Issue metadata
Sign in to add a comment
|
various gray gradient artifacts color banding, jitter, etc.
Reported by
realgran...@gmail.com,
May 28 2017
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (X11; CrOS x86_64 9460.50.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3071.71 Safari/537.36 Platform: 9460.50.0 (Official Build) beta-channel cave Example URL: https://jsfiddle.net/wy5pz3wa/ Steps to reproduce the problem: 1. create a complex gray gradient or open a picture with gray gradient (not all images have this effect though) 2. scroll/resize window 3. What is the expected behavior? smooth and stable gradient What went wrong? the gradient is uneven, exhibits weird color banding and the colors bounce when scrolled/resized Does it occur on multiple sites: Yes Is it a problem with a plugin? No Did this work before? N/A Does this work in other browsers? Yes Chrome version: 59.0.3071.71 Channel: beta OS Version: 9460.50.0 Flash Version: Shockwave Flash 26.0 r0 this can be especially seen at https://auth0.com/learn header when scrolling the page with smooth scroll touchpad found on arm and intel chromebooks (veyron_jaq, falco, cave) stable and beta
,
May 31 2017
Looks fine on Mac Version 60.0.3112.0 (Official Build) canary (64-bit) and Mac Version 59.0.3071.61 (Official Build) beta (64-bit). Could the test team please try to reproduce on one of the listed CrOS devices?
,
Jun 14 2017
Tested this issue on Chrome OS using chrome version stable # 59.0.3071.91/9460.60.0 by following steps mentioned below. 1.Opened https://jsfiddle.net/wy5pz3wa/ and then resized the window 2.Observed smooth and stable gradient and Not observed uneven gradient, weird color banding and the colors bouncing when resized/Scrolled (Please refer Video) 3.Opened https://auth0.com/learn header and Scrolled the page with smooth scroll touchpad and did not observe gray gradient artifacts color banding, jitter, etc. Note: As we don't have Veyron_jaq, Cave devices tested this Issue on Falco Songsuk@ : Could you please help in testing this issue on Veyron_jaq, Cave devices if its available there. Thanks in advance!
,
Jun 19 2017
The NextAction date has arrived: 2017-06-19
,
Jun 19 2017
Reproduced this again on cave and others: Steps to reproduce: 1. open the https://auth0.com/learn web page 2. scroll SLOWLY using two-finger scroll to see jumping gradient colors 3. observe the image with EYES, not by recording internal chrome buffers see attached vids shot from the phone camera compared to firefox, auth0/learn: (showing Chrome, then switching to FF, then Chrome, FF, Chrome): https://goo.gl/photos/NQQZ9iF8cYrsMdQ98 chrome, jsfiddle: https://goo.gl/photos/MnuGPT2psHTLDHgJA same with FF: https://goo.gl/photos/5CN8WqVkWQoYdJuD6 videos are not very crisp since I do not have the right equipment with me but still you can see the color bouncing in chrome
,
Jun 19 2017
Thank you for providing more feedback. Adding requester "schenney@chromium.org" to the cc list and removing "Needs-Feedback" label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Jun 19 2017
This is only occurring on one device family, suggesting a graphics card issue, or maybe on that device things switch to software rendering and that has the artifacts. Over to the GPU for further triage, as this issue must be somewhere in the raster pipeline.
,
Nov 8 2017
This is intended. This is an artifact of the cheaper display panels used on some devices which display at 6bpp instead of 8bpp like higher end panels. Even though we enable dithering, which improves the situation a little, the banding is still visible.
,
Nov 8 2017
Thank you very much for explanation! I would greatly appreciate just in case you could add a short clarification - does ASUS Chromebook Flip C302 fall into this category also (because this is the one I am using and where I noticed this thing)? It is a $1200 computer and still comes with a cheaper 6bpp low-end display panel. That is a nice issue to mention in the reviews!
,
Nov 8 2017
I see the banding on the C302 as well. It's a higher end display, but still 6 bit (notice the line that says "6 bits per primary color channel"):
$ edid-decode < edid41.bin
Extracted contents:
header: 00 ff ff ff ff ff ff 00
serial number: 06 af 6d 22 00 00 00 00 00 19
version: 01 04
basic params: 95 1c 10 78 02
chroma info: 16 85 95 59 58 96 28 1d 50 54
established: 00 00 00
standard: 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01
descriptor 1: 14 37 80 b8 70 38 24 40 10 10 3e 00 14 9b 10 00 00 18
descriptor 2: 00 00 00 0f 00 00 00 00 00 00 00 00 00 00 00 00 00 20
descriptor 3: 00 00 00 fe 00 41 55 4f 0a 20 20 20 20 20 20 20 20 20
descriptor 4: 00 00 00 fe 00 42 31 32 35 48 41 4e 30 32 2e 32 20 0a
extensions: 00
checksum: 5e
Manufacturer: AUO Model 226d Serial Number 0
Made week 0 of 2015
EDID version: 1.4
Digital display
6 bits per primary color channel
DisplayPort interface
Maximum image size: 28 cm x 16 cm
Gamma: 2.20
Supported color formats: RGB 4:4:4
First detailed timing is preferred timing
Established timings supported:
Standard timings supported:
Detailed mode: Clock 141.000 MHz, 276 mm x 155 mm
1920 1936 1952 2104 hborder 0
1080 1083 1097 1116 vborder 0
-hsync -vsync
Manufacturer-specified data, tag 15
ASCII string: AUO
ASCII string: B125HAN02.2
Checksum: 0x5e
EDID block does NOT conform to EDID 1.3!
Missing name descriptor
Missing monitor ranges
Overall, the gray gradient used on that website is so narrow that banding is hard to avoid. I even see a little bit of banding on my HP LP3065 which is natively 8 bit...
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by cbiesin...@chromium.org
, May 30 2017Labels: Needs-Bisect