Allow FMP trace event to appear slightly before the FCP #1404
Conversation
|
So there's good news and bad news. 👍 The good news is that everyone that needs to sign a CLA (the pull request submitter and all commit authors) have done so. Everything is all good there. 😕 The bad news is that it appears that one or more commits were authored by someone other than the pull request submitter. We need to confirm that they're okay with their commits being contributed to this project. Please have them confirm that here in the pull request. Note to project maintainer: This is a terminal state, meaning the |
|
@googlebot cool thanks |
|
you have some camel case eslint errors |
| e.name === 'firstMeaningfulPaint' && e.ts >= (firstFCP.ts - fCPfMP_TOLERANCE)); | ||
|
|
||
| // Sometimes fMP is triggered before fCP | ||
| // (this happends when above the fold is being swapped with JS) |
There was a problem hiding this comment.
happens
(also not sure what this means. Fold?)
| const firstMeaningfulPaint = frameEvents.find(e => | ||
| e.name === 'firstMeaningfulPaint' && e.ts >= firstFCP.ts); | ||
| e.name === 'firstMeaningfulPaint' && e.ts >= (firstFCP.ts - fCPfMP_TOLERANCE)); |
There was a problem hiding this comment.
if there's an FMP event a few ms before FCP (within tolerance) and one after FCP, is the first one always better here?
There was a problem hiding this comment.
this shouldn't happen as we filter by frame. but if so, yeah sure. :)
| @@ -0,0 +1,77 @@ | |||
| { | |||
There was a problem hiding this comment.
hooray for minimal trace fixtures :)
paulirish
force-pushed
the
fmptolerance
branch
from
63b0b8e
to
073b820
Compare
| } | ||
|
|
||
| // Sometimes fMP is triggered before fCP | ||
| // (this happends when above the fold is being swapped with JS) |
There was a problem hiding this comment.
rebase or something got mixed up here
paulirish
force-pushed
the
fmptolerance
branch
from
073b820
to
3dac393
Compare
|
Woah, FCP should never happen after FMP. |
|
@tdresser there's an example trace with FMP before FCP in www.nfl.com_2017-01-04_09-44-25-0.zip from #1253 (comment) |
|
@tdresser here is another one |
This is the core problem of #1253.
As @wardpeet deduced, we are hitting these because sometimes FMP trace events are added before FCP
This now allows there to be 32ms of tolerance between the two, as we found traces where FMP is just 0.7ms before FCP.
Fixes #1253.