Benchmarked: Ubuntu vs Vista vs Win7

[zridling]

From TuxRadar.com:
A lot of people have been chattering about the improvements Windows 7 brings for Windows users, but how does it compare to Ubuntu in real-world tests? We put Ubuntu 8.10, Windows Vista and Windows 7 through their paces in both 32-bit and 64-bit tests to see just how well Ubuntu faces the new contender. And, just for luck, we threw in a few tests using Jaunty Jackalope with ext4.

When Windows users say that Windows 7 is easier to install than ever, what do they really mean? When they say it's faster, is it just in their heads, or is Microsoft really making big strides forward? And, perhaps most importantly, when Linux benchmarkers show us how screamingly fast ext4 is compared to ext3, how well do those figures actually transfer to end users?

These are the questions we wanted to answer, so we asked Dell to provide us with a high-spec machine to give all the operating systems room to perform to their max. Our test machine packed an Intel Core i7 920, which in layman's terms has four cores running at 2.67GHz with hyperthreading and 8MB of L3 cache. It also had 6GB of RAM, plus two 500GB of hard drives with 16MB of cache.

[attach]

[Eóin]

For the most part it's the final test which interests me most and Windows wins out there. And honestly, who gets a quad core, 6gb machine and only runs a bunch of file copying tests, pathetic I got to say.

[mouser]

Every benchmark ever created is attacked for being "not representative", so i don't think its fair to knock such things -- the site presented a mostly interesting set of comparisons.

The only benchmark i honestly view as absolutely irrelevant is installation time.. Unless you plan on reinstalling your OS once a week, i don't see why it really matters how long it takes you to install the OS.

[Eóin]

True it's easy to attack benchmarks, but these seem like a particularly eclectic set of choices to me.

There are a tonne of cross platform apps out there that they could have run. Sure then people would go and poke holes in those particular setups but it would still be a huge leap closer to "real-world tests".

[mouser]

There are a tonne of cross platform apps out there that they could have run. Sure then people would go and poke holes in those particular setups but it would still be a huge leap closer to "real-world tests".

fair enough -- they could have tested real work applications, that would be helpful.  that's not really a knock on the benchmarks they ran, but rather a need for additional benchmarks.

[f0dder]

Hmmm, were the file copy tests done on the same disk as the OS was installed to? In that case, it's a flawed benchmark (disks are slower towards the higher sector counts, and Windows takes up more space than linux). It also seems like a bad idea to turn off write caching... I wonder whether this means "turn off HD write caching" for linux but additionally "turn off FS write cache" for Windows? (there's a big difference between the two).

Obviously Windows does have to worry about some things that Linux doesn't, namely DRM checks

Oh ffs, STOP THROWING THAT BLANKET STATEMENT.

Also, they forget to mention which mount options are used for ext3/ext4 - like, does it run with journalling, and in that case, which kind? If they compare unjournalled ext3 to NTFS, no wonder NTFS loses out on lots-of-small-files

I'd love to see more 'scientific' benchmarks, as I'm pretty sure NTFS isn't always the best filesystem... but comparisons do have to be fair (ie., comparing to a journalled FS, copying from/to "comparable" locations (to avoid harddrive-speed issues), et cetera). This benchmark looks somewhat pseudo to me, but at least not as pseudo as the ones from that Adrian Kingsley-Hughes guy.

BTW: imho benchmark results shouldn't be averaged, as that means abnormal spikes will influence the result. Instead, the best timings should be chosen...

[Lashiec]

So, how many mouse clicks do I save during the installation routine if I upgrade the computer?

[zridling]

Our test machine packed an Intel Core i7 920, which in layman's terms has four cores running at 2.67GHz with hyperthreading and 8MB of L3 cache. It also had 6GB of RAM, plus two 500GB of hard drives with 16MB of cache.

[attach]

[f0dder]

My machine is a Q6600 quadcore (w/o HT) running 4 cores at 3.01GHz, and 8GB of ram, sporting 2x74GB raptor drives with 16MB of cache, nya-nyyyyyyyyya! (But obviously with less cache memory, less effective logical CPUs, less online storage, and slower RAM).

[bgd77]

I installed the 64 bit version of Windows 7 on a virtual machine (on a 32 bit host). The installation worked like a charm. And it seamed pretty fast to me (compared with XP or Vista). And compared with Ubuntu, it seemed to be about the same. But this is just my personal subjective opinion. Next time I will use a clock. 

[Edvard]

I can't wait to try Ubuntu 9.04, and I hope it doesn't disappoint. I finally admitted 2 months ago that Intrepid Ibex has been nothing but CRAP for me and I wouldn't use it if I hadn't installed everything on it already. 

Ext4 looks like a step in the right direction until either BtrFS gets it's legs  or ZFS gets a GPL-compatible license.

[40hz]

I can't wait to try Ubuntu 9.04, and I hope it doesn't disappoint. I finally admitted 2 months ago that Intrepid Ibex has been nothing but CRAP for me...

Ditto. I went back to LTS 8.0.4-2

(I also ditched the Pulse Audio Server, but that's a topic for a whole separate slag fest. )

Here's hoping that Ubuntu's new Juvenile Jackass (or whatever they decide to call it) is better than Ibex.

[Hirudin]

Argh, I'd LOVE to go on a rant here (actually I wrote and deleted a small rant) but I'll just say...

I'm glad I didn't read that article, it probably would have taken me at least 14 scroll-wheel turns to get the the bottom, and frankly that's just too much! Other Linux vs. Windows articles only require 8 scroll-wheel turns.

[urlwolf]

Well, actually it looks like they have improved IO quite a lot for Win 7. I wonder how they did that, as they are still using NTFS...

[zridling]

I can't wait to try Ubuntu 9.04, and I hope it doesn't disappoint. I finally admitted 2 months ago that Intrepid Ibex has been nothing but CRAP for me and I wouldn't use it if I hadn't installed everything on it already.    Ext4 looks like a step in the right direction until either BtrFS gets it's legs  or ZFS gets a GPL-compatible license.

8.10 punked me good. So I'm using openSUSE for the meantime. However, the latest Fedora 11 beta offers both BtrFS and Ext4 filesystems as defaults if you want.

[f0dder]

Well, actually it looks like they have improved IO quite a lot for Win 7. I wonder how they did that, as they are still using NTFS...

I'm not taking anything in that article serious since they disabled write caching

Btw, Windows (at least client versions) is usually way too conservative wrt. disk caching. Probably because so many morons are "OMFG IT USES THE MEMORY I HAVE INSTALLED IN MY SYSTEM!11!1!", coming from a total lack of understanding of how filesystem cache works.

[Edvard]

Here's hoping that Ubuntu's new Juvenile Jackass (or whatever they decide to call it) is better than Ibex.

8.10 punked me good. So I'm using openSUSE for the meantime. However, the latest Fedora 11 beta offers both BtrFS and Ext4 filesystems as defaults if you want.

from the Btrfs wiki:

Btrfs is under heavy development, and is not suitable for any uses other than benchmarking and review. The Btrfs disk format is not yet finalized, but it will only be changed if a critical bug is found and no workarounds are possible.

I can wait...

I'm not taking anything in that article serious since they disabled write caching

Btw, Windows (at least client versions) is usually way too conservative wrt. disk caching. Probably because so many morons are "OMFG IT USES THE MEMORY I HAVE INSTALLED IN MY SYSTEM!11!1!", coming from a total lack of understanding of how filesystem cache works.

I agree. Maybe they just wanted to "take some lead out of the gloves" to measure raw throughput? It does seem kind of unfair to not let a particular OS show off some of it's best tricks...