Does hiding sensors help battery/CPU utilization?

[snovvman]

My observation is inconclusive, so I figure I ask here--

I am running HWINFO on a laptop and would like to minimize, if possible, its CPU utilization and therefore maximizing battery life. HWINFO is paired with Rainformer and they provide a great set of data for me. At the same time, there are many sensors that I do not need. Were I to hide those sensors under Layout, will that help reduce HWINFO's CPU utilization?

Thanks.

 

[Martin]

Reading sensor data should have no significant impact on system utilization or power consumption, but there can be some exceptions. Such can be disk SMART sensors in case the underlying drivers perform excessive actions to read the data. If you enable the "Profiling Time" column in HWiNFO, it will tell you how much time it takes to read each sensor/value. Even though this does not indicate the system load associated with reading a given sensor (as the time spent might include idle time), it can show you reading which sensors might cost more resources.
Anyway, if you don't need certain sensors and are concerned about utilization just hide them.

 

[snovvman]

Such can be disk SMART sensors in case the underlying drivers perform excessive actions to read the data. If you enable the "Profiling Time" column in HWiNFO, it will tell you how much time it takes to read each sensor/value. Even though this does not indicate the system load associated with reading a given sensor (as the time spent might include idle time), it can show you reading which sensors might cost more resources.

Thank you, Martin. I enabled Profiling Time. I see that SMART takes 2-10ms while CPU is essentially zero. I have set the polling frequency to 3 cycles to try it out. The real interesting story is that the GPU takes somewhere between 8-45ms. The individual data points under the GPU, however, all show zero. Is there a way to find out which specific sensor(s) is/are using resources?

 

[Martin]

I'm afraid that's not possible. This is because some of the calls made a done globally for the whole GPU and not for a particular value.