I have two laptops with Ubuntu 16.04 installed (kernel 4.4), both using metacity and gnome-*.

Laptop A - Lenovo W530

Core™ i7-3630QM 2.4GHz quad-core hyper-threading enabled

Integrated Graphics

single HDD

Laptop B - CybertronPC

Core™ i7-6700K 4.0GHz quad-core hyper-threading enabled

NVIDIA Quadros M1000M with proprietary driver installed

RAID-0 using 2 SSDs

Yet B is slower than A when running almost anything. In particular I ran a Python program that is CPU-bound (generate a random number 100 million times) and it takes twice as long in B.

I tried htop/top and couldn't pinpoint any single process taking up CPU in B. But B's htop usage bars seem definitely higher. How can I profile this situation and figure out where is the bottleneck? Any help would be much appreciated.

Upon further research I realized that potentially my RAID 0 is not hardware based, since laptops don't usually have it. I assumed that it was hardware based since it appears in BIOS. Any idea how to determine whether RAID 0 is hurting me?

  • Try Ubuntu 16.10, it has a newer kernel and better support for those 6th generation Intel CPUs. Also make sure you have the proprietary nVidia drivers installed for maximum performance. – Byte Commander Jan 27 '17 at 18:41
  • I already installed nVidia proprietary drivers. Are you saying 16.04 wouldn't work well with 6th generation CPUs, any links that you could point me to documenting this issue? Thanks! – Puppy Cyber Jan 27 '17 at 18:47
  • 1
    I personally have an i5-6200U and Ubuntu 16.04 works subjectively well with it. However, some special features of that CPU series like some low-power states and other stuff might not yet be well supported with the kernel versions 16.04 uses. 16.10 has a newer kernel version with improved support. What features exactly may be missing and how that affects your performance - I don't know. Entering the keywords "ubuntu linux skylake" or something similar into your preferred search engine will show you lots of articles. – Byte Commander Jan 27 '17 at 19:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.