Closed Snuffleupagus closed 2 weeks ago
Command cmd_test from @Snuffleupagus received. Current queue size: 0
Live output at: http://54.241.84.105:8877/6e0ea539a1a9e89/output.txt
Command cmd_test from @Snuffleupagus received. Current queue size: 0
Live output at: http://54.193.163.58:8877/c7c5f7714d511d0/output.txt
Full output at http://54.241.84.105:8877/6e0ea539a1a9e89/output.txt
Total script time: 28.63 mins
different ref/snapshot: 13
different first/second rendering: 1
Image differences available at: http://54.241.84.105:8877/6e0ea539a1a9e89/reftest-analyzer.html#web=eq.log
Full output at http://54.193.163.58:8877/c7c5f7714d511d0/output.txt
Total script time: 44.11 mins
different ref/snapshot: 3
Image differences available at: http://54.193.163.58:8877/c7c5f7714d511d0/reftest-analyzer.html#web=eq.log
Missed one case...
/botio test
Command cmd_test from @Snuffleupagus received. Current queue size: 0
Live output at: http://54.241.84.105:8877/4f8a8bd7a5fea7f/output.txt
Command cmd_test from @Snuffleupagus received. Current queue size: 0
Live output at: http://54.193.163.58:8877/205590cf8f15436/output.txt
Full output at http://54.241.84.105:8877/4f8a8bd7a5fea7f/output.txt
Total script time: 28.68 mins
different ref/snapshot: 17
different first/second rendering: 2
Image differences available at: http://54.241.84.105:8877/4f8a8bd7a5fea7f/reftest-analyzer.html#web=eq.log
Full output at http://54.193.163.58:8877/205590cf8f15436/output.txt
Total script time: 42.02 mins
different ref/snapshot: 9
Image differences available at: http://54.193.163.58:8877/205590cf8f15436/reftest-analyzer.html#web=eq.log
Locally I get less "severe" movement when testing on Windows, perhaps this is worse with devicePixelRatio === 1
since I believe the bots use that?
Perhaps the QueueOptimizer
logic is also affecting things, now that we're correctly handling cached imageMask-data as well?
In any case, given that this patch ought to be correct are we OK with the ref-test movement in the latest round of testing?
I have checked locally that this indeed solves the issue. I have also checked the last round of reference tests and didn't find anything that was either noticeable with the naked eye or that was unexpected movement, and given that I agree that this patch should be correct I think we can do this. Thanks!
/botio makeref
Command cmd_makeref from @timvandermeij received. Current queue size: 0
Live output at: http://54.193.163.58:8877/018642573633292/output.txt
Command cmd_makeref from @timvandermeij received. Current queue size: 0
Live output at: http://54.241.84.105:8877/2645d88b795f3f7/output.txt
Full output at http://54.241.84.105:8877/2645d88b795f3f7/output.txt
Total script time: 19.88 mins
Full output at http://54.193.163.58:8877/018642573633292/output.txt
Total script time: 25.13 mins
/botio test