[x265] TComTrQuant: lambda for each Cb and Cr

[Steve Borho]

On Tue, Jun 10, 2014 at 11:26 AM, Steve Borho <steve at borho.org> wrote:
> On Sat, Jun 7, 2014 at 9:01 PM, Satoshi Nakagawa <nakagawa424 at oki.com> wrote:
>> # HG changeset patch
>> # User Satoshi Nakagawa <nakagawa424 at oki.com>
>> # Date 1402192642 -32400
>> #      Sun Jun 08 10:57:22 2014 +0900
>> # Node ID 188e115f07427c759f47154a864467be21b5b6a1
>> # Parent  e5656f1e190453efa84732269b259a6dee608ff9
>> TComTrQuant: lambda for each Cb and Cr
>
> Queued for testing.  This commit message only describes about a third
> of what this patch does, but it does look like it shouldn't change
> outputs if the chroma weights are not changed from the default.

Our testing showed this commit (or perhaps the tskip changes, since
they were tested together) changed the encoder outputs.  Was it
intentional to change outputs with default chroma weights?

-- 
Steve Borho