On Sun, Feb 01, 2015 at 07:44:49PM -0300, James Almer wrote:
> On 01/02/15 7:27 PM, Michael Niedermayer wrote:
> > On Sun, Feb 01, 2015 at 08:54:44PM +0100, Mickaël Raulet wrote:
> >> something like this.
> >>
> >> https://github.com/OpenHEVC/openHEVC/commit/5d9f79edef2c11b915bdac3a025b59a32082f409
On 01/02/15 7:27 PM, Michael Niedermayer wrote:
> On Sun, Feb 01, 2015 at 08:54:44PM +0100, Mickaël Raulet wrote:
>> something like this.
>>
>> https://github.com/OpenHEVC/openHEVC/commit/5d9f79edef2c11b915bdac3a025b59a32082f409
>
> should we try to apply this one ?
>
> also we should reduce the
Hi,
2015-02-01 23:27 GMT+01:00 Michael Niedermayer :
> On Sun, Feb 01, 2015 at 08:54:44PM +0100, Mickaël Raulet wrote:
>> something like this.
>>
>> https://github.com/OpenHEVC/openHEVC/commit/5d9f79edef2c11b915bdac3a025b59a32082f409
>
> should we try to apply this one ?
I'm in the process of doi
On Sun, Feb 01, 2015 at 08:54:44PM +0100, Mickaël Raulet wrote:
> something like this.
>
> https://github.com/OpenHEVC/openHEVC/commit/5d9f79edef2c11b915bdac3a025b59a32082f409
should we try to apply this one ?
also we should reduce the differences to openhevc, are there any
other changes which w
Hi,
2015-02-01 20:54 GMT+01:00 Mickaël Raulet :
> something like this.
>
> https://github.com/OpenHEVC/openHEVC/commit/5d9f79edef2c11b915bdac3a025b59a32082f409
If I understand correctly, SAO EO needs data from before SAO on top
and left. I had hoped something simpler than this. Nice that the abov
something like this.
https://github.com/OpenHEVC/openHEVC/commit/5d9f79edef2c11b915bdac3a025b59a32082f409
Mickaël
2015-02-01 20:37 GMT+01:00 Christophe Gisquet
:
> Hi,
>
> 2015-02-01 20:00 GMT+01:00 James Almer :
> > From: Christophe Gisquet
> >
> > For band filter, source and destination are
Hi,
2015-02-01 20:00 GMT+01:00 James Almer :
> From: Christophe Gisquet
>
> For band filter, source and destination are aligned (except for 16x16 ctbs),
> and otherwise, they are most often aligned. Overall, the total width is also
> too small for amortizing memcpy.
There's probably much better
From: Christophe Gisquet
For band filter, source and destination are aligned (except for 16x16 ctbs),
and otherwise, they are most often aligned. Overall, the total width is also
too small for amortizing memcpy.
Timings (using an intrinsic version of edge filters):
B/32 B/64 E/
