Hello,

Your comparison used CRF encoding and the results showed decreased bitrate
as well as decreased PSNR. This is not sufficient to make a case for
improved compression efficiency. The thorough metric to use to judge this
would be BD-SSIM/BD-PSNR. You could also use --bitrate encoding, ensuring
similar bitrates for a less accurate comparison. A couple more points

a) SSIM is a more preferred metric with this team, since it correlates
better to visual quality
b) For selected sequences, visually comparing the 2 encodes would also be
very useful.




On Thu, Jul 28, 2016 at 9:30 PM, N Vijay Anand <
[email protected]> wrote:

> Min, others,
>
> Can we do comparison this way:
>
> 1. Generate bitstream with x265.exe at data rates (achieved with the
> patch).
>     Compare the psnr's in this scenario.
> 2. Generate bitstream with patch at data rates (that's reported by
> x265.exe)
>     and compare psnr's.
>
> Im sure there's an ineresting case here.
>
> Vijay
> --
> Principal Architect and Director,
> TriSpace Technologies Pvt Ltd.,
>
>
>
> _______________________________________________
> x265-devel mailing list
> [email protected]
> https://mailman.videolan.org/listinfo/x265-devel
>
>


-- 
Deepthi
_______________________________________________
x265-devel mailing list
[email protected]
https://mailman.videolan.org/listinfo/x265-devel

Reply via email to