HandBrake for Windows support
An Activity Log
for support requests. Please read How-to get an activity log?
for details on how and why this should be provided.
- New User
- Posts: 1
- Joined: Tue Dec 04, 2018 5:09 am
Description of problem or question:
Poor CPU Utilization
Steps to reproduce the problem (If Applicable):
Run Handbrake on a server with more than 3 instances at a time with >4 CPUs per server
HandBrake version (e.g., 1.0.0):
Operating system and version (e.g., Ubuntu 16.04 LTS, macOS 10.13 High Sierra, Windows 10 Creators Update):
Windows Server 64Bit Datacenter
HandBrake Activity Log ***required***
(see How-to get an activity log
Hello, I was wonder if I was doing something wrong to fully utilize my 4x Dell R930 4xE7-4830 v3, 512 GB DDR4. It seems as though after three instances there are MASSIVE diminishing returns.
- Veteran User
- Posts: 6329
- Joined: Sat Apr 21, 2007 8:54 pm
Try disabling the decomb detect/decomb deinterlace, since your source doesn't appear to be interlaced or telecined based on framerate.
Also upgrade to the latest release, or nightly, since 1.1.1 is not supported.
Given your lack of specificity about what the utilization/returns are, it's hard to say any more.
- HandBrake Team
- Posts: 9361
- Joined: Sun Dec 24, 2006 1:05 pm
The biggest issue you'll have is decode. Low clock speed xeon's are not ideal for this typically but since your using x265 which is a very CPU intensive, you should get away with it.
I wouldn't have thought memory throughput would be an issue at 1080p, but it's difficult to say. We haven't yet enabled numa on x265 which should allow it to better handle multi-socket machines.
What sort of drop-offs do you see with the 4th instance?