Re: [FFmpeg-devel] [PATCH v2 1/2] avfilter/vf_blackdetect: add AVX2 SIMD version

From: Kacper Michajlow <kasper93@gmail.com>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH v2 1/2] avfilter/vf_blackdetect: add AVX2 SIMD version
Date: Fri, 18 Jul 2025 14:28:24 +0200
Message-ID: <CABPLASS4erKuF0bziTreZjCr962KcOniuq=7m49vzHzQRE5DOA@mail.gmail.com> (raw)
In-Reply-To: <CABGuwEnr25LE3+Mq4ZuY9vmzQ+Enp77J23+3s3KQAdbbsMBVsw@mail.gmail.com>

On Fri, 18 Jul 2025 at 14:14, Kieran Kunhya via ffmpeg-devel
<ffmpeg-devel@ffmpeg.org> wrote:
>
> > blackdetect8_c:                                        820.8 ( 1.00x)
> > blackdetect8_avx2:                                     219.2 ( 3.74x)
> > blackdetect16_c:                                       372.8 ( 1.00x)
> > blackdetect16_avx2:                                    201.4 ( 1.85x)
> >
> > Again, sorry for being pedantic here, but it gives the wrong
> > impression especially if you look at this from outside.
>
> Also misleading as far as I understand because GCC doesn't have
> runtime detection like FFmpeg.

It's not misleading if you say in the commit description (or anywhere)
what you are comparing against. Nice of you to conveniently strip this
part from my email.

> march=core-avx2 -ftree-vectorize:

I know it's good publicity to say something is 100x faster. Though,
I'm looking at this from a technical point of view, not the Twitter
point of view.

- Kacper
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".