Tiger Lake + Renoir On Ubuntu Linux For Battery vs. AC Performance
Given the recent Intel presentation alleging AMD Ryzen laptop performance being worse on battery relative to the AC vs. battery performance seen with Intel EVO notebooks featuring Tiger Lake processors, I ran a mini comparison on my side to see whether there is any merit to Intel's information when testing under Ubuntu Linux.
Here is just some initial data on my side when benchmarking AMD Ryzen "Renoir" versus Intel Core i7 "Tiger Lake" when running Ubuntu 20.10 and comparing the AC power versus battery performance.
It's just a small comparison though for lack of Renoir / Tiger Lake systems. With most major laptop vendors not pursuing the admittedly small Linux marketshare, it's not easy obtaining review samples for a site devoted to Linux. That paired with the abundance of ad-block users in the Linux/open-source community is detrimental to the site's resources and the ability to buy many laptops for testing. Thus for this testing based on my current hardware is just one Renoir and one Tiger Lake laptop. (If you wish to show your support and allow for more laptop testing, consider joining Phoronix Premium and there is the ongoing Black Friday deal.) The two devices also aren't in the same class with one being a cheap Ryzen 5 4500U notebook and the other being a Core i7 1165G7 in a premium Dell XPS 13 with Intel EVO certification but alas that's the only hardware available for this quick comparison at this time.
Both laptops at least had 16GB RAM, 256GB NVMe solid-state storage, integrated graphics, and similar screens. Both laptops saw fresh installs of Ubuntu 20.10 followed by installing all available stable release updates. The only other modifications made to the operating system were disabling of the automatic suspend mode when on battery. For maximizing the battery life when running on battery, both laptops also had their screen brightness set to their minimums.
While the Ryzen 5 4500U isn't close to being AMD's top-end Renoir SKU, as shown in past benchmarks it can generally stand its ground well against the Core i7 1165G7. But even still, the main focus of this article is to look at the relative impact of performance with each device on AC power versus battery. Again, due to limited hardware available, take these numbers as you wish are just some data points in response to Intel's recent presentation.
All the benchmarks, of course, were driven via the Phoronix Test Suite. During the benchmarking process the Phoronix Test Suite was monitoring the peak CPU frequency every second (the highest recorded CPU frequency on any of the CPU cores), the CPU core temperature, and also the battery power consumption rate when not on AC power.
Given the number of web browser benchmarks used, first up are a few tests within Mozilla Firefox and Google Chrome. The other tests picked out for this article aimed to be real-world and representative tests along with some overlap to Intel's testing.
First up was MotionMark running in Google Chrome. The high-end Core i7 1165G7 saw basically the same performance whether on AC or battery while the Ryzen 5 4500U was at roughly 92% the performance when operating on AC power. That's out-of-the-box on Ubuntu 20.10 without any modifications to the power management behavior, etc. Both laptop BIOS configurations were at their defaults too.
When looking at the CPU peak frequency, the clock behavior was basically the same for Tiger Lake whether on AC or battery. The mid-range Ryzen 5 4500U when operating on battery though had a 2.5GHz average speed for this test compared to 2.9GHz when connected to AC power. Both modes did successfully hit the 3.9~4.0GHz rated boost clock of the Ryzen 5 4500U.
Though one benefit of that behavior for the Renoir laptop when on battery power is a lower CPU temperature, which is good news if you are mobile and running the laptop on your lap.
But let's move on and look at more browser tests...