The Power Efficiency Between Ubuntu 19.04, Clear Linux & openSUSE Tumbleweed With CompuLab's Airtop 3

[phoronix]

Phoronix: The Power Efficiency Between Ubuntu 19.04, Clear Linux & openSUSE Tumbleweed With CompuLab's Airtop 3

With CompuLab's incredibly well engineered Airtop 3 fan-less computer that is built to meet rugged industrial requirements while being loaded with an 8-core/16-thread Xeon CPU, NVIDIA Quadro RTX 4000 graphics, 64GB of RAM, and NVMe solid-state storage, here is an interesting benchmark comparison of Ubuntu 19.04, Clear Linux, and openSUSE Tumbleweed. Given the interesting system under test, not only is the raw performance being looked at but also the performance-per-Watt / AC power consumption and CPU thermal differences between these Linux operating systems.

http://www.phoronix.com/vr.php?view=28288

[linuxgeex]

The total kWh consumed to complete the entire test suite, at the wall outlet, would be a much better indicator of overall platform efficiency with the installed software. I bet Clear would beat Ubuntu by a small margin, and SuSE would lose by a large margin.

[andyprough]

Once again, these are not directly relevant comparisons since you left Tumbleweed on btrfs and KDE while running Clear and Ubuntu with EXT4 and Gnome. These tests only beg the question of what results you would get with Tumbleweed on EXT4 with Gnome, and what the power draw would be.

[tildearrow]

The total kWh consumed to complete the entire test suite, at the wall outlet, would be a much better indicator of overall platform efficiency with the installed software. I bet Clear would beat Ubuntu by a small margin, and SuSE would lose by a large margin.

Is OpenSUSE really that power-hungry...? I was thinking of upgrading my server to it...

[skeevy420]

Once again, these are not directly relevant comparisons since you left Tumbleweed on btrfs and KDE while running Clear and Ubuntu with EXT4 and Gnome. These tests only beg the question of what results you would get with Tumbleweed on EXT4 with Gnome, and what the power draw would be.

Yes they are. They're OOTB end-user comparisons. If Joe Dumbass goes and installs any of those distributions, that's what he'd get so that's what is compared.

[skeevy420]

Is OpenSUSE really that power-hungry...? I was thinking of upgrading my server to it...

On a desktop on first boot/the first day or two? Maybe. balooctl isn't known for being light on resources.

[andyprough]

If Joe Dumbass goes and installs any of those distributions

Joe Dumbass may be relevant to your use case, but it is not to mine.

[AsuMagic]

Once again, these are not directly relevant comparisons since you left Tumbleweed on btrfs and KDE while running Clear and Ubuntu with EXT4 and Gnome. These tests only beg the question of what results you would get with Tumbleweed on EXT4 with Gnome, and what the power draw would be.

Why? Defaults are defaults.

[skeevy420]

Joe Dumbass may be relevant to your use case, but it is not to mine.

When comparing a range of distributions on a fixed piece of hardware to see the power differences based on what the those distributions provide by default, the Joe Dumbass experience is exactly what needs to be tested because that's what regular users, not advanced users like ourselves, will use if they're actually capable of flashing a USB stick and installing Linux.

[horizonbrave]

Hi,
I'm Joe Dumbass.
And yes I use Linux everyday on my desktop.
Planning on staying dumb? Not sure, mummy say..