Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Official Low GOP topic, series 4
  • 586 Replies sorted by
  • LPowellLPowell

    @balazer The oscillation you see when the bitrate is forced up against the limit is the encoder switching in and out of Fallback Mode. When the encoder is working within safe limits, it uses T1 quantizer tables on its I-frames. In Fallback Mode it resorts to using T4 tables, which are very coarse and dramatically lower the QP factor of the encoder. The reason it oscillates is because the reduced bitrate of Fallback Mode relieves the internal pressure that forced it into Fallback Mode in the first place. But it then hits the ceiling once again and the cycle repeats.

    Fallback Mode is actually a safety valve that protects the encoder from failing at excessive bitrates. When it happens only on occasional bitrate peaks, its effects are too fleeting to be noticed. If you disable Fallback Mode with a Top Setting higher than Video Bitrate, the encoder may seize up at peak bitrates.

    The major obstacle to lowering the1080p QP to improve 24p image quality is the destabilizing effect it has on 1080i mode. The technique I devised in Flow Motion that allows 24H and FHS modes to peacefully coexist becomes very finicky when you try to push it much beyond 100Mbps or force it to work at low QP levels. That's why I couldn't simply tweak it up to say 120Mbps to make room for a lower QP.

    P.S. - at shutter speeds of 1/60 or slower, a reasonable amount of non-blurred motion can be handled solely by the keyframe rate. For sports action, you'd want 60p anyway, where Flow Motion works generously at 12 keyframes per second.

  • DisdausdebilDisdausdebil

    hey guys, here's a clip with Quantum 100 in action! Check it out! :)

  • JRaineyJRainey

    Cool

  • duartixduartix

    I'm beginning to wonder what advantage do B-Frames really bring...

    Let's face it, in their nature they make sense and can bring serious savings when there is motion in the footage that is non linear with time.

    Consider this timeline:

    1 23 4 56 7

    I BB P BB P

    Sure, a B-Frame at point #6 will most of the times look more similar to that P frame at point #7 than to the P-Frame at #4, but not by much, and you need to use at least a 2(+) B-Frame structure for that to occur.

    In reality how often does that happen?

    In reality how often do you have complex occlusions that occur within a 3 frame span?

    Now, adding insult to injury, let me show you what happened a few days ago when I was investigating a 2fps timelapse patch:

    I - 160K

    B - 8K

    B - 8K

    P - 3K

    B - 8K

    B - 8K

    P - 3K

    Surprised? I was! But it's normal, you see? Most of these frames are identical so the P/B frames are encoding null changes. Now why are the B-Frames so much bigger in this extreme case? Probably because their block addressing structure is a lot more complex, referring past and future frames something that doesn't need to be saved on a P-Frame.

    Now, if B-Frames are so much complex and they explore very particular cases of footage, how come they are so "efficient"?

    I suspect the answer lies in the quantization. B-Frames are small because they are heavily quantized and I mean a lot more heavily than P-Frames. I also suspect that given the same bitrate for the same footage, you could possibly get more OVERALL quality by using a 2 B-Frame structure but you wouldn't get more CONSTANT quality. I don't know, I haven't got the data, just a strong suspicion. :(

    @LPowell might have a better formed opinion on this...

  • jefelemurjefelemur

    Driftwood´s quantum 10 Nikkor 24mm 1,8 / Nikkor 55mm 2,0 Iso 3200 / 6400 B&W Noise reduction -2 MTS-ClipWarp-Final Cut-Export Image secuence-edit Dxo Film Pack3-Final cut ; syncro audio.

    http://www.notodofilmfest.com/#/Buscador/Ficha/31233/ http://www.notodofilmfest.com/index.php?corto=31233

  • driftwooddriftwood

    @Disdausdebil Wonderful colouring and superb editing. Love it. @jefelemur Funny! I'm like that when I want a shave! Good luck in the festival comp.

  • sebbensebben

    @balazer How does Cake compare to Quantum 100?

  • kodakmomentkodakmoment

    @LPowell what's the use of peaceful coexistence of 24p and 50i modes? My assumption would be that anyone shooting a project in 24p (and hopefully soon 25p) would not need a good 50i mode in the same patch and vice versa. Or is it for the folks who use de-interlaced 50i for slow motion? I would personally prefer 30p and 50p or 60p at 720 resolution and could perfectly live without a good 50i mode, or without an interlaced mode all together. To me the most useful modes in a patch would be one that allows for the best quality for handheld or moving scenes, regardless of the data rate, a second mode for static shots with best quality and one mode for rather long takes (interviews, debates, speeches etc.).That kind of recording doesn't have to be top quality. For anything else I don't see the point in saving card space. After transcoding to prores hq, dnxhd or cineform one ends up with high data rates anyhow. So maybe a one-patch-serves-all is not so useful taking hard disk prices into account?

  • RJHRJH

    Hey Guys, I shot this on an older Driftwood patch, think it was AQuarius. It's only a rough cut, lot's of elements missing, haven't had a chance to touch the sound, and no grading but take a look if you get a moment as I entered it into Empire and Jameson's Done in 60 Seconds competition and it's only just been included so we're almost a week behind in views. I am working on the final version now which will actually be longer than 1 minute. I hope you enjoy it:

    Blade Runner in 60 Seconds

  • x_worpig_xx_worpig_x

    @jefelemur One of the better films i've seen in a long time. Loved it.

  • driftwooddriftwood

    @RJH Brilliant...brilliant!

  • csynccsync

    @Disdausdebil - really wonderful. Any chance of a workflow description?

  • RJHRJH

    @driftwood Cheers mate! Thanks for your excellent patch, I'm looking forward to playing with Quantum on the next project.

  • rikyxxxrikyxxx

    @kodakmoment

    I'm one of the dudes using (properly) de-interlaced 1080i50 for SloMo and I can guarantee you that 720p50 hasn't got the same level of details :-)

  • kodakmomentkodakmoment

    @rikyxxx

    But why would you need a high quality 50i mode for that if you are loosing a lot of detail to de-interlacing in the first place? And what's the proper way to de-inderlace for you?

  • spacewigspacewig

    @jefelemur Really enjoyed it... hold on to that actor for future projects - apart from talent there's something very likable and compelling about him.

  • rikyxxxrikyxxx

    @kodakmoment

    Are you asking me why I don't use 1080p24 instead of 1080i50?

    Because with the right de-interlacing method (For instance BOB, Temporal & Spatial) I can get a quite good 1080p50 that means very smooth SloMo when slowed down.

  • mozesmozes

    i agree with @rikyxxx when 50i converted to 50p you get a incredible clean and smooth image.
    i use virtualdub with the bob doubler plugin for that,
    i wish there was a workaround for it in sony vegas pro, there is, but not so good as with Vdub.

  • artiswarartiswar

    Quick test of Quantum 9b after leaving cinema class at university. Shot by myself on a tripod, pulling over every few miles and repositioning.

    Shot on a 28mm/f2 Lomo, Sandisk 95/64GB card, and a cheapo Davidson and Stanford Tripod. Graded only with ColorGHears. About a 15 minute edit/grade process.

    Astonished at the quality. The night shots where around 1600-3200 ISO and the grain is there, but completely bearable.

  • balazerbalazer

    @sebben, I tested Cake against Quantum 100, and they are very very close in quality. Cake bit rates average less than half of Quantum 100's.

    I have started a new topic to discuss constant quality VBR approaches: http://www.personal-view.com/talks/discussion/2123/gh2-constant-quality-variable-bit-rate-encoder-settings

  • mozesmozes

    Just used Quantum 9b in sunny daylight, fails on all (high)settings when shutter go's up 1/400.
    Iso was set on 160.
    i didn't see that coming a was very surprised, all my test before just where without any problem
    I did use the natural color preset, and manual white balances...

    @artiswar wow, that looks good, and nice music by the way....

  • artiswarartiswar

    @mozes - Thanks! Little tribute to "Drive" I suppose. Quality soundtrack, that movie has.

  • LPowellLPowell

    @kodakmoment - "What's the use of peaceful coexistence of 24p and 50i modes? My assumption would be that anyone shooting a project in 24p (and hopefully soon 25p) would not need a good 50i mode..."

    If you have no personal use for FSH interlaced modes, then of course you may choose to sacrifice those modes in favor of tweaking the most our of 24p modes. For the 100Mbps Flow Motion Patch, my goal was to achieve optimal short-GOP performance without sacrificing any of the camera's modes or features. This makes the patch suitable for the widest possible range of interests and safe for use by novices as well as experts.

  • LPowellLPowell

    @duartix - "...if B-Frames are so much complex and they explore very particular cases of footage, how come they are so "efficient"?"

    As you suspect, much of the compression efficiency of B-frames has been achieved by using coarser quantizer tables than the tables used in I-frames. The justification for that is that the low-level residual data encoded by B-frames doesn't need to be encoded with as much precision as intra macroblock data.

    The flaw in that theory is what happens when the encoder fails to find suitable motion vectors for a sizable number of macroblocks in a B-frame. It is then forced to encode those macroblocks in intra-mode using the B-frame's coarse quantizer tables. For a consumer AVCHD encoder, this sacrifice of image quality for the sake of maintaining compression efficiency is evidently considered to be an acceptable tradeoff.

    Fortunately, Vitaliy has provided patches in the current PTool v3.63d that allow the firmware's built-in quantizer tables to be reassigned among the I, P, and B-frames used in the various video modes. I took advantage of this feature in my 100Mbps Flow Motion Patch to boost the quantizer table quality of P and B-frames to precision levels comparable to that of I-frames. Upcoming versions of PTool will allow me to craft fully customized quantizer tables that fine-tune the core encoding engine for optimal motion picture quality in each video mode.

  • driftwooddriftwood

    Listen up motherfuckers, if you want shit, then record with shit, if Ive wasted 3 months of my life trying to find fuckin A pictures out of this camera, then fuckit, tell me youre a better man than I... nuff said. Right? Stay with the plan.

This topic is closed.
← All Discussions