[x264-devel] bug report

[Attila József]

Thank you

All the "non smart" amiko boxes has very similar software (and maybe same
decoder), thats why all of them affected by this issue. Even the "amiko
impulse 265" (that box is works well with x265, the above x264 error also
occurs here ), and even the "amiko 8142 twin c", even though this one is
very new product.

I dont think they fix the older models. The amiko support is very poor. The
biggest issues will fixed, but the others is not.

All the boxes works well with nvenc, or cable tv company used hardware
encoders. There were mystical errors earlier with some options on the
encoder, but complete fails with play h264 streams not.

One more questions. If this commit is stay, than what can i do besides, I
use an older version of x264 or revert manually that commit every time? Is
there any other solution for that?

Henrik Gramner <henrik at gramner.com> ezt írta (időpont: 2019. jún. 24., H,
17:36):

> On Mon, Jun 24, 2019 at 5:07 PM Attila József <user1894 at gmail.com> wrote:
> >
> > Hi.
> >
> > I want to report an interesting bug.
> >
> > I use ffmpeg with x264 lib for lot of tasks. Nowadays, i realize that
> some encoded video is not playable on some set-top boxes anymore. I
> started  lokking for the error. After lot of compilling, and testing, i
> found exactly one commit, that related to this problem. Before that commit,
> encoded video works perfectly on every device. After this, only with
> baseline profile, and/or with interlaced fields.
> >
> > the commit:
> > ----
> > 92d36908cbafd2a6edf7e61d69f341027b57f6f8
> >
> > Author: Yusuke Nakamura <muken.the.vfrmaniac at gmail.com>
> > Date:   Mon Apr 9 11:01:28 2018 +0900
> >
> >     Signal Progressive and Constrained profiles
> >
> >     Progressive High, Constrained High, and Progressive High 10.
> >
> >     Even in Main profile, constraint_set4_flag is now set to 1 if
> progressive,
> >     and constraint_set5_flag is set to 1 if no B-slices are present.
> > ----
> >
> > According to my experiences, many amiko boxes affected to this problem,
> like:
> > amiko mini hd
> > amiko 8140c se
> > amiko 8142 twin c
> > amiko impulse
> > Amiko Mini HD265
> >
> > Please help me.
>
> It's an issue with your decoder(s) as they fail if one of the
> previously reserved bits is now 1 instead of 0 which they really
> should be ignoring, so you should probably send a bug report to the
> manufacturer.
>
> It seems there are a few different devices with the same issue though,
> so maybe we should just revert this change since I don't have much
> faith in decoder manufactures fixing bugs, especially for older
> products.
> _______________________________________________
> x264-devel mailing list
> x264-devel at videolan.org
> https://mailman.videolan.org/listinfo/x264-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x264-devel/attachments/20190624/77b34e72/attachment.html>