This is probably blur related. We should retry after https://github.com/servo/webrender/pull/1961 lands.
status-firefox57: --- → unaffected
status-firefox58: --- → unaffected
Priority: -- → P2
I believe #1961 landed and this page is still pretty bad (and renders incorrectly).
This is indeed blur related. The two <a> elements for "IPFS" and for "CRDTs" each have an ::after pseudo element which has box-shadow: -350px -20px 1000px 22px #23e0f7 set on it. That's a 1000px radius blur.
I take a look.
Assignee: nobody → ethlin
Looks like the performance bottleneck is not box-shadow anymore. I sent a PR 2096 to fix the rendering problem. I will keep investigating the next performance bottleneck in this page.  https://github.com/servo/webrender/pull/2096
I should say the blur is not the bottleneck. According to my local test, we spend 25ms on GPU, and clip takes more than 10ms. I think it's because the clip shape is ellipse and webrender doesn't handle ellipse clip efficiently. I'll have a PR to fix it.
PR 2158 also helps. Right now the GPU time is about 11ms on my MBP.  https://github.com/servo/webrender/pull/2158
The FPS of scrolling on my device is about 50~60fps now. Jeff, do you think we can close this bug?
I only get around 30fps on that page and see 30+ ms composites: https://perfht.ml/2nE69al
Priority: P1 → P3
This is better but still not great.
Jeff -- What issues remain? How bad are they? How difficult are they to hit?
I tried this out on some older nvidia desktop hardware and still got 30ms composites. So we could definitely do better.
Depends on: 1494775
Status: ASSIGNED → RESOLVED
Last Resolved: 23 days ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.