Re: [FFmpeg-user] Levels and Primaries
Thank you guys for posting it, I've done just as requested but the dpx that is attached to the mail is stalling it because of it's size. It's awaiting moderation. I've also provided an mxf rendered by our studio with episode using the exact same dpx file. However that one is also awaiting moderation. In the mean time I'll just continue providing links as well to speed up the conversation. The dpx file can be found in the zip file that I refered to in my original post. Here you can find the episode rendered mxf file of the colorbars (Episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrMGZQX19oa1pPREU/edit?usp=sharing> ). And here are the generated waveforms from it: lightworks (lightworks waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrNzFWTGtZenRjLWM/edit?usp=sharing> ) ffmpeg (ffmpeg waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrblVYUGlEajF3cGc/edit?usp=sharing> ) smoke (smoke waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrbDBJbkpueDZDdE0/edit?usp=sharing> ). -- 42
On 3 February 2014 18:36, Rio Kierkels <riokierkels@gmail.com> wrote:
Thank you guys for posting it, I've done just as requested but the dpx that is attached to the mail is stalling it because of it's size. It's awaiting moderation. I've also provided an mxf rendered by our studio with episode using the exact same dpx file. However that one is also awaiting moderation.
In the mean time I'll just continue providing links as well to speed up the conversation. The dpx file can be found in the zip file that I refered to in my original post.
Here you can find the episode rendered mxf file of the colorbars (Episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrMGZQX19oa1pPREU/edit?usp=sharing> ). And here are the generated waveforms from it: lightworks (lightworks waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrNzFWTGtZenRjLWM/edit?usp=sharing> ) ffmpeg (ffmpeg waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrblVYUGlEajF3cGc/edit?usp=sharing> ) smoke (smoke waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrbDBJbkpueDZDdE0/edit?usp=sharing> ).
-- 42
Well my posts have bounced so I'll resort to posting links again. -- 42
On Mon, Feb 3, 2014, at 10:32 AM, Rio Kierkels wrote:
Well my posts have bounced so I'll resort to posting links again.
Sorry, I had to reject your messages because they were much too large (one was ~25 MB if I recall correctly). This mailing list has a limit of 100k. I understand it may complicate matters when it comes to providing console outputs and sample files, and I will likely increase this limit, but this is a high volume list with a large number of recipients. This should have resulted in a reply to you with the rejection notice indicating the message size limitation.
On 3 February 2014 21:02, Lou <lou@lrcd.com> wrote:
On Mon, Feb 3, 2014, at 10:32 AM, Rio Kierkels wrote:
Well my posts have bounced so I'll resort to posting links again.
Sorry, I had to reject your messages because they were much too large (one was ~25 MB if I recall correctly). This mailing list has a limit of 100k. I understand it may complicate matters when it comes to providing console outputs and sample files, and I will likely increase this limit, but this is a high volume list with a large number of recipients.
This should have resulted in a reply to you with the rejection notice indicating the message size limitation.
Yes thank you Lou, I did get it and I understand the size limitations.
From now on I'll post every text output here, It just seemed a bit excessive to add them to the original post. I'll repost the one with the dpx file but only a link to the file.
-- 42
Thanks Andy, If this is true it's good news for ffmpeg, bad news for the other apps and our current delivery pipeline. My next step would be to create a very basic mxf parser to determine the Y' values but as I've said my C skills are sub par to my python skills. Does anybody have some other nice tool to determine levels to cross check this. Or maybe someone who can verify that the ffmpeg mxf parser and mpeg2video decoder doesn't do anything to the levels when converting it to a waveform? On 4 February 2014 01:04, Phil Rhodes <phil_rhodes@rocketmail.com> wrote:
It just seemed a bit excessive to add them to the original post.
It is.
P _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user
-- 42
On 2/4/14, Rio Kierkels <riokierkels@gmail.com> wrote:
Thanks Andy,
If this is true it's good news for ffmpeg, bad news for the other apps and our current delivery pipeline. My next step would be to create a very basic mxf parser to determine the Y' values but as I've said my C skills are sub par to my python skills. Does anybody have some other nice tool to determine levels to cross check this. Or maybe someone who can verify that the ffmpeg mxf parser and mpeg2video decoder doesn't do anything to the levels when converting it to a waveform?
The only thing that could change waveform is colorspace change. It's not histogram filter job to play with colorspace stuff. You can see if swscale was used to do colorspace conversion with verbose ffmpeg output.
On 4 February 2014 01:04, Phil Rhodes <phil_rhodes@rocketmail.com> wrote:
It just seemed a bit excessive to add them to the original post.
It is.
P _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user
-- 42 _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user
Yes if I prepend the histogram filter with format=rgb24 I get a nice red green blue colored waveform, nice to know you can switch output styles. On 4 February 2014 12:05, Paul B Mahol <onemda@gmail.com> wrote:
On 2/4/14, Rio Kierkels <riokierkels@gmail.com> wrote:
Thanks Andy,
If this is true it's good news for ffmpeg, bad news for the other apps and our current delivery pipeline. My next step would be to create a very basic mxf parser to determine the Y' values but as I've said my C skills are sub par to my python skills. Does anybody have some other nice tool to determine levels to cross check this. Or maybe someone who can verify that the ffmpeg mxf parser and mpeg2video decoder doesn't do anything to the levels when converting it to a waveform?
The only thing that could change waveform is colorspace change. It's not histogram filter job to play with colorspace stuff. You can see if swscale was used to do colorspace conversion with verbose ffmpeg output.
On 4 February 2014 01:04, Phil Rhodes <phil_rhodes@rocketmail.com>
wrote:
It just seemed a bit excessive to add them to the original post.
It is.
P _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user
-- 42 _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user
_______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user
-- 42
Rio Kierkels wrote:
Thanks Andy,
If this is true it's good news for ffmpeg, bad news for the other apps and our current delivery pipeline. My next step would be to create a very basic mxf parser to determine the Y' values but as I've said my C skills are sub par to my python skills. Does anybody have some other nice tool to determine levels to cross check this. Or maybe someone who can verify that the ffmpeg mxf parser and mpeg2video decoder doesn't do anything to the levels when converting it to a waveform?
My C skills aren't up to much either so maybe you should have more confidence in something you write in python. Attached is what I used to look (cue someone spots blindingly obvious bug) of course looking means decoding with ffmpeg to rawvideo either to a file or a named pipe. So if you compile the file - gcc -Wall yuvp-counter.c -o count-yuv My sanity test file is 100% bars (+ subsampling artifacts) from ftp://ftp.tek.com/tv/test/streams/Element/MPEG-Video/625/100b_400.m2v for 422 or for 420 ftp://ftp.tek.com/tv/test/streams/Element/MPEG-Video/625/100b_060.m2v Save & make a named pipe somewhere - mkfifo myfifo In one terminal/konsole/whatever run ffmpeg -i 100b_400.m2v -f rawvideo -y myfifo Look at the output to get width, height and format (must be planar) and from another term set wide so you have a couple of hundred chars run eg. ./count-yuv 704 576 422 myfifo To me the output from that looks sane so I assume my counter works sometimes at least :-)
Rio Kierkels wrote:
Thank you guys for posting it, I've done just as requested but the dpx that is attached to the mail is stalling it because of it's size. It's awaiting moderation. I've also provided an mxf rendered by our studio with episode using the exact same dpx file. However that one is also awaiting moderation.
In the mean time I'll just continue providing links as well to speed up the conversation. The dpx file can be found in the zip file that I refered to in my original post.
Here you can find the episode rendered mxf file of the colorbars (Episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrMGZQX19oa1pPREU/edit?usp=sharing>
). Well I don't know anything about dpx or mxf other than what I just read on wikipedia :-) I do however have some crap C I wrote once to count yuv and it shows the levels on this as being wrong in that black is around 30 and white 218. If I use your ffmpeg command to make the mxf from the bars in the zip then the levels are roughly right (off by 1 and more spread than this). Given my limited understanding of dpx I am not sure how 709 and 601 apply to it - I thought they were matricies to convert to/from RGB and YUV, yet dpx seems more complicated than that (ie. it's not just RGB??).
And here are the generated waveforms from it: lightworks (lightworks waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrNzFWTGtZenRjLWM/edit?usp=sharing>
)
ffmpeg (ffmpeg waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrblVYUGlEajF3cGc/edit?usp=sharing>
)
smoke (smoke waveform episode rendered mxf colorbars<https://drive.google.com/file/d/0B9hzJJkCcuUrbDBJbkpueDZDdE0/edit?usp=sharing>
). So it looks like 0 and 100% with these tools are 16 and 235/240.
participants (5)
-
Andy Furniss -
Lou -
Paul B Mahol -
Phil Rhodes -
Rio Kierkels