Re: [FFmpeg-user] Reordering frames in corrupt GoPro videos

2024-07-17 Thread Mark Filipak
On 17/07/2024 16.54, Vincent Deconinck wrote: On Fri, Jul 12, 2024 at 1:04 AM Mark Filipak wrote: Try this: -filter_complex "shuffleframes=" It's worked for me. Hi Mark and Paul, Sorry for the delay. Thanks very much for the "shuffleframes" filter tip, this seems the right tool to use,

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Paul B Mahol
On Wed, Jul 17, 2024 at 10:16 PM Carl Zwanzig wrote: > On 7/17/2024 12:43 PM, Mark Filipak wrote: > >> frame; some projectors would use a three-blade shutter to get > >> supposedly smoother-looking motion. > > > Actually, z!, a double or triple shutter doesn't improve motion. It > > reduces/elimi

Re: [FFmpeg-user] Reordering frames in corrupt GoPro videos

2024-07-17 Thread Vincent Deconinck
On Fri, Jul 12, 2024 at 1:04 AM Mark Filipak wrote: > Try this: > -filter_complex "shuffleframes=" > It's worked for me. > Hi Mark and Paul, Sorry for the delay. Thanks very much for the "shuffleframes" filter tip, this seems the right tool to use, even if it implies reencoding. My tests co

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Carl Zwanzig
On 7/17/2024 12:43 PM, Mark Filipak wrote: frame; some projectors would use a three-blade shutter to get supposedly smoother-looking motion. Actually, z!, a double or triple shutter doesn't improve motion. It reduces/eliminates perceived flicker. Please read that again, I did not say it "imp

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Mark Filipak
On 17/07/2024 15.32, Carl Zwanzig wrote: On 7/17/2024 5:46 AM, Oliver Fromme wrote: The typical cinema frame rate (23.976 fps) is the minimum that's commonly accepted, In a real film cinema it's an exact 24fps however the shutter is quartered so there are two flashes per frame; some projector

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Carl Zwanzig
On 7/17/2024 5:46 AM, Oliver Fromme wrote: The typical cinema frame rate (23.976 fps) is the minimum that's commonly accepted, In a real film cinema it's an exact 24fps however the shutter is quartered so there are two flashes per frame; some projectors would use a three-blade shutter to get

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Mark Filipak
On 17/07/2024 14.46, Oliver Fromme wrote: Mark Filipak wrote: > On 17/07/2024 13.23, Oliver Fromme wrote: > > Mark Filipak wrote: > > > What if there's no such thing as frame in the future? Just think about it. > > > > What are you going to use instead? > > Pels. Pel is just anoth

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Oliver Fromme
Mark Filipak wrote: > On 17/07/2024 13.23, Oliver Fromme wrote: > > Mark Filipak wrote: > > > What if there's no such thing as frame in the future? Just think about > > > it. > > > > What are you going to use instead? > > Pels. Pel is just another name for pixel: https://en.wikipedia.o

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Mark Filipak
On 17/07/2024 13.23, Oliver Fromme wrote: Mark Filipak wrote: > What if there's no such thing as frame in the future? Just think about it. What are you going to use instead? Pels. You somehow need to organize screen updates -- that is, when the contents of the screen change. You have to d

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Oliver Fromme
Mark Filipak wrote: > What if there's no such thing as frame in the future? Just think about it. What are you going to use instead? You somehow need to organize screen updates -- that is, when the contents of the screen change. You have to describe the change somehow, and attach a time stamp to

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Mark Filipak
On 17/07/2024 12.30, Oliver Fromme wrote: Mark Filipak wrote: > The weighting of Y Cr Cb are still based on CRT phosphors, not reality. HSV is reality. No, that's wrong, HSV is not "reality". I'm afraid it's much more complicated than that. Oliver, if we continue, you will resent me 'wasti

[FFmpeg-user] GOPs in AVC

2024-07-17 Thread Mark Filipak
I'm sorry my posts are so technical. AVC is technical. I have 'attacked' my 40 second m2ts with a hex editor because other tools are inadequate. I have not found any GOPs. I've found GOP-like start codes (0x00 00 01 B8) but the next 4 bytes are not correct for a GOP header. For example: 00 00

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Oliver Fromme
Mark Filipak wrote: > The weighting of Y Cr Cb are still based on CRT phosphors, not reality. HSV > is reality. No, that's wrong, HSV is not "reality". I'm afraid it's much more complicated than that. RGB, YCbCr and HSV are just color models (sometimes also called color spaces, but that's an

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Mark Filipak
On 17/07/2024 08.46, Oliver Fromme wrote: I'm not sure this is on-topic for the ffmpeg users list ... It is, or rather, it will be some day... Imagine video without frames, imagine how it would work, imagine pixel-by-pixel random access and you're on your way. There would be no game play or

Re: [FFmpeg-user] The future of video

2024-07-17 Thread Oliver Fromme
I'm not sure this is on-topic for the ffmpeg users list ... Anyway, I'll try to comment on some of your points. Mark Filipak wrote: >I interviewed at NVIDIA when it was a start-up. My interview wound up > being an argument over > sprite graphics, which they held in contempt. I did not get

Re: [FFmpeg-user] pict_type="I"

2024-07-17 Thread Oliver Fromme
Mark Filipak wrote: > Thank you for your summary, Oliver. That has really, really helped. As you > know, H.264 is > prescriptive, not descriptive, which makes understanding it very difficult. > > It also lacks the strings: "I[- ]frame" "P[- ]frame" "B[- ]frame" "I[- > ]picture" "P[- ]pict

Re: [FFmpeg-user] The future of video

2024-07-17 Thread MacFH - C E Macfarlane - News
On 17/07/2024 01:01, Mark Filipak wrote: On 16/07/2024 19.40, MacFH - C E Macfarlane - News wrote: On 17/07/2024 00:02, Mark Filipak wrote:    To match human persistence of vision, the refresh rate of individual pels actually needs to be no greater than 10 or 12 refreshes per second. Only th