View RSS Feed

oldcpu's meandering thoughts on Computers, GNU/Linux and openSUSE

OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience

Rating: 2 votes, 4.50 average.
I recently purchased a Lenovo X1 Carbon 9th generation, with an Intel Core-i7-1165G7, 16GB RAM, 1 TB SSD and a 1920x1200 non-touch display, where I successfully installed openSUSE LEAP-15.3 on this machine.

My plan is that this will be a long blog entry, with multiple posts over a period of time as I gradually update software associated with this laptop. If I have issues I will post for 'help' elsewhere in this forum.

INTRODUCTION:

This laptop replaces an ancient Toshiba Z930 (w/core-i7-3667U) which is significantly slower – which means any improvement looks good to me, and I am not the most demanding person for high performance.

Note that I am a basic user, with video conferencing, and video editing / rendering being my most demanding potential use, although typically I do some video editing/rendering on a separate desktop PC. Only when traveling ‘on the road’ do I render with the ultrabook. So it is highly possible aspects that I am FULLY satisfied with on this laptop, won’t please others who are more demanding users.

I primarily purchased the X1 carbon Gen-9 for (1) light weight, (2) being very slim, and (3) having many external interfaces. Likely I will keep this for 5 to 8 years, which is another reason I went for the X1 Carbon Gen-9 as opposed to an older cheaper device.

In deciding on this device, I was encouraged by the fact that Lenovo do sell this laptop with either Fedora or Ubuntu pre-installed, instead of Windows-10, if one wishes to purchase a GNU/Linux version. In the case of this laptop, Lenovo do not sell it with openSUSE installed, although in the past Lenovo have sold other computers with SLED and SLES reference:

https://support.lenovo.com/us/en/solutions/pd031426

I purchased a version of this Lenovo X1 Carbon Gen-9 with Windows-10 pre-installed. Part of the bargain I struck with my wife (in her agreeing to my purchase such a new Lenovo) was that if I could not get openSUSE to work with this laptop, then she would get the laptop, and she is an MS-Windows user.

While I have installed variants of SuSE and openSUSE many times, this is my 1st openSUSE LEAP-15.3 install. While I have been using openSUSE (and SuSE-Pro prior) since 2003, until now I had yet until now to try LEAP-15.3.

INSTALLATION PREPARATION :

Pre-install LiveUSB checks

Prior to installing LEAP-15.3, I tested booting the laptop to different live-GNU/Linux USB variants (press F12 on boot to see the possibility to boot to a USB device):

* Knoppix-9.1 – liveUSB booted fine [Microknoppix Kernel: 5.10.10-64 x86_64 bits]. wifi worked. Sound did not work (Alsa info: http://alsa-project.org/db/?f=6803b75cc84178e5e5e2af1f55fe287f1029d9e4) . I did not know how to install sof-firmware under Knoppix so I could not confirm that would work. A full shutdown with “shutdown -h now” worked. Bluetooth was not identified in the app ‘inxi’. [ OpenGL: renderer: Mesa Intel Xe Graphics (TGL GT2) v: 4.6 Mesa 20.3.4 ]

* openSUSE-LEAP-15.3 – live USB booted fine. [Kernel: 5.3.18-59.19-default x86_64 bits]. Wifi worked. Sound only worked after installing sof-firmware and restarting sound with YaST. ( ALSA information before sof-firmware installed: http://alsa-project.org/db/?f=d004e9...f5b0bca527e709 and ALSA info after sof-firmware installed (audio now works): http://alsa-project.org/db/?f=e57ada...be20bf5c5fa99b ) . A full shutdown failed (only partly worked). Bluetooth was not identified in the app ‘inxi’. [ OpenGL: renderer: Mesa Intel Xe Graphics (TGL GT2) v: 4.6 Mesa 20.2.4 ]

* openSUSE-Tumbleweed (20210924 version)- liveUSB booted fine. [Kernel: 5.14.6-1-default x86_64 bits:]. Sound only worked after installing sof-firmware and restarting sound with YaST ( ALSA information before installing sof-firmware: http://alsa-project.org/db/?f=eae33c...3f3700aac0cff8 ) . Identified bluetooth but advised (in the app ‘inxi’) that “This feature requires one of these tools: hciconfig/bt-adapter”. A full shutdown with “shutdown -h now” worked. [ OpenGL: renderer: Mesa Intel Xe Graphics (TGL GT2) v: 4.6 Mesa 21.2.2 ]

Wifi worked with those live-USBs ( using the iwlwifi driver). Audio worked only after sof-firmare installed (for openSUSE liveUSBs).

Here is the file structure initially (as seen from a GNU/Linux boot USB with ‘parted -l’ ) BEFORE any SSD carving/partition creating (ie with no GNU/Liinux installed)
Code:
 root@Microknoppix:/home/knoppix# parted -l
  
 Model: SAMSUNG MZVL21T0HCLR-00BL7 (nvme)
  Disk /dev/nvme0n1: 1024GB

 Sector size (logical/physical): 512B/512B
 Partition Table: gpt
 Disk Flags:  
  
 Number  Start   End     Size    File system  Name                          Flags
  1      1049kB  274MB   273MB   fat32        EFI system partition          boot, hidden, esp
  2      274MB   290MB   16.8MB               Microsoft reserved partition  msftres
  3      290MB   1023GB  1023GB               Basic data partition          msftdata
  4      1023GB  1024GB  1049MB  ntfs         Basic data partition          hidden, diag
BIOS versions

I went to both the Lenovo support site, and the GNU/Linux ‘fwupd.org’ app site (which provides Lenovo firmware for installation under GNU/Linux) to see what versions of the BIOS were available:

On the ‘fwupd.org’ site v.1.47 was the latest version: https://fwupd.org/lvfs/search?value=X1+Carbon

Here is the link to v.1.47: https://fwupd.org/lvfs/devices/com.l...ETXXW.firmware

I also went to the Lenovo support site, and I noted a v.1.47 BIOS update there also:
https://pcsupport.lenovo.com/th/en/p...1-yoga-6th-gen

I checked my laptop, and it only had an older version 1.42 of UEFI BIOS (2021/06/15) where I also noted the latest was version 1.47 (issued 2021/09/15). Checking the change logs, I saw updates relevant to GNU/Linux and I knew I had a BIOS update to perform. One update in the UEFI BIOS update change logs in particular caught my eye (for the v.1.47 update): Fixed an issue where thermal throttling may happens on Linux OS”.

I had not used fwupd app before. Given this was a new install of openSUSE LEAP-15.3, where I had not read of even 1 person before me installing openSUSE on this 9th gen Lenovo X1 carbon, I decided to play safe and do the firmware update from Windows-10.

However Lenovo recommended (if updating the BIOS from MS-Windows) to only do so with the latest Windows updates in place, so that meant I had to update MS-Windows-10 first, which I did (and that took a while).

Carving up the SSD for GNU/Linux

After the MS-Windows update, and after updating the BIOS to v.1.47, I re-booted to Windows-10 and I immediately disabled hibernation, disabled page file, disable system protection, defragged the SSD (where defragging was likely not needed), and reduced the MS-Windows partition to about 80 GB (using Windows-10 disk management). I could have reduced it further to 65 GB had I wished. This left most of the remainder of the SSD for openSUSE (aside from some MS-Windows recovery partitions).

I then booted to the Knoppix-9.1 liveUSB and used ‘gparted’ to carve up the newly created unallocated SSD space to / (~25 GB) and /home (~890 GB). I also created a ~16 GB swap (possibly not needed).
Code:
  Model: NVMe Device (nvme)
 Disk /dev/nvme0n1: 1024GB
 Sector size (logical/physical): 512B/512B
 Partition Table: gpt
 Disk Flags:  
 
 
 Number  Start   End     Size    File system     Name                          Flags
  1      1049kB  274MB   273MB   fat32           EFI system partition          boot, hidden, esp
  2      274MB   290MB   16.8MB                  Microsoft reserved partition  msftres
  3      290MB   84.2GB  83.9GB                  Basic data partition          msftdata
  4      84.2GB  111GB   26.9GB  ext4            root
  6      111GB   129GB   17.6GB  linux-swap(v1)  swap                          swap
  7      129GB   1023GB  894GB   ext4            home
  5      1023GB  1024GB  1049MB  ntfs            Basic data partition          hidden, diag
Once this was created I then proceeded to successfully install openSUSE LEAP-15.3 (with secure boot disabled).

.... to be continued ....

Submit "OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience" to Digg Submit "OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience" to del.icio.us Submit "OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience" to StumbleUpon Submit "OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience" to Google Submit "OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience" to Facebook Submit "OpenSUSE LEAP-15.3 installation on Lenovo X1 Carbon Gen-9 – my experience" to Twitter

Categories
Uncategorized

Comments

Page 2 of 2 FirstFirst 12
  1. oldcpu's Avatar
    Quote Originally Posted by karlmistelberger
    Did you ever consider real stress testing?

    Intel® oneAPI Math Kernel Library (oneMKL) Benchmarks Suite

    https://www.intel.com/content/www/us...rks-suite.html
    In short - no.

    Thanks for the suggestion, but I confess that running benchmarks is not a particular interest of mine. Suffice to say this Lenovo X1C9 is a massive improvement over the i7-3667U CPU in my old Toshiba Z930 (where I purchased that Toshiba in June-2013). Further the nvme SSD in this Carbon X1G9 is faster than the SSD in my old Toshiba Z930 - where this is no surprise as that Toshiba hardware (and its still functioning SSD) is over 8-years old.

    After reading your post, I did download the Intel® oneAPI Math Kernel Library (oneMKL) Benchmarks Suite tarball (software), and looked inside the tarball. Its not intuitively obvious as to how to edit/run the script for the Tiger Lake Core-i7-1065G7 CPU in my Carbon X1G9 (nor run on my old core-i7-4770 CPU based desktop PC nor on my Toshiba Z930 core-i7-3667U laptop, nor on my even older core-i7-920 CPU based desktop PC either) and given my not having a strong interest in benchmarks, and also given 1st editing the benchmark script followed by running the benchmark script is recommended by the readme.txt file that comes with that package, I note that it is not intuitively obvious how to run (where for me a study of the tarball is needed). Hence I elected not to spend the time to try and figure it out. So likely I will not pursue this further.

    A benchmark program would need be much more obvious on how to run before I took the time to run such.
  2. karlmistelberger's Avatar
    [QUOTE=oldcpu;bt1280]In short - no.

    Thanks for the suggestion, but I confess that running benchmarks is not a particular interest of mine. Suffice to say this Lenovo X1C9 is a massive improvement over the i7-3667U CPU in my old Toshiba Z930 (where I purchased that Toshiba in June-2013). Further the nvme SSD in this Carbon X1G9 is faster than the SSD in my old Toshiba Z930 - where this is no surprise as that Toshiba hardware (and its still functioning SSD) is over 8-years old.

    After reading your post, I did download the Intel® oneAPI Math Kernel Library (oneMKL) Benchmarks Suite tarball (software), and looked inside the tarball. Its not intuitively obvious as to how to edit/run the script for the Tiger Lake Core-i7-1065G7 CPU in my Carbon X1G9 (nor run on my old core-i7-4770 CPU based desktop PC nor on my Toshiba Z930 core-i7-3667U laptop, nor on my even older core-i7-920 CPU based desktop PC either) and given my not having a strong interest in benchmarks, and also given 1st editing the benchmark script followed by running the benchmark script is recommended by the readme.txt file that comes with that package, I note that it is not intuitively obvious how to run (where for me a study of the tarball is needed). Hence I elected not to spend the time to try and figure it out. So likely I will not pursue this further.

    A benchmark program would need be much more obvious on how to run before I took the time to run such.[/QUOTE]
    No intuition required. Run ./runme_xeon64 and post output:

    karl@i3-4130:~/Downloads/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack> cat lin_xeon64.txt
    Do 21. Okt 12:16:21 CEST 2021
    Sample data file lininput_xeon64.

    Current date/time: Thu Oct 21 12:16:21 2021

    CPU frequency: 3.398 GHz
    Number of CPUs: 1
    Number of cores: 2
    Number of threads: 2

    Parameters are set to:

    Number of tests: 15
    Number of equations to solve (problem size) : 1000 2000 5000 10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000
    Leading dimension of array : 1000 2000 5008 10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000
    Number of trials to run : 4 2 2 2 2 2 2 2 2 2 1 1 1 1 1
    Data alignment value (in Kbytes) : 4 4 4 4 4 4 4 4 4 4 4 1 1 1 1

    Maximum memory requested that can be used=16200901024, at the size=45000

    =================== Timing linear equation system solver ===================

    Size LDA Align. Time(s) GFlops Residual Residual(norm) Check
    1000 1000 4 0.014 49.4279 1.150968e-12 3.449968e-02 pass
    1000 1000 4 0.012 56.4134 1.150968e-12 3.449968e-02 pass
    1000 1000 4 0.012 55.8050 1.150968e-12 3.449968e-02 pass
    1000 1000 4 0.012 56.0257 1.150968e-12 3.449968e-02 pass
    2000 2000 4 0.085 63.0172 4.198086e-12 3.329359e-02 pass
    2000 2000 4 0.084 63.5465 4.198086e-12 3.329359e-02 pass
    5000 5008 4 1.026 81.2428 2.548853e-11 3.354599e-02 pass
    5000 5008 4 1.025 81.3156 2.548853e-11 3.354599e-02 pass
    10000 10000 4 7.465 89.3281 9.941109e-11 3.352730e-02 pass
    10000 10000 4 7.462 89.3694 9.941109e-11 3.352730e-02 pass
    15000 15000 4 27.141 82.9164 1.931753e-10 2.912586e-02 pass
    15000 15000 4 24.620 91.4057 1.931753e-10 2.912586e-02 pass
    18000 18008 4 43.625 89.1372 3.179195e-10 3.367161e-02 pass
    18000 18008 4 47.449 81.9546 3.179195e-10 3.367161e-02 pass
    20000 20016 4 201.547 26.4659 3.447581e-10 2.966315e-02 pass
    20000 20016 4 57.388 92.9489 3.447581e-10 2.966315e-02 pass
    22000 22008 4 75.615 93.8924 4.128938e-10 2.933052e-02 pass
    22000 22008 4 76.457 92.8573 4.128938e-10 2.933052e-02 pass
    25000 25000 4 109.717 94.9528 6.607437e-10 3.641903e-02 pass
    25000 25000 4 109.437 95.1960 6.607437e-10 3.641903e-02 pass
    26000 26000 4 122.875 95.3704 5.872723e-10 2.991647e-02 pass
    26000 26000 4 123.091 95.2033 5.872723e-10 2.991647e-02 pass
    27000 27000 4 137.248 95.6185 5.860798e-10 2.780977e-02 pass
    30000 30000 1 187.889 95.8111 7.834016e-10 3.015989e-02 pass
    35000 35000 1 323.327 88.4113 1.271780e-09 3.601975e-02 pass
    40000 40000 1 471.701 90.4595 1.371931e-09 2.976481e-02 pass
    Done: Do 21. Okt 12:54:41 CEST 2021
    karl@i3-4130:~/Downloads/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack>
  3. oldcpu's Avatar
    OK, thanks.

    As I noted I am not nominally one for benchmarks.

    There are so many different benchmarks and to know which are most applicable as to how I use a computer is not something I have spent time to research. My very fuzzy understanding is assessing a computer's performance to assess benefit for an individual's use, depends on the same individual's nominal use of the computer.

    I believe there are CPU bench marks for mathematical calculations, benchmarks for video rendering, hard-drive/SSD/nvme storage benchmarks, benchmarks for I/O (ie USB I/O, thunderbolt I/O etc ... ), graphical display benchmarks, and even GPU calcuation benchmarks (?)(for machine learning and such).

    Reading the specs and skimming through various reviews on the Lenovo X1 Carbon Generation-9 (X1C9) I am very confident it is a significant step up from my older laptop, a Toshiba Z930 (where that Toshiba came out in 2012).

    Since you asked about a specific benchmark (and provided the command to run such) I ran that command. Also since my interest was an improvement over my old laptop (and not so much against any one else's device) I ran that against my older laptop.

    This is the benchmark against my new Lenovo X1C9 which is likely more relevant to this thread, and also more likely of interest to anyone reading this thread:

    Code:
    oldcpu@X1-Carbon-G9:~/rpms/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack> ./runme_xeon64
    This is a SAMPLE run script for running a shared-memory version of
    Intel(R) Distribution for LINPACK* Benchmark. Change it to reflect
    the correct number of CPUs/threads, problem input files, etc..
    *Other names and brands may be claimed as the property of others.
    Sat 13 Nov 2021 05:32:45 PM +07
    Sample data file lininput_xeon64.
    
    Current date/time: Sat Nov 13 17:32:45 2021
    
    CPU frequency:    4.088 GHz
    Number of CPUs: 1
    Number of cores: 4
    Number of threads: 4
    
    Parameters are set to:
    
    Number of tests: 15
    Number of equations to solve (problem size) : 1000  2000  5000  10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000
    Leading dimension of array                  : 1000  2000  5008  10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000
    Number of trials to run                     : 4     2     2     2     2     2     2     2     2     2     1     1     1     1     1    
    Data alignment value (in Kbytes)            : 4     4     4     4     4     4     4     4     4     4     4     1     1     1     1    
    
    Maximum memory requested that can be used=16200901024, at the size=45000
    
    =================== Timing linear equation system solver ===================
    
    Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
    1000   1000   4      0.005      135.8550 1.052710e-12 3.143422e-02   pass
    1000   1000   4      0.004      169.0633 1.052710e-12 3.143422e-02   pass
    1000   1000   4      0.004      166.8704 1.052710e-12 3.143422e-02   pass
    1000   1000   4      0.004      165.6777 1.052710e-12 3.143422e-02   pass
    2000   2000   4      0.029      184.4839 4.597892e-12 3.588984e-02   pass
    2000   2000   4      0.028      189.7658 4.597892e-12 3.588984e-02   pass
    5000   5008   4      0.398      209.4370 2.669148e-11 3.543370e-02   pass
    5000   5008   4      0.402      207.6614 2.669148e-11 3.543370e-02   pass
    10000  10000  4      3.169      210.4173 9.466089e-11 3.192014e-02   pass
    10000  10000  4      3.321      200.8272 9.466089e-11 3.192014e-02   pass
    15000  15000  4      16.591     135.6422 2.224910e-10 3.365899e-02   pass
    15000  15000  4      17.794     126.4729 2.224910e-10 3.365899e-02   pass
    18000  18008  4      30.179     128.8540 2.999166e-10 3.174851e-02   pass
    18000  18008  4      30.020     129.5363 2.999166e-10 3.174851e-02   pass
    20000  20016  4      40.980     130.1637 3.406316e-10 2.914867e-02   pass
    20000  20016  4      40.953     130.2486 3.406316e-10 2.914867e-02   pass
    22000  22008  4      54.296     130.7576 4.453436e-10 3.156515e-02   pass
    22000  22008  4      54.281     130.7935 4.453436e-10 3.156515e-02   pass
    25000  25000  4      79.603     130.8735 5.747349e-10 3.169419e-02   pass
    25000  25000  4      79.572     130.9236 5.747349e-10 3.169419e-02   pass
    26000  26000  4      89.467     130.9832 5.959303e-10 3.043206e-02   pass
    26000  26000  4      89.481     130.9622 5.959303e-10 3.043206e-02   pass
    27000  27000  4      99.950     131.3000 7.263368e-10 3.444243e-02   pass
    30000  30000  1      136.744    131.6459 8.100526e-10 3.118683e-02   pass
    35000  35000  1      217.075    131.6864 1.095406e-09 3.095182e-02   pass
    40000  40000  1      323.903    131.7366 1.273924e-09 2.766933e-02   pass
    45000  45000  1      506.997    119.8311 1.813956e-09 3.114033e-02   pass
    
    Performance Summary (GFlops)
    
    Size   LDA    Align.  Average  Maximal
    1000   1000   4       159.3666 169.0633
    2000   2000   4       187.1248 189.7658
    5000   5008   4       208.5492 209.4370
    10000  10000  4       205.6223 210.4173
    15000  15000  4       131.0575 135.6422
    18000  18008  4       129.1951 129.5363
    20000  20016  4       130.2061 130.2486
    22000  22008  4       130.7756 130.7935
    25000  25000  4       130.8985 130.9236
    26000  26000  4       130.9727 130.9832
    27000  27000  4       131.3000 131.3000
    30000  30000  1       131.6459 131.6459
    35000  35000  1       131.6864 131.6864
    40000  40000  1       131.7366 131.7366
    45000  45000  1       119.8311 119.8311
    
    Residual checks PASSED
    
    End of tests
    
    Done: Sat 13 Nov 2021 06:07:08 PM +07
    oldcpu@X1-Carbon-G9:~/rpms/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack>
    For my own edification, I also ran that 'benchmark' against very old Toshbia Z930 laptop (which has a Core-i7 3667U CPU). I note my Toshiba Z930 was not able to run the full benchmark range :
    Code:
    oldcpu@localhost:~/rpms/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack> ./runme_xeon64
    This is a SAMPLE run script for running a shared-memory version of
    Intel(R) Distribution for LINPACK* Benchmark. Change it to reflect
    the correct number of CPUs/threads, problem input files, etc..
    *Other names and brands may be claimed as the property of others.
    Sun Nov 14 09:01:31 +07 2021
    Sample data file lininput_xeon64.
    
    Current date/time: Sun Nov 14 09:01:31 2021
    
    CPU frequency:    3.192 GHz
    Number of CPUs: 1
    Number of cores: 2
    Number of threads: 2
    
    Parameters are set to:
    
    Number of tests: 15
    Number of equations to solve (problem size) : 1000  2000  5000  10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000
    Leading dimension of array                  : 1000  2000  5008  10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000
    Number of trials to run                     : 4     2     2     2     2     2     2     2     2     2     1     1     1     1     1    
    Data alignment value (in Kbytes)            : 4     4     4     4     4     4     4     4     4     4     4     1     1     1     1    
    
    Maximum memory requested that can be used=12800801024, at the size=40000
    
    =================== Timing linear equation system solver ===================
    
    Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
    1000   1000   4      0.032      20.8308  1.108169e-12 3.311275e-02   pass
    1000   1000   4      0.028      23.8896  1.108169e-12 3.311275e-02   pass
    1000   1000   4      0.025      26.6411  1.108169e-12 3.311275e-02   pass
    1000   1000   4      0.028      24.0372  1.108169e-12 3.311275e-02   pass
    2000   2000   4      0.156      34.2321  4.780135e-12 3.797053e-02   pass
    2000   2000   4      0.156      34.1773  4.780135e-12 3.797053e-02   pass
    5000   5008   4      2.188      38.1164  2.585496e-11 3.402825e-02   pass
    5000   5008   4      2.256      36.9612  2.585496e-11 3.402825e-02   pass
    10000  10000  4      17.274     38.6045  8.918638e-11 3.006692e-02   pass
    10000  10000  4      17.394     38.3398  8.918638e-11 3.006692e-02   pass
    15000  15000  4      78.718     28.5889  2.052349e-10 3.094689e-02   pass
    15000  15000  4      83.614     26.9148  2.052349e-10 3.094689e-02   pass
    18000  18008  4      152.687    25.4681  2.830931e-10 2.992661e-02   pass
    18000  18008  4      144.801    26.8552  2.830931e-10 2.992661e-02   pass
    20000  20016  4      208.456    25.5887  3.773184e-10 3.239039e-02   pass
    20000  20016  4      198.428    26.8820  3.773184e-10 3.239039e-02   pass
    22000  22008  4      268.268    26.4647  5.157639e-10 3.658542e-02   pass
    22000  22008  4      271.300    26.1690  5.157639e-10 3.658542e-02   pass
    25000  25000  4      396.423    26.2798  5.611069e-10 3.098006e-02   pass
    25000  25000  4      412.766    25.2393  5.611069e-10 3.098006e-02   pass
    26000  26000  4      474.254    24.7097  6.411192e-10 3.263811e-02   pass
    26000  26000  4      465.019    25.2005  6.411192e-10 3.263811e-02   pass
    27000  27000  4      660.732    19.8620  6.859950e-10 3.255930e-02   pass
    30000  30000  1      1169.022   15.3990  7.734124e-10 2.978534e-02   pass
    35000  35000  1      2918.538   9.7946   1.032016e-09 2.921610e-02   pass
    ./runme_xeon64: line 41:  2711 Killed                  $numacmd ./xlinpack_$arch lininput_$arch
    Done: Sun Nov 14 11:34:59 +07 2021
    oldcpu@localhost:~/rpms/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack>
    I ran the benchmark twice to be certain, and both times it stopped at the same place with my old Toshiba Z930.
  4. oldcpu's Avatar
    Continuing from the above post ...

    From the perspective of that benchmark, it does appear I obtain a significant performance gain with the Lenovo X1C9 over the old Toshiba Z930, which is enough to satisfy myself. This was what I wanted in buying the newer laptop.

    I also like the superior interface functionality of this X1C9 in regards to Thunderbolt ... and its newer Bluetooth specification, and of course its superior display. It also has a larger nvme storage (1TB) than the 256GB SSD storage on my Toshiba. The X1C9's speakers can put out much more volume than the old Toshiba Z930 (meaning I may not carry a small USB portable travel speaker like I did with the Toshiba). I believe this laptop should be superior in every respect to my older laptop, except possibly weight, where this Lenovo X1C9 is about the same weight (possibly a smidgen heavier) than my old Toshiba Z930. Obtaining a light weight is very very important to me, and I came close to buying the lighter Lenovo Nano instead, but in the end went for the X1C9 mainly due to the X1C9 having more external interfaces..

    One very nice feature of this X1C9 is that GNU/Linux is supported by Lenovo for this laptop. While Lenovo nominally support Fedora and Ubuntu for this laptop (as they sell the X1C9 with those distros installed), on the Lenovo Linux forum they will support other distributions questions - which I believe is excellent on the part of Lenovo.

    As you can tell - at present time, I am a happy customer (of Lenovo).
    .
  5. karlmistelberger's Avatar
    Many thanks for running the test and thoughtful commenting. Linpack is easy to use, but widely ignored by computer magazines. However users may even test their smartphones: https://media.ccc.de/v/35c3-9703-supermuc-ng#t=1988 I run it on all machines I assemble, no matter whether users will stress their machines to the limit. Modern hardware performs thermal throttling thus avoiding overheating.
Page 2 of 2 FirstFirst 12