[FFmpeg-devel] [PATCH] ffplay: support rotated video.

[Clément Bœsch]

On Fri, May 23, 2014 at 02:40:35AM -0400, Don Moir wrote:
> >>I guess what you are saying it might be something like "90.5" or something.
> 
> >Yes. 90.999 would raise 90 too unless you round btw.
> 
> >>Could be one degree off. Not sure it would display where someone would
> >>notice a one degree difference.
> 
> >There is no reason to drop that accuracy. The orientation matrix in MOV
> >has a relatively high one, there is no reason to drop it just to save 3
> >strcmp on 2-3 characters called only once at the beginning of the program.
> 
> Thanks for your explanation and not arguing about the strcmp. I was not
> aware it could have fractional degrees. Now that I do though, I probably am
> still going to drop that fraction. I doubt it makes a difference but I might
> just see what it does if I add a fractional degree to it like .99. Maybe you
> could also fake it in the code to see what kind of difference it might make.
> 
> Not sure why a video would have a fractional degree unless it was a bug.
> Just trying to be clear on it and dealing with width height display issues
> etc.

Well, if the format has an orientation matrix with enough precision to say
"this video stream should be rotated by 90.954123113 degrees" then it
probably means that, otherwise the format would just have contained 2 bits
for the 4 combinations of 0, 90, 180 and 270 degrees.

You want to see the difference? Well, try ffplay -f lavfi testsrc -vf
rotate="90.999*PI/180"

I wonder why you are arguing so much for dropping that precision. And no I
won't make that code rotate by 90 when the format says 90.0001 unless it's
stated in the specifications.

-- 
Clément B.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 490 bytes
Desc: not available
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20140523/21e22308/attachment.asc>