03 March 2012

It's Here.

I present to you, fellow readers of this blog, the soon-to-be second fastest[1] student owned computer on MIT campus:
The beginning of an ultimate case mod for the ultimate hackintosh

Projected specifications:
  • Two Intel Xeon X5650 @ 4.0GHz
  • 24GiB DDR3-1600 RAM
  • Two AMD Radeon HD 6870
[1] The fastest will most certainly be this baby, if not for beefier GPU hardware then the soon-to-be liquid nitrogen pots sitting on top of the dual hexes. I also need to mention Cyril Lan's rig, which is powered by a 2500K @ 4.8GHz with dual AMD Radeon HD 7970s, but for most nonspecialized applications, CPUs dominate, sadly.


  1. what if (!!) someone outside EC has something ridiculous too?

  2. As I mentioned in the citation, Cyril Lan has a beast rig that will trounce this box in GPU performance. I haven't heard of anyone else who spends absurd amounts on hardware, though.

  3. It next step up from dual 4GHz+ hex-cores is quad 8-core Intel, and I've yet to meet someone with a 32-core system.
    Now, it is possible, but unlikely that someone on campus has a 48-core AMD box, if they happen to be Folding@Home fanatics...

  4. If anyone owns a 48+ core box, please post a self portrait of yourself with the box and a screen connected to the box with a screenshot of wPrime and CPU-Z!

  5. I seriously doubt this is the fastest computer on campus. I would be shocked if there are not several uber-performance-nerds on west campus you don't know about.

  6. …all the more motivation to start the Collegiate Overclocking League to expose those west campus overclockers…

  7. Hey...don't forget about the GPU fanatics. Though it won't matter for applications you buy, but for those of your own that you write you can get a lot more speed for your money with a nice GPU.

    My desktop + GTX-580 blows away the school cluster on my LBM codes.

    Just sayin'.

    1. I forgot to mention that you'll be obliterated once my friend (linked in my post) puts seven GTX460s in his SR-2.

    2. Also, for compute nothing beats the GTX 460 768 MB, which is under $100/TFLOPS right now and supports all the relevant CUDA/OpenCL-related updates.

    3. Haha! I never said mine was the fastest out there. :-)

      I get time on a machine with 6 C2070s that leaves mine in the dust too on my multi-GPU codes. (though per GPU, the GTX-580 beats the C2070 as long as the smaller card can hold the problem in memory)

      I have to agree though; dollar-for-dollar it's hard to beat the GTX460.

  8. I'm against getting a second GPU because I haven't done anything GPU intensive to ``warrant'' getting another one. Another limiting factor is that I insist on running OS X and using the stock drivers, which greatly limits the space of GPUs available.

  9. Have you seen this?


    I have also a Hackintosh, with a gtx580 gpu! Goes like hell! :)