Score:0

TLP Wifi works BETTER on battery than on AC

cn flag

I installed TLP a couple of months ago. (I was having a problem with my Bluetooth disconnecting after 3 seconds of inactivity, installing TLP and going to /etc/tlp.conf and changing USB_BLACKLIST_BTUSB from 0 to 1 fixed it)

I had not used my laptop in the last 10/15 days (during the holidays) and updated everything as soon as I got back. I am using Ubuntu 20.04.

I have the impression that updating, my wifi started working terribly, disconnecting all the time. Surprisingly, the wifi seemed fine if I was on the battery, and I discovered that it worked normal simply if a ran sudo tlp bat instead of sudo tlp start independently on being or not connected to AC.

I was able to fix the issue last night by going to /etc/tlp.conf and changing RUNTIME_PM_ON_AC to auto.

This morning I updated the BIOS to see if it fixed it, and so far it seems that setting RUNTIME_PM_ON_AC back to the default value no longer influences the wifi performance, and my wifi connection is stable again in both tlp modes.

However, by making several tests on the speed of the connection, I discovered that my internet speed is about 20% faster if I run sudo tlp bat compared to sudo tlp ac.

All of these issues seem to be related, but I cannot understand what is happening. I have been using Ubuntu for 10 years, and never got such a weird behavior.

Score:0
id flag

This seems to be a problem with the Wi-Fi driver. Why wouldn't you want to stay with RUNTIME_PM_ON_AC=auto if the performance is better? You may also consider using RUNTIME_PM_ENABLE to set only the Wi-Fi card to auto https://linrunner.de/tlp/settings/runtimepm.html#runtime-pm-enable

Raphael Chinchilla avatar
cn flag
Changing `RUNTIME_PM_ON_AC=auto` no longer influences the quality of the wifi. It no longer keeps disconnecting and connecting, but the internet speed is smaller on ac than on battery.
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.