Intel Core i5 2500K Linux Performance

[V!NCENT]

The best game that was created after the 90's is GoldenEye: Source. Bar none. It has the best gameplay of an FPS since Wolfenstein3D. Bloom&co affects your fragging by 0%. If I stare at a light source and you're shooting your rocketlauncher at mee when I hold a Walter; you're a dead man.

Also CPU's are fast enough. Researchers should use OpenCL and consortium for the realy real work. Period.

So my laptop choice will have a minimum requirement of an AMD dedicated GPU. My current HD5770 is already doing 17.000.000 rays/s with fglrx with GLSL photon mapping and I'mlooking forward to building trees on Fusion and rendering ERPT on a Radeon.

My 0 cents.

[TrevorPH]

Btw; do you use a dedicated graphics card or the internal one? And from which repository do you get your current kernel? Thanks..

I bought mine as a command line only server so that I could have a test bed for stuff that I'm working on that will run on a Dell M610 blade server. It has no GUI.

For the kernel for Centos 5 you can get 2.6.36.2 from the ELRepo - usual caveats apply, if it breaks your system you get to keep all the pieces

[buzz]

However, for Linux users seeking to utilize the next-generation Intel HD graphics found on these new CPUs, it meant problems.

It only meant problems due to something wrong with your setup or drivers. As I pointed on on the comments for the graphics article, everything works fine with linux - every single one of your tests that failed, works fine here. I would have expected to see some correction here, as any new user reading the article would think that opengl and x acceleration isnt working well.

as other have pointed out, the cpu comparisons were wrong also. the different between graphics between the K and non K cpu is rather important.

I would ask again that something be amended on the graphics article. it is just plain wrong, and no doubt putting linux users of getting a h67 board to use the onboard graphics.

[buzz]

The only problem continuing to challenge us is the Intel HD Graphics support with Sandy Bridge, which is something we are continuing to tackle and by the time Ubuntu 11.04 rolls around it will hopefully be a pleasant "out of the box" experience for those running this new hardware.

if you need any help getting this working I am happy to advise. on an ubuntu maverick system, you can have accelerated X, compiz, and working opengl in around 10 mins, with no need to compile anything. I put the details on the comments for the other article. you just need

ppa:xorg-edgers/ppa - which includes needed linux 2.6.37 kernel as well as updated X packages
ppa:jools/sandybridge - for a compiz without blacklisted sandybridge ids

apt-get update/dist-upgrade install the new kernel reboot and you're done.

[sc3252]

Is there any good reason why Intel VT-D isn't in the K series, other than they want you to feel like you are missing something if you want to overclock? Also does anyone know if there is a performance benefit of having it over vt-x?

[buzz]

woops.

http://www.anandtech.com/show/4142/intel-discovers-bug-in-6series-chipset-begins-recall

[tamale]

how did you guys overclock and then run in ubuntu? every time I set the multipliers up and boot into ubuntu, I'm back at the default speeds (at least according to cat /proc/cpuinfo).. when I boot into windows, I'm running at the expected 4.7GHz.. what gives?

asus p8p67 deluxe
i7 2600k
ubuntu 10.10 / windows 7 dual boot

[buzz]

how did you guys overclock and then run in ubuntu? every time I set the multipliers up and boot into ubuntu, I'm back at the default speeds (at least according to cat /proc/cpuinfo).. when I boot into windows, I'm running at the expected 4.7GHz.. what gives?

asus p8p67 deluxe
i7 2600k
ubuntu 10.10 / windows 7 dual boot

I think there is an issue with it reporting the wrong information only. saw on another forum some firmware update on their board fixed it (i think). memory is a little vague. only have a h67 myself so no overclocking the cpu of my i5-2500k (nor interest to do so luckily )

[Michael]

Using the drm-intel-next kernel the experience is improved a bit for me... but I can still cause tiling corruption under certain cases, but I do have a few test profiles now working correctly most of the time.

Intel's Jesse Barnes has now also been able to reproduce my SNB Linux problems.

[Kenni]

Hmm, I've just upgraded one of my CentOS 5.5 systems with a Core i7 2600 + Intel DH67CL motherboard (with H67 chipset). Due to the problems with the integrated graphics mentioned in the last Sandy Bridge Phoronix article, I moved over the old and reliable Geforce 7xxx PCI-e 16x graphics card from the old system and disabled the integrated graphics permanently in the BIOS.

The system got exactly a few seconds past GRUB and then it crashed. I further disabled all the fancy stuff; USB3, onboard audio, SATA 3.0, etc. and after a few tries the system booted X11 and then it crashed after 20-30 seconds.

I still have several things to test in order to draw any conclusions, but it's not looking good so far

New BIOS, memtest86 and test of a newer Linux distribution is up next.

I finally identified the issue, the onboard SATA 3.0 has issues. Once I moved the HDDs to the SATA 2.0 controller, the system got rock stable.

It didn't matter if I ran Ubuntu 10.10, Ubuntu 11.04 Alpha1 or CentOS (with kernel 2.6.18 or 2.6.37), they all crashed eventually.

I don't know if it's a local HW problem on my board, but if your Sandy Bridge Intel board is unstable, try to skip the SATA 3.0 ports to see if it helps.