| Summary: | Slow render on Khadas Vim2 Max and NanoPC T3+ Octa-core 64-bit ARM | ||
|---|---|---|---|
| Product: | [Applications] kdenlive | Reporter: | NicoD <dekerfnico> |
| Component: | Rendering & Export | Assignee: | Jean-Baptiste Mardelle <jb> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | normal | CC: | fritzibaby |
| Priority: | NOR | ||
| Version First Reported In: | 18.04.1 | ||
| Target Milestone: | --- | ||
| Platform: | Ubuntu | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
|
Description
NicoD
2018-05-29 11:13:44 UTC
Some extra information. I tried to compile version 18.04 with the instructions here: https://community.kde.org/Kdenlive/Development I chose refactoring_timeline I'll try again with the other. For some strange reason I ended up with version 17.11.70. There it performs even worse. Result with the NanoPC T3+ was 2h49m17s. What I forgot to mention was that the places where it performs very badly are places with a dissolve and 2 x difference transitions. The places where there is no transition or effect it performs good. I still hope you can help me solve this problem. I would expect these boards to do this in under an hour. That's why I bought them, I spend a lot of money on both. I don't want this to be in vain. Thank you for your attention. I'll keep posting my progress here too. Greetings. NicoD I was able to install version 18.04.1 It is completely the same as with version 17.12.3 So between version 17.11.70 and 17.12.3 this has improved quite a bit. But no difference in 17.12.3 and 18.04.1 The time was 1h24m10s Again the last part, a difference transition of 1 minute long between 2 1080p video's. This takes more than 35minutes. Only 1 core is maxed, the others do almost nothing. The rest of the project all the cores are at average at 70%. Sometimes it goes up, sometimes a bit down. But compared to the last bit it goes fast. Greetings. Thank you for reporting. We marked this as a dublicate of 401186 *** This bug has been marked as a duplicate of bug 401186 *** |