On 1/25/19 2:01 PM, Liam Proven wrote:
I do not like to see this Phoronix-bashing. It is a useful and valuable site, as the only one in the *large* field of PC performance comparison sites and communities which is Linux-based. The Linux world _needs_ this.
I don't think the site is valuable as long as the benchmark methods remain unscientific and questionable.
If we Linux users wish to see Linux continue to make headway against Windows, then sites like this are _necessary_.
Most Linux users don't care about competing with Windows. Linux has already surpassed Windows in all markets that are relevant. The only market left is the desktop market but there is not money to be made which is why the market has been on a decline for several years now.
If someone spends their own money on computers, then they want to know which are fast and which are slow, which are better or worse value for money. There *are* differences.
Sure. But what do you gain if the benchmarks produce incorrect or misleading results?
As well as differences between makes and models, there are also differences between performance under Windows and under Linux -- obviously.
Those performance differences are negligible for 95% of the userbase as the CPU is idling most of the time anyway. For users which seriously care about performance (e.g. HPC users), there are better and more scientific ways of running benchmarks.
Phoronix is the _only_ site doing such comparisons.
That is a valuable service.
There are tons of sites and computer magazines that do benchmarks.
If, perhaps, aspects of the design of openSUSE make it seem slower in benchmarks, either on first boot or in general, well, that is going to make it seem less appealing to people who want to get more performance out of their hardware by running Linux instead of Windows.
No user is going to run Linux over Windows for performance reasons. In fact, I would argue that the majority of users choose their operating system over the applications they are running. What do I gain by installing Linux when I need to work with the Adobe Creative Suite?
That being the case, there's only one answer: eliminate those performance differentials. Make it faster.
There is no point in trying to make it faster when the benchmark methods to determine the performance were flawed in the first place.
The answer is *not* to deride the people doing the measurements, because people do care about those measurements. That's life. Deal with it.
If the methods are completely flawed, it absolutely is. If someone tried to publish scientific paper based on the work methods that Phoronix is using, it would get torn into pieces by the scientific community in no time. Adrian -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org