New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 796831 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner: ----
Closed: Jan 2018
Cc:
Components:
EstimatedDays: ----
NextAction: 2018-01-30
OS: Windows
Pri: 3
Type: Bug



Sign in to add a comment

Pointer Lock movementX/Y unusable

Reported by lama12...@gmail.com, Dec 21 2017

Issue description

UserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.84 Safari/537.36

Steps to reproduce the problem:
http://deepdreamers.org/canvas.html

1. Draw circles on different systems, some will bug
2. 
3. 

I don't have this problem myself, but two people I know got this bug.

What is the expected behavior?
Nice and smooth circles

What went wrong?
Lots of relative coordinates got artifacts from real cursor position, which is probably hidden and centered in middle of screen

Did this work before? N/A 

Does this work in other browsers? N/A

Chrome version: 63.0.3239.84  Channel: n/a
OS Version: 10.0
Flash Version: 

I looked at the other Pointer Lock issues, but this seems to be unique.
 
canvas.html
1.1 KB View Download
282f8de902e236bcce54e39d46eb5c80.png
84.6 KB View Download
Labels: Needs-Triage-M63

Comment 2 by lama12...@gmail.com, Dec 21 2017

What is Triage M63? Can't find any information about that. My mate has a Logitech G500. He tested different DPI:

Works: 125hz, 200hz, 250hz, 333hz
Bug: 500hz, 1000hz
Unable to reproduce the issue on reported chrome version 63.0.3239.84 and on the latest canary 65.0.3305.0 using Windows 10 with the below mentioned steps.
1) Launched Chrome reported version
2) Navigated to URL: http://deepdreamers.org/canvas.html
3) Drawn the circles in the path provided
4) Able to draw smooth and nice circles

@Reporter: Please have a look at the attached screen cast and let us know whether we have missed any steps in reproduicng the issue, try to test this issue by creating new person which don't have any apps or extensions in it and let us know if the issue still persists.

Thanks!
796831.mp4
1.2 MB View Download
Cc: viswatej...@techmahindra.com sc00335...@techmahindra.com
Labels: Needs-Feedback Triaged-ET

Comment 5 by lama12...@gmail.com, Dec 28 2017

Hi, did you test with a high DPI mouse with the polling rate set between 500hz and 1000hz?

And thanks for testing aswell. :)
Project Member

Comment 6 by sheriffbot@chromium.org, Dec 28 2017

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "sc00335628@techmahindra.com" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
As per comment#5 from the reporter, we need to test this issue with a high DPI mouse by setting polling rate between 500hz and 1000hz, hence adding the label as TE-NeedsTriageFromMTV, can anyone from the MTV team please have a look at this issue.

Thanks!
Labels: TE-NeedsTriageFromMTV
Cc: dtapu...@chromium.org
Can you try this URI https://rbyers.github.io/paint.html?

When you load this page press 'c' on the keyboard. You'll see two sets of points; 1 the pointer events and 2 the events with pointer histories; ie (calling https://developer.mozilla.org/en-US/docs/Web/API/PointerEvent/getCoalescedEvents API).



Hi, he tested it with his mouse now, but the result is the same as on my PC.

Chrome, without "c": https://i.gyazo.com/7101bdf4c8c257abbb5ae18ada7dee1b.png

Chrome, with "c": https://i.gyazo.com/d9e0c3988f5345e2675425d2b0500c75.png

Firefox, "c" has no effect: https://i.gyazo.com/4e0ee889d56cbcf3cf964006a7168d84.png

It is kinda what I expected tho, because the bug only occurs in PointerLock.
Cc: eirage@chromium.org nzolghadr@chromium.org
So the problem you are seeing is with movementX/Y on some systems. Could this be because those systems are using HiDPI displays?

In your example you are assuming that CSS pixels == movementX/Y and that isn't necessarily the case you need to currently divide it by window.devicePixelRatio.

We have a few bugs open as to what coordinate system movementX/Y should be relative to.

Does loading this page: http://output.jsbin.com/natigaz work on the system?
Labels: Needs-Feedback
The re-center may be same as  crbug.com/781182 . Could you try if it's reproduce on canary or M64?

And there is another issue only happen on high rolling rate mouse, it might be similar: crbug.com/780404 
NextAction: 2018-01-18
lama...@ we are still waiting for feedback from you.
Hi, sorry for the late response. Latest Canary does not fix it. However, we figured something out...

His multi-monitor setup has:

1) 144hz screen
2) and a 60hz screen
3) and a >1000hz mouse...

canvas.html bugs on the 144hz screen, but it works on the 60hz screen.

We tested mouse hertz on this site: https://zowie.benq.com/en/support/mouse-rate-checker.html

He got >1030hz there for his mouse.

So I don't know if the mouse is the relevant thing here... maybe it's the different screen hertz, or only the mouse... or both together.

One strange thing aswell... we tested this WebGL for FPS: http://seemore.playcanvas.com/

He got 63 FPS on the 60hz screen and 145/146 FPS on the 144hz screen. My hardware setup only gives me max 60 FPS on my 60hz screen.

Lots of factors here, IMHO you should try to setup this kind hardware for yourself and see if it bugs for you, so you can debug it without problems.
Project Member

Comment 17 by sheriffbot@chromium.org, Jan 14 2018

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "dtapuska@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
Ok, I got feedback from the second guy, he has a single-monitor setup, 60 FPS only.

He tested with Canary 65.0.3315.3 and the canvas.html demo works for him now. But he says it feels "spongy" and inprecise (spongy like before, but without the re-center bug).
Hi lama, could you provide the scale setting for the monitors? What's the result if you type 'window.devicePixelRatio' in chrome devtool console?
And I didn't get what you mean by "spongy", could you describe it more or attach a screen shot. It'll be really helpful for figuring out the issue, thanks!
The NextAction date has arrived: 2018-01-18
Labels: -Pri-2 Needs-Feedback Pri-3
NextAction: 2018-01-30
lama...@ we are still waiting feedback from you.
Status: WontFix (was: Unconfirmed)
lama...@ since there was no feedback I'm going to close this for now. But feel free to comment if you have additional feedback. If you happen to come back could you also try
http://output.jsbin.com/natigaz see if there is any difference. Also if you see some sort of a polygon shape instead of a circle can you try pointerevents coalesced events and draw those. That will solve that issue as well. 
The NextAction date has arrived: 2018-01-30

Sign in to add a comment