Issue metadata
Sign in to add a comment
|
Opacity property visually 'thins' red/blue color text
Reported by
alyst...@gmail.com,
Jul 13
|
||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.99 Safari/537.36 Example URL: https://codepen.io/alystair/pen/qyOoVX Steps to reproduce the problem: Color a piece of text red or blue, then add any level of opacity, more noticeable on dark backgrounds What is the expected behavior? Text should render the same as any other color/opacity. What went wrong? Affected text becomes visually 'thinner'. Does it occur on multiple sites: Yes Is it a problem with a plugin? No Did this work before? No Does this work in other browsers? Yes Chrome version: 67.0.3396.99 Channel: stable OS Version: 6.1 (Windows 7, Windows Server 2008 R2) Flash Version: Test URL contains screenshots
,
Jul 14
Can anyone in font rendering take a look?
,
Jul 15
,
Jul 16
Able to reproduce the issue on Win-10 using chrome reported version #67.0.3396.99 and latest canary #69.0.3492.0. Issue is not seen in OS-mac and OS-Linux. This is a non-regression issue as it is observed from M60 old builds. Hence, marking it as untriaged to get more inputs from dev team. Thanks...!!
,
Jul 16
Able to reproduce the issue on Win-10 using chrome reported version #67.0.3396.99 and latest canary #69.0.3492.0. Issue is not seen in OS-mac and OS-Linux. This is a non-regression issue as it is observed from M60 old builds. Hence, marking it as untriaged to get more inputs from dev team. Thanks...!!
,
Jul 16
,
Jul 18
Odd that it only happens with opacity, if it was an LCD smoothing issue I'd expect it to apply to all text and be dependent on the geometry of the screen. Any ideas paint team?
,
Jul 19
bungeman@, any insight on (a) why this might be Windows only, and (b) why fints with opacity might render differently. It seems this is most likely a Skia rendering issue given we just set paint properties in Blink.
,
Jul 30
The NextAction date has arrived: 2018-07-30
,
Jul 30
Ping, bungeman@ could you reply to comment #8. Or drott@ could you reply?
,
Aug 13
The NextAction date has arrived: 2018-08-13
,
Aug 13
Still looking for input from bungeman@ or drott@
,
Aug 14
I started looking into this, but I haven't been able to get an skp of the problematic bits yet. None of the methods at https://www.chromium.org/developers/how-tos/trace-event-profiling-tool/saving-skp-s-from-chromium seem to work for this. With option 1 I get an skp, but only layer0; it appears the opacity sends this into a layer but currently the layers aren't outputing that way. Using option 2 just crashes recording. What I really need to know here is if blink is still requesting lcd or not (it should be) and an actual example so I can tell if the issue is something to do with the text itself or the way the layers are being combined.
,
Aug 27
The NextAction date has arrived: 2018-08-27
,
Aug 27
fmalita@, could you look into how the paint is setup for the text? |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by alyst...@gmail.com
, Jul 13