HandBrake

I have a SuperMicro 7037A-i with 2 XEON E5-2665s' & 128GB of RAM.

My question is about CPU utilization...
Other than launching the process with a high priority, is there any configuration that can be done to optimize HandBrake's CPU settings?

A single HandBrake instance will only load the system to ~20%, and a 2-pass BRD conversion will take ~15 hours to complete.

At this point I am simply running 3 instances of HandBrake, to best utilize the CPUs.

Is there a better solution?

THX!

JWS

Failed to mention...
Running on Win 7 Ultimate x64

My encode settings are custom, and geared for quality, not speed.

CPU utilisation will depend a lot on your source & settings, can you post a log so we can see what's going on?

I believe there are parts of HB that are not multi-threaded so there might not be much that can be done.

SC

JWS wrote:is there any configuration that can be done to optimize HandBrake's CPU settings?

Yes.

An activity log would be required for any specific suggestions.

In general:
- Input format with a threaded decoder
- No filters or threaded filters only
- High vertical resolution
- High total resolution
- Slow encoder settings

Hello,
same Question is my Issue.
Here are the log Files.

Activity Log
https://drive.google.com/file/d/0B_o_L2 ... sp=sharing

Film Logs
https://drive.google.com/file/d/0B_o_L2 ... sp=sharing
https://drive.google.com/file/d/0B_o_L2 ... sp=sharing
https://drive.google.com/file/d/0B_o_L2 ... sp=sharing
https://drive.google.com/file/d/0B_o_L2 ... sp=sharing
https://drive.google.com/file/d/0B_o_L2 ... sp=sharing

JWS wrote: ↑Tue Apr 26, 2016 4:30 pm launching the process with a high priority...
I am simply running 3 instances of HandBrake,

Interesting me thinks....

Ok...

And that is the solution?
Nobody who is able to read out the log files and can tell me the reason?
When I want to recode 10 files over night it´s better to run more Instances than put them all in a queue?

Really?

Do those sources really contain interlaced content? if not, I'd turn off decomb that'll help speed things up a small amount.

Regardless, your very unlikely to max out those CPU's with once instance. Running multiple instances is the best option. Have a separate queue of work for each.