INTERACT FORUM
Windows => Television => Topic started by: eapool on March 15, 2014, 07:21:51 am
-
Lately, I have noticed an increase in CPU usage while recording. Today I finally had some time to run some tests. What I am seeing is that between 19.0.108 and 19.0.114, the CPU usage jumped while recording an OTA show using my HDhomerun tuners. In version 108, I was seeing about 8% usage by media center, in version 114, it jumps to about 25%. This becomes a problem when I am trying to record more than a couple programs at one time. This morning, I rolled back to 104 and started moving forward until I saw the increase in CPU usage. I was not watching anything, just recording one show (the Today show on NBC).
I am recording in .ts, with windows 8.1.
My question: Was this a planned event, or did something break between these versions?
Thanks,
Alex
-
Are you really referring to MC18, or was that a typo and you are referring to MC19 respectively?
-
Yea, sorry, it is MC19, not enough coffee yet. - I fixed my original post.
-
There was this change in build 110:
Changed: When recording a television show in TS format, MC will use buffering to reduce the number of IO hits. This should fix the video quality issue in television shows recorded in TS format.
I can not imagine this causes CPU usage to significantly increase. Is it possible?
-
Have you had a chance to attempt to reproduce this? It is very consistent on my machine. If I try to record more than two things at a time, the CPU usage hits 100%. At this point, the recording become pixelated and unwatchable. I am hoping to put a second STB into the mix this weekend, but it may be too much for my machine to handle.
Thanks,
Alex
-
I can confirm there is a problem since build 110. CPU usage was significantly increased for recording in TS format. For a single recording on my computer, it uses 25-27% of CPU capacity. With build 109 and before, the usage is 12-14%.
-
This should be fixed in the latest build, 19.0.124.
http://yabb.jriver.com/interact/index.php?topic=88373.0
-
Brief testing this morning indicates this is fixed.
Thank you.
Alex