I thought about bug 1901641 yesterday, I have an idea to solve bug 1901641 and this bug altogether. Now we are going to use the last sampled animation value on the __compositor__ as the start value of a new transition in this bug. And the problem caused bug 1901641 is that the start time of the new transition is stable if there's busy state on the main-thread. That's right? Then, in this bug we can also use the last Vsync timestamp on the __compositor__ as the start time of the new transition. I suppose it should solve both bugs altogether. Maybe the start time should be the current timestamp? maybe?
Bug 1626165 Comment 32 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
I thought about bug 1901641 yesterday, I have an idea to solve bug 1901641 and this bug altogether. Now we are going to use the last sampled animation value on the __compositor__ as the start value of a new transition in this bug. And the problem caused bug 1901641 is that the start time of the new transition is stale if there's busy state on the main-thread. That's right? Then, in this bug we can also use the last Vsync timestamp on the __compositor__ as the start time of the new transition. I suppose it should solve both bugs altogether. Maybe the start time should be the current timestamp? maybe?