Make QualityScaler more responsive to downgrades.

Make QualityScaler more responsive to downgrades.

Permits going from HD to QVGA in 6 seconds instead of 10. Also adds
windows for going up quickly in the beginning of a call (before any
downscaling happens due to bad quality).

BUG=webrtc:5678
R=glaznev@webrtc.org, stefan@webrtc.org

Committed: https://crrev.com/85829fd90cc4e7a91c9857921b19e8fc126aeb60
Cr-Commit-Position: refs/heads/master@{#12219}

[pbos-webrtc, 2016-03-24 13:26:00]

PTAL, I think this one should definitely be submitted "before the other one".

[pbos-webrtc, 2016-03-24 13:26:19]

On 2016/03/24 13:26:00, pbos-webrtc wrote:
> PTAL, I think this one should definitely be submitted "before the other one".

This also adds a fast-upscale initially, so time to HD is reduced.

[AlexG, 2016-03-24 18:13:49]

lgtm

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
File webrtc/modules/video_coding/utility/quality_scaler.cc (right):

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
webrtc/modules/video_coding/utility/quality_scaler.cc:183:
ReportFramerate(framerate_);
May worth to move num_samples_upscale_, num_samples_downscale_ update to a
separate function to make it a little cleaner - why scale adjusting should also
report framerate change?

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
File webrtc/modules/video_coding/utility/quality_scaler.h (right):

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
webrtc/modules/video_coding/utility/quality_scaler.h:56: int
measure_upscale_seconds_;
nit: measure_seconds_upscale_ to be in sync with other naming?

[pbos-webrtc, 2016-03-29 10:26:08]

feedback

[pbos-webrtc, 2016-03-29 10:37:22]

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
File webrtc/modules/video_coding/utility/quality_scaler.cc (right):

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
webrtc/modules/video_coding/utility/quality_scaler.cc:183:
ReportFramerate(framerate_);
On 2016/03/24 18:13:49, AlexG wrote:
> May worth to move num_samples_upscale_, num_samples_downscale_ update to a
> separate function to make it a little cleaner - why scale adjusting should
also
> report framerate change?

Definitely, I was being lazy. Done, thanks.

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
File webrtc/modules/video_coding/utility/quality_scaler.h (right):

https://codereview.webrtc.org/1830593003/diff/1/webrtc/modules/video_coding/u...
webrtc/modules/video_coding/utility/quality_scaler.h:56: int
measure_upscale_seconds_;
On 2016/03/24 18:13:49, AlexG wrote:
> nit: measure_seconds_upscale_ to be in sync with other naming?

Done.

[stefan-webrtc, 2016-04-01 13:35:11]

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/api/java/jni/andro...
File webrtc/api/java/jni/androidmediaencoder_jni.cc (right):

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/api/java/jni/andro...
webrtc/api/java/jni/androidmediaencoder_jni.cc:387: codec_settings->height,
codec_settings->maxFramerate);
Does this even build? Too many arguments AFAICT.

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/modules/video_codi...
File webrtc/modules/video_coding/utility/quality_scaler.cc (right):

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/modules/video_codi...
webrtc/modules/video_coding/utility/quality_scaler.cc:188:
measure_seconds_upscale_ = kMeasureSecondsUpscale;
Should there be a time limit at which time we also switch to the longer window?
E.g., after 20 seconds?

[pbos-webrtc, 2016-04-04 12:20:42]

remove extra height argument

[pbos-webrtc, 2016-04-04 12:32:39]

PTAL

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/api/java/jni/andro...
File webrtc/api/java/jni/androidmediaencoder_jni.cc (right):

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/api/java/jni/andro...
webrtc/api/java/jni/androidmediaencoder_jni.cc:387: codec_settings->height,
codec_settings->maxFramerate);
On 2016/04/01 13:35:10, stefan-webrtc (holmer) wrote:
> Does this even build? Too many arguments AFAICT.

Bad rebase, done.

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/modules/video_codi...
File webrtc/modules/video_coding/utility/quality_scaler.cc (right):

https://codereview.webrtc.org/1830593003/diff/20001/webrtc/modules/video_codi...
webrtc/modules/video_coding/utility/quality_scaler.cc:188:
measure_seconds_upscale_ = kMeasureSecondsUpscale;
On 2016/04/01 13:35:10, stefan-webrtc (holmer) wrote:
> Should there be a time limit at which time we also switch to the longer
window?
> E.g., after 20 seconds?

Don't think it's really worth the code complexity for a corner case where we'd
go up wrong once (and it could still be right)? WDYT?

[stefan-webrtc, 2016-04-04 14:31:04]

lgtm

[pbos-webrtc, 2016-04-04 16:11:22]

Description was changed from

==========
Make QualityScaler more responsive to downgrades.

Permits going from HD to QVGA in 6 seconds instead of 10. Also adds
windows for going up quickly in the beginning of a call (before any
downscaling happens due to bad quality).

BUG=webrtc:5678
R=glaznev@webrtc.org, stefan@webrtc.org
==========

to

==========
Make QualityScaler more responsive to downgrades.

Permits going from HD to QVGA in 6 seconds instead of 10. Also adds
windows for going up quickly in the beginning of a call (before any
downscaling happens due to bad quality).

BUG=webrtc:5678
R=glaznev@webrtc.org, stefan@webrtc.org

Committed:
https://chromium.googlesource.com/external/webrtc/+/85829fd90cc4e7a91c9857921...
==========

[pbos-webrtc, 2016-04-04 16:11:24]

Committed patchset #3 (id:40001) manually as
85829fd90cc4e7a91c9857921b19e8fc126aeb60 (presubmit successful).

[commit-bot: I haz the power, 2016-04-04 16:11:24]

Description was changed from

==========
Make QualityScaler more responsive to downgrades.

Permits going from HD to QVGA in 6 seconds instead of 10. Also adds
windows for going up quickly in the beginning of a call (before any
downscaling happens due to bad quality).

BUG=webrtc:5678
R=glaznev@webrtc.org, stefan@webrtc.org

Committed:
https://chromium.googlesource.com/external/webrtc/+/85829fd90cc4e7a91c9857921...
==========

to

==========
Make QualityScaler more responsive to downgrades.

Permits going from HD to QVGA in 6 seconds instead of 10. Also adds
windows for going up quickly in the beginning of a call (before any
downscaling happens due to bad quality).

BUG=webrtc:5678
R=glaznev@webrtc.org, stefan@webrtc.org

Committed: https://crrev.com/85829fd90cc4e7a91c9857921b19e8fc126aeb60
Cr-Commit-Position: refs/heads/master@{#12219}
==========

[commit-bot: I haz the power, 2016-04-04 16:11:26]

Patchset 3 (id:??) landed as
https://crrev.com/85829fd90cc4e7a91c9857921b19e8fc126aeb60
Cr-Commit-Position: refs/heads/master@{#12219}

[phoglund, 2016-04-12 13:25:35]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.webrtc.org/1880103002/ by phoglund@webrtc.org.

The reason for reverting is: Speculative revert: want to see if this causes the
regression in https://crbug.com/602621.