Max bit rate higher than expected

Rich Parry wrote on 12/6/2017, 1:37 PM

Analyzing a rendered file using VP13, MainConcept AVC/ACC .mp4 VBR codec (2 pass), I set average bit rate to 28Mbps and changed only the maximum bit rate for the test, 35Mbps and 28Mbps I was surprised to find that in both cases the final maximum bps far exceeded the maximum value I set.

  • With maximum set to 35Mbps that actual max is 50.5Mbps
  • With maximum set to 28Mbps the actual max is 41.3Mbps

This post isn't so much a question as an observation that surprised me. I assumed when I set a maximum bit rate, the final value would be close to what I set, especially when using 2 pass rendering.

    CPU Intel i9-13900K Raptor Lake

    Heat Sink Noctua  NH-D15 chromas, Black

    MB ASUS ProArt Z790 Creator WiFi

    OS Drive Samsung 990 PRO  NVME M.2 SSD 1TB

    Data Drive Samsung 870 EVO SATA 4TB

    Backup Drive Samsung 870 EVO SATA 4TB

    RAM Corsair Vengeance DDR5 64GB

    GPU ASUS NVDIA GeForce GTX 1080 Ti

    Case Fractal Torrent Black E-ATX

    PSU Corsair HX1000i 80 Plus Platinum

    OS MicroSoft Windows 11 Pro

    Rich in San Diego, CA

    Comments

    3POINT wrote on 12/6/2017, 2:07 PM

    When I read your diagrams, I can only discover a few peaks which exceed the max bitrate. The rest is just what you have set..

    Which sense does it make to set average bitrate and max bitrate to the same level (28Mbps)? You can use CBR for that purpose. 2Pass rendering will only give a closer final value to the average bitrate, not to the max bitrate. Max bitrate is depending on how much details and movement there is in your video.

    Rich Parry wrote on 12/6/2017, 2:36 PM

    3Point,

    Good points. As you mention, setting avg. and max. to same value doesn't make sense other than it was a test. For me, the results of this test show that the Max bit rate you set isn't the "hard limit" I was expecting. Wonder what CBR will yield, but that is a different test.

    To be clear, this is all academic, I just thought it an interesting test, I don't think it will affect my workflow other than to make me think about my settings a little harder.

    Thanks for your comments and insight.

    Rich

    Last changed by Rich Parry on 12/6/2017, 2:37 PM, changed a total of 2 times.

    CPU Intel i9-13900K Raptor Lake

    Heat Sink Noctua  NH-D15 chromas, Black

    MB ASUS ProArt Z790 Creator WiFi

    OS Drive Samsung 990 PRO  NVME M.2 SSD 1TB

    Data Drive Samsung 870 EVO SATA 4TB

    Backup Drive Samsung 870 EVO SATA 4TB

    RAM Corsair Vengeance DDR5 64GB

    GPU ASUS NVDIA GeForce GTX 1080 Ti

    Case Fractal Torrent Black E-ATX

    PSU Corsair HX1000i 80 Plus Platinum

    OS MicroSoft Windows 11 Pro

    Rich in San Diego, CA

    Marco. wrote on 12/6/2017, 2:37 PM

    When using VBR encoding, what you define for maximum and average rates are bit/sec, not bits/frame.

    Bitrate Viewer uses frame based (counted to seconds) analyze instead in its standard setting. Try setting Bitrate Viewer to "Second based" analyze and you'd probably will see different results.

    Edit:
    Nevermind, just saw you'd already set Bitrate Viewer to "Second based".

    3POINT wrote on 12/6/2017, 3:30 PM

    Just also try Sony AVC, there you only set the max bitrate. The average bitrate is calculated automatically and depends on amount of movement and details in your video. Be also astonished about quality and speed of render.

    Musicvid wrote on 12/6/2017, 4:41 PM

    Notice that the spikes coincide with apparent scene cuts. Automatic placement of new I-frames at the cut points may account for the momentary bitrate peaks.

    One way to test this would be a sideshow with dissimilar images and sharp cuts.

    Rich Parry wrote on 12/7/2017, 9:44 AM

    As another test I rendered a video using VP13, MainConcept AVC/ACC .mp4 codec (2 pass) with CBR of 28Mbps. I assumed the bit rate would be fairly constant, which it is, but there are still peaks above the CBR value set. Once again, this was an academic exercise for me, it merely confirms the maximum bit rate may exceed the value you set which I didn't know ... so the test was a success, I learned something.

    CPU Intel i9-13900K Raptor Lake

    Heat Sink Noctua  NH-D15 chromas, Black

    MB ASUS ProArt Z790 Creator WiFi

    OS Drive Samsung 990 PRO  NVME M.2 SSD 1TB

    Data Drive Samsung 870 EVO SATA 4TB

    Backup Drive Samsung 870 EVO SATA 4TB

    RAM Corsair Vengeance DDR5 64GB

    GPU ASUS NVDIA GeForce GTX 1080 Ti

    Case Fractal Torrent Black E-ATX

    PSU Corsair HX1000i 80 Plus Platinum

    OS MicroSoft Windows 11 Pro

    Rich in San Diego, CA