-15T15:57:00.000000Z
    material_package_umid: 0x060A2B340101010501010D43130000002F421A006388058008004602013001DD
    timecode        : 00:57:06;38
    encoder         : Lavf57.83.100
    Stream #0:0: Video: wrapped_avframe, yuv422p, 1280x720 [SAR 1:1 DAR 16:9], q=2-31, 200 kb/s, 59.94 fps, 59.94 tbn, 59.94 tbc
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
      encoder         : Lavc57.107.100 wrapped_avframe
    Stream #0:1: Audio: pcm_s16le, 48000 Hz, mono, s16, 768 kb/s
    Metadata:
      file_package_umid: 0x060A2B340101010501010D431300000030421A006388058008004602013001DD
      encoder         : Lavc57.107.100 pcm_s16le
frame=    1 fps=0.0 q=-0.0 Lsize=N/A time=00:00:00.01 bitrate=N/A speed=1.58x    
video:1kB audio:2kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown

fifonik schrieb am 25.09.2020 um 01:50 Uhr

The command above is executed by FFMetrics to get ref's media info.

I supplied your output to the routine that parses such output and everything worked as it should. No errors, crashes, lockouts and returned parsed properties as they should be.

Without having problematic media, I can only try to add more debug output, unfortunately.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

fifonik schrieb am 27.09.2020 um 02:56 Uhr

@jason-b Sorry, I copy/pasted wrong frmpeg command.

To get media info the FFMetrics should execute the following:

ffmpeg -hide_banner -nostdin -i sample.mxf -frames:v 30 -filter:v idet -f null -

Could you check if this works and provide its output?

BTW, i was able to re-produce "locking" issue with one of file (mkv) I have on my PC.

Fixed that and PMed you a link to new build. Please check it and let me know if this issue is fixed for you.

Thanks.

Zuletzt geändert von fifonik am 27.09.2020, 03:29, insgesamt 1-mal geändert.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

fifonik schrieb am 28.09.2020 um 07:55 Uhr

Version 0.8.3b published. Changes:

- Change: Long file specs on the list truncating with ellipses in the middle of path (full file specs still can be visible as hints on mouse hover)

- Change: On getting media info now 60 frames analyzed (changed from 30)

- Bugfix: Program freezing while getting media info when some files added as ref #4 (deadlock while redirecting process strout/stderr)

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

fifonik schrieb am 09.10.2020 um 08:05 Uhr

Version 0.8.5b published. Changes:

- New: Colour format displayed in Ref's media info

- New: Media info displayed not only for Ref, but for processed files as well (full info available as ToolTip)

- Change: When Ref changes, calculated metrics/graphs are resetting

- Change: Calc button does not reset already calculated metrics/graphs

- Change: Colours are tied to file numbers. When you tick/untick files, lines on graphs will not change colours

- Change: '20 seconds' and '2 minutes' entries added to dropdown with processing lengths

- Change: FFMpeg version detection routine improved

- Bugfix: Manual editing Ref's path does not work correctly

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

fifonik schrieb am 24.10.2020 um 09:29 Uhr

Version 0.8.6b published. Changes:

- New: Program also get & display (in ToolTip) min, max, avg, std. dev. pop values based on frames' info.

- Change: webm added on video file extensions list

- Bugfix: version detection fail for some ffmpeg builds

Zuletzt geändert von fifonik am 24.10.2020, 09:30, insgesamt 1-mal geändert.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

almosely schrieb am 02.12.2020 um 23:37 Uhr

Hello fifonik,

recently I discovered your tool and started to use it frequently to optimize my encodings. A few things I realized - perhaps you could make them better.

1. (VMAF via) ffmpeg is blocking my pc almost entirely. It works fine, when I start your tool, then open the task manager and change the priority to lower as normal. Then ffmetrics calls ffmpeg with the same prio (lower as normal) and I can work with my pc almost normally. I tried to call ffmetrics with low prio at first place, but that did not start ffmpeg with low prio too. Maybe you could call ffmpeg internally with a lower prio?

2. Sometimes the displayed values (VMAF) within the GUI do not match the values displayed within the corresponding pop-ups.

3. I work with raw-files (.264, .hevc) and everytime I want to select one of these files I have to change to display "all files (*.*)". Maybe you could change this selection permanently to "all files" or you could take .264 and .hevc into account when leaving that option at default ("Vide Files"; there's a missing "o" in your GUI)?

And I think, the option "process ### seconds" does not work, if I remember correctly. But I do not need this.

Anyway: A great thank you for creating and providing your fine tool!! :-)

fifonik schrieb am 03.12.2020 um 04:24 Uhr

Thanks for feedback.

1. VMAF calculation is resource expensive so your suggestion makes sense. Will implement it.

2. Number in GUI field is rounded, while in tooltip it is displayed as it was received from ffmpeg. However, I just did multiple runs and found that numbers in tooltip not updating. Will fix it. Thanks.

3. *.h264 is already on the list. I will add *.264 & *.hevc. Also, will fix the typo. BTW, you can simply drag & drop files from explorer.

I just checked the "### seconds" issue you reported and it works as designed in .the latest 0.8.6b. I need more details if it does not work in your case (you can log commands and run from command line manually to double check).

Zuletzt geändert von fifonik am 03.12.2020, 09:14, insgesamt 2-mal geändert.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

fifonik schrieb am 03.12.2020 um 08:34 Uhr

Version 0.8.7b published. Changes:

-    New:    More detailed media info displayed per file in Tooltips
-    Change:    FFMpeg detection routing now checks available filters instead of configuration options (#7)
-    Change:    264 and hevc were added to video file extensions list
-    Change:    FFMpeg calculation process priority changed to "Below Normal" (media info analyze FFMpeg process priority was not changes and it is still "Normal")
-    Bugfix:    Metrics' values in Tooltips not updating on 2nd+ runs
-    Bugfix:    Metrics' values not cleared after "Reset"
-    Bugfix:    Tooltip with blank metric values still shown after "Reset"
-    Bugfix:    Typo in "Vide Files"

Zuletzt geändert von fifonik am 03.12.2020, 09:12, insgesamt 2-mal geändert.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

almosely schrieb am 03.12.2020 um 13:21 Uhr

Thank you very much! :-) As I said, I don't need the limited run function ("### seconds") because I work with avisynth and create my test-files fitting my needs. But I remember - at least I think so - that there was a crash or an endless run or a hung-up etc. when testing this function in the very beginning.

fifonik schrieb am 03.12.2020 um 22:58 Uhr

@almosely

I use this "### seconds" feature often during program testing to speed up process and it works fine in my case.

So, I'm not able to investigate/fix the issue until more details provided so I can re-produce it. Sorry.

 

BTW, Have you already checked the new version? Were your other issues resolved?

Thanks.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

almosely schrieb am 04.12.2020 um 13:18 Uhr

Hello fifonik,

I tested nearly 2 weeks every day countless files, but the last days I had no time (birthday and wedding). I will test more in a few days.

But I did a short test right now. FFMetrics is called with normal prio, ffmpeg with lower as normal. I let analyze two files and ffmpeg was call with lower as normal twice - so this seems to work fine :-)

The spelling ("o" in "Vide") is also corrected.

My raw-files (.264 and .hevc) are displayed now, without changing to "all files".

I did not make a check for differences within the GUI and pop-ups. I noted the lowes "peaks" within the pop-ups only the first analyzings. But later, with more experience, I realized that I don't need that info and started to only use the shortened VMAF-info within the GUI. So, if there are still differences between them two infos, I do not realize them.

I only mentioned that the "### seconds" option produced an error is because it produced an error :-) I do not use it, I only tested it at the very beginning of using your tool. I don't know the exact constellation anymore by which the error occured. I think, I let analyze PSNR, SSIM and VMAF and used long files. A short test with a short clip (3 min) reduced to 30 sec analyzing time and only VMAF did not produce an error right now.

dr0 schrieb am 15.01.2021 um 18:25 Uhr

What a great little program! Just what I was looking for!

I'm having difficulties getting VMAF to work tho. No matter what settings I apply, it always returns with an error.

I downloaded ffmpeg from here.

My program folder looks like this:

My vmaf-models folder looks like this:

Am I doing something wrong?

fifonik schrieb am 15.01.2021 um 22:51 Uhr

What is in FFMetrics.log?

If you use recent ffmpeg build, it have new version of libvmaf that expect json models, not pkl. FFMetrics does not work with json models yet. I'm aware about this issue and new version that support hson model will be published soon.

In a meantine, you can try to use a bit older ffmpeg build.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

JN- schrieb am 16.01.2021 um 01:51 Uhr

@dr0 As @fifonik suggests. The 19th. November 2020 version is the most recent ffmpeg that uses the .pkl files.

---------------------------------------------

VFR2CFR, Variable frame rate to Constant frame rate link to zip here.

Copies Video Converts Audio to AAC, link to zip here.

Convert 2 Lossless, link to ZIP here.

Convert Odd 2 Even (frame size), link to ZIP here

Benchmarking Continued thread + link to zip here

Codec Render Quality tables zip

---------------------------------------------

PC ... Corsair case, own build ...

CPU .. i9 9900K, iGpu UHD 630

Memory .. 32GB DDR4

Graphics card .. MSI RTX 2080 ti

Graphics driver .. latest studio

PSU .. Corsair 850i

Mboard .. Asus Z390 Code

 

Laptop… XMG

i9-11900k, iGpu n/a

Memory 64GB DDR4

Graphics card … Laptop RTX 3080

dr0 schrieb am 16.01.2021 um 16:50 Uhr

Thanks for your help guys. The previous FFmpeg release is indeed working with VMAF.

Is it normal that VMAF doesn't have a live graph and progress bar update (just frame number counter) while processing?

fifonik schrieb am 16.01.2021 um 23:12 Uhr

Is it normal that VMAF doesn't have a live graph and progress bar update (just frame number counter) while processing?

Yes, this is normal.

FFMpeg does not report metrics during VMAF calculation and write log only when process ended.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

dr0 schrieb am 17.01.2021 um 14:12 Uhr

One more minor thing that I noticed is that when I maximize FFmetrics' window to better see the graphs, a thin black line appears on the right side of the window.

I only tried to use this program in Windows 7, so I can't tell if this kind of behavior occurs in newer versions of Windows. Of course, it's really insignificant, but if this can be fixed the program would look even more polished than it is now.

fifonik schrieb am 18.01.2021 um 22:19 Uhr

I do not see such issue on my 3 PCs (different GPUs, screen resolutions, DPI scaling). So I think it is somehow related to Win7 (I have Win10 on my PCs).

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B450M MORTAR TITANIUM, CPU: AMD Ryzen 5700X, RAM: G'Skill 32 GB DDR4@3200, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

musko schrieb am 20.01.2021 um 16:25 Uhr

I'm on Win10 and same graphic issue as @dr0
This appear only in maximized window.