FreeBSD: A Faster Platform For Linux Gaming Than Linux?

[Apopas]

How so? You could've used any of the following:

Naah... you really deserved something extraordinary...

[gilboa]

As much as I enjoy reading the overwhelmingly idiotic comments about Gentoo / Linux / Ubuntu and FreeBSD, I guess its time to say:
http://www.phoronix.com/scan.php?pag...desktops&num=1

It was a sensational story for the sake of, well, writing the sensational story.
I can't possibly believe that someone @Phoronix done two bencharmks in the same week, one ignoring the WM differences between kwin and Unity, and a couple of days later, a second benhmark comparing Unity to Unit2D and GNOME shell showing the huge effect a WM can have on gaming performance without connecting the dots.

Beyond that:
1. Kudos to the nVidia driver team, period. (Whether or not Linux / BSD / Windows is 5% slower or faster is completely irrelevant)
2. Last time I checked (7.3), the 64bit driver was noticeably slower than the Linux driver; I guess its time to install FreeBSD 8.2. (Or should I wait for 9?)
3. Beyond rethinking the very, very, very, questionable benchmarking methodology, isn't it high time to reconsider the selection of Ubunutu as master of all things?
As an alternative, I propose the following:
a. On each release cycle compare the 3 top distribution (e.g. Fedora, Ubunutu, OpenSUSE, etc).
b. Determine who's the best performing distribution.
c. Use *winner* against BSD/Windows/MAC/Android/DOS.
d. Sound's almost, err, logical, isn't it?

- Gilboa

[russofris]

@gilboa

You've hit the nail on the head, though I am curious to know the motive behind the article's publication.

Re: Gentoo posters.

Gentoo is a source distribution and package manager and fits together based on a combination of Arch and compiler flags. It simply cannot be compared to other distributions in any given performance comparison as it's extremely difficult to reach a consensus on which arch, cflags, makeopts, and overlays constitute a Gentoo installation. We all agree that it is fairly trivial to make a single target application perform faster than those provided by binary distributions. I applaud the efforts of Gentoo users, and view the top performance results of Gentoo installations to represent the potential performance that can be achieved by other performance seeking enthusiasts. That said, we should balance the results with the fact that many Gentoo users do not know what they are doing, have horribly mis-configured their installations to the point of barely working, and perform horribly at all but a few targeted tasks. Gentoo isn't the solution for performance issues, it's the path to the solution.

[gilboa]

russofris,

The target of the article was to generate a lot of traffic by gaining attention (E.g. Slashdor) - and it worked.

As for Gentoo, I fully agree.

- Gilboa

[Apopas]

@gilboa

it is called Ubuntu...

[kebabbert]

Interesting benchmarks. It seems that BSD devs did something right, when you consider that all Linux code has to be interpreted and then run on another OS - and STILL it runs faster on BSD. Cudos to BSD devs. What is wrong with Linux, why is it slower?

So, if you want to use Linux software with heavy graphics, you should install BSD instead. It would be interesting to see if BSD also runs other Linux software faster: software that is not graphics heavy, but just pure Linux software. "BSD runs faster Linux than Linux"?




UPDATE: As someone wrote:
"What's funny is that people are finding any reason they can to dismiss the benchmarks (my favorite is claiming the hardware is different, when it's not).

Meanwhile, nobody seemed to have a problem with Phoronix's previous benchmark showing Wine/Cedega games running faster on Linux than on Windows:
http://www.phoronix.com/scan.php?pag...item=681&num=2
The difference now is that Linux is on the losing end of the benchmark, so it simply must be incorrect in some way."



When Linux wins, everything is correct and good and fair. When Linux looses, something is wrong and the benchmark is bad and unfair. What is this behaviour called?

[gilboa]

@gilboa

it is called Ubuntu...

‏I know. (Typed the post at 1am after a very long workday.)

- Gilboa

[gilboa]

When Linux wins, everything is correct and good and fair. When Linux looses, something is wrong and the benchmark is bad and unfair. What is this behaviour called?

So, in short, you can't/won't compete with rather common logic (read: please compare oranges to oranges) so you switch to a wild generalized accusation (You Linux fanboys only cry when Linux loses, etc, etc, etc).

Am I to assume that you concede the fact that benchmark is badly conducted?

- Gilboa

[kebabbert]

So, in short, you can't/won't compete with rather common logic (read: please compare oranges to oranges) so you switch to a wild generalized accusation (You Linux fanboys only cry when Linux loses, etc, etc, etc).

Am I to assume that you concede the fact that benchmark is badly conducted?

- Gilboa

I am quoting someone else from slashdot, who observed this behavior. I am not the only one to found this out: When Linux wins, everything is good and fair. When Linux looses, something is wrong. Isn't this a bit odd?

Regarding if this benchmark is bad or good:
I know Linux people here that compared a single core 1GHz old SPARC vs 2.4GHz dual core x86 Linux, and drew the conclusion that "This benchmark is fair and correct: Linux is clearly faster than Solaris". When I objected, he called me names such as "FUD, Troll, Idiot, etc"

I also know a company that migrated 800 old SPARC servers to 4.000 new shiny Intel dual core 2.4GHz servers - and presented the same conclusion "Linux is much faster than Solaris". Fair and correct, yes?

So, we see that Linux people thinks such benchmarks vs antique Solaris servers are fair and good, and we also see that Linux people thinks this benchmark vs BSD is badly conducted. What conclusion can you draw regarding how Linux benchmark should be done?

The problem is BSD is running on same hardware, how can it be fair and correct then? They should have ran BSD on 800MHz Pentium 3 with 128 MB RAM, and only then Linux people would have called it a fair and correct benchmark. Why are Linux people afraid of using benchmarks on same hardware? Linux should always run on twice as fast hardware? So, what is this behaviour from Linux people, called?

[tbyte]

So, in short, you can't/won't compete with rather common logic (read: please compare oranges to oranges) so you switch to a wild generalized accusation (You Linux fanboys only cry when Linux loses, etc, etc, etc).

Am I to assume that you concede the fact that benchmark is badly conducted?

- Gilboa

Do You know what the meaning of "linux shill" is, Bilbo ?
Measuring by the level of your logic I'm picturing you as a low level windows support.