If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Announcement
Collapse
No announcement yet.
FreeBSD: A Faster Platform For Linux Gaming Than Linux?
You're not only comparing Linux vs FreeBSD. But also Unity vs KDE. And there's also some hardware differences. And then you come to the conclusion that the performance differences are due to the OS used?
Do most users switch to a 'lighter' DM when playing games? Equally do most users switch off compositing before running a game (yes I know Kwin can do that automatically)
The only thing I don't like about this article is the title it really should be "PC-BSD: A Faster Platform for Linux Gaming than Ubuntu"
Everything else is using the default settings (plus the Nvidia blob) to see how fast games run
You don't get more fairer than that really
I just want to point out that PC-BSD ships with KDE.
You're not only comparing Linux vs FreeBSD. But also Unity vs KDE. And there's also some hardware differences. And then you come to the conclusion that the performance differences are due to the OS used?
Ridiculous!
Linux versus Mac OS X comparisons compare not only Linux and Mac OS X, but whatever desktop environment Linux uses with the Mac OS X desktop environment. It is fair for a Ubuntu versus PC-BSD comparison. As for Linux versus FreeBSD, it would probably be better to use KDE on both systems.
So.. FreeBSD is faster in a small set of circumstances (nVidia GPU with closed drivers), but doesn't run most of the games in others (ATI, OS drivers). I'm not sure that counts as a win for FreeBSD.
Seriously, if I wanted a tainted kernel with a KDE workspace and great gaming performance, I'd install windows. Even works with ATIs closed drivers!
Still, these numbers are interesting - if there's a bottleneck in linux, it needs to be found and eliminated. If only the numbers were useful in determining that.
As others have mentioned, adding another distribution would be important to determine whether FreeBSD is faster than "Linux" or just faster than "Ubuntu". The article isn't too clear about possible differences due to different window managers, which always raises a red flag. It's difficult to trust the data when such issues aren't documented.
But, of course, the point of the headline is to generate clicks and ad-revenue, not to deliver useful numbers or research the actual causes for the difference. Why am I expecting more?
I would like to see FreeBSD against Gentoo Linux, which I assume that should be one of the best performance Linux distribution.
Only if you've set it up properly, and even then it might not make a large difference. Gentoo has its strengths in customisation and clean package management; performance is just a by-product of being a source-based distro. So you might see better performance in cpu-bound games that are compiled from source, I doubt you'll see much difference with something like Unigine's Heaven benchmark. That assumes all other things are equal of course, but they won't be: Gentoo being a rolling release will have different xserver versions, kernel, gcc, etc etc etc.
So using two somewhat different setups you find that the Nvidia BSD driver is faster than the Nvidia Linux driver?
Here's the thing, unless someone compares a barebone (low-end WM; composition disabled; services disabled; etc) on both ends, even this relatively benign statement above cannot be verified.
... The sad part of it is that AFAICS Michael is *indeed* trying to be productive, but somehow, his benchmark methodology always leaves something major behind
- Gilboa
Last edited by gilboa; 07 September 2011, 10:40 AM.
Comment