Minisforum MS-R1

[geerlingguy]

Basic information

• Board URL (official): https://s.minisforum.com/4hRFZbj
• Board purchased from: Provided by Minisforum
• Board purchase date: November 10, 2025
• Board specs (as tested): 64GB RAM, 1TB HDD
• Board price (as tested): $695.90 (MSRP $869)

Linux/system information

# output of `screenfetch`
         _,met$$$$$gg.           OS: Debian 12 bookworm
      ,g$$$$$$$$$$$$$$$P.        Kernel: aarch64 Linux 6.6.10-cix-build-generic
    ,g$$P""       """Y$$.".      Shell: bash 5.2.15
   ,$$P'              `$$$.      DE: GNOME 43.6
  ',$$P       ,ggs.     `$$b:    WM Theme: 
  `d$$'     ,$P"'   .    $$$     Icon Theme: Adwaita
   $$P      d$'     ,    $$P     Disk: 12G / 925G (2%)
   $$:      $$.   -    ,d$$'     RAM: 2056MiB / 63833MiB
   $$\;      Y$b._   _,d$P'      Uptime: 2m
   Y$$.    `.`"Y$$$$P"'          Packages: 1433
   `$$b      "-.__               GTK Theme: Adwaita [GTK2/3]
    `Y$$                         Font: Cantarell 11
     `Y$$.                       Resolution: 3840x2160
       `$$b.                     WM: Mutter
         `Y$$b.                  CPU: CIX P1 CP8180 @ 12x 2.6GHz
            `"Y$b._              mini@ms-r1
                `""""            
                                

# output of `uname -a`
Linux ms-r1 6.6.10-cix-build-generic #2 SMP PREEMPT Fri Oct 10 10:21:15 UTC 2025 aarch64 GNU/Linux

Benchmark results

CPU

• Geekbench 6: (1336 single / 6773 multi - https://browser.geekbench.com/v6/cpu/14893385)
• 143.03 Gflops at 39W for 3.67 Gflops/W (geerlingguy/top500-benchmark HPL result)

Power

• Idle power draw (at wall): 17 W (18.1 W with 10 Gbps Network)
• Maximum simulated power draw (stress-ng --matrix 0): 30.3 W
• During Geekbench multicore benchmark: 34.7 W
• During top500 HPL benchmark: 40 W

Disk

KINGSTON OM8TAP41024K1-A00

Benchmark	Result
iozone 4K random read	70.36 MB/s
iozone 4K random write	253.85 MB/s
iozone 1M random read	2783.00 MB/s
iozone 1M random write	4678.96 MB/s
iozone 1M sequential read	3218.89 MB/s
iozone 1M sequential write	4678.47 MB/s

Network

iperf3 results:

Built-in 10 Gbps NIC (Realtek RTL8127)

• iperf3 -c $SERVER_IP: 9.40 Gbps
• iperf3 -c $SERVER_IP --reverse: 3.66 Gbps
• iperf3 -c $SERVER_IP --bidir: 7.87 Gbps up, 1.97 Gbps down

Built-in WiFi 6E (Mediatek MT7922 802.11ax)

• iperf3 -c $SERVER_IP: 925 Mbps
• iperf3 -c $SERVER_IP --reverse: 659 Mbps
• iperf3 -c $SERVER_IP --bidir: 497 Mbps up, 385 Mbps down

(Be sure to test all interfaces, noting any that are non-functional.)

GPU

glmark2

glmark2-es2-wayland results:

=======================================================
    glmark2 2023.01
=======================================================
    OpenGL Information
    GL_VENDOR:      ARM
    GL_RENDERER:    Mali-G720-Immortalis
    GL_VERSION:     OpenGL ES 3.2 v1.r53p0-00eac0.c707efa0cfa034b363bc93f9b6749cb5
    Surface Config: buf=32 r=8 g=8 b=8 a=8 depth=24 stencil=0 samples=0
    Surface Size:   800x600 windowed
=======================================================
[build] use-vbo=false: FPS: 8009 FrameTime: 0.125 ms
[build] use-vbo=true: FPS: 8127 FrameTime: 0.123 ms
[texture] texture-filter=nearest: FPS: 8954 FrameTime: 0.112 ms
[texture] texture-filter=linear: FPS: 9057 FrameTime: 0.110 ms
[texture] texture-filter=mipmap: FPS: 7269 FrameTime: 0.138 ms
[shading] shading=gouraud: FPS: 7464 FrameTime: 0.134 ms
[shading] shading=blinn-phong-inf: FPS: 7733 FrameTime: 0.129 ms
[shading] shading=phong: FPS: 7362 FrameTime: 0.136 ms
[shading] shading=cel: FPS: 7340 FrameTime: 0.136 ms
[bump] bump-render=high-poly: FPS: 4475 FrameTime: 0.223 ms
[bump] bump-render=normals: FPS: 8431 FrameTime: 0.119 ms
[bump] bump-render=height: FPS: 8068 FrameTime: 0.124 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 7375 FrameTime: 0.136 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 6361 FrameTime: 0.157 ms
[pulsar] light=false:quads=5:texture=false: FPS: 8637 FrameTime: 0.116 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 3448 FrameTime: 0.290 ms
[desktop] effect=shadow:windows=4: FPS: 5792 FrameTime: 0.173 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 1033 FrameTime: 0.968 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 967 FrameTime: 1.035 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 1504 FrameTime: 0.665 ms
[ideas] speed=duration: FPS: 3383 FrameTime: 0.296 ms
[jellyfish] <default>: FPS: 7069 FrameTime: 0.141 ms
[terrain] <default>: FPS: 540 FrameTime: 1.852 ms
[shadow] <default>: FPS: 6465 FrameTime: 0.155 ms
[refract] <default>: FPS: 1637 FrameTime: 0.611 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 8265 FrameTime: 0.121 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 7309 FrameTime: 0.137 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 8383 FrameTime: 0.119 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 7999 FrameTime: 0.125 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 7227 FrameTime: 0.138 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 7884 FrameTime: 0.127 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 7811 FrameTime: 0.128 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 7313 FrameTime: 0.137 ms
=======================================================
                                  glmark2 Score: 6322 
=======================================================

vkmark

vkmark results (had to be compiled as it wasn't available on the default Debian 12 package repository):

mini@ms-r1:~/Downloads/vkmark$ DISPLAY=:0 build/src/vkmark --winsys-dir=build/src --data-dir=data
Authorization required, but no authorization protocol specified

Authorization required, but no authorization protocol specified

Error: Failed to find a connected drm connector

I also tried using vulkaninfo, but got:

mini@ms-r1:~/Downloads/vkmark$ vulkaninfo 
Segmentation fault

Apparently some users got Vulkan going on the Radxa Orion O6, but reported incredibly slow inference with llama.cpp, suggesting the memory connection between the iGPU and the LPDDR5 RAM is not so great.

GravityMark

GravityMark result: 3,037

AI / LLM Inference

Basic ollama LLM model inference results:

System	CPU/GPU	Model	Eval Rate	Power (Peak)
Minisforum MS-R1	CPU	llama3.2:3b	12.12 Tokens/s	35 W
Minisforum MS-R1	CPU	llama3.1:8b	5.78 Tokens/s	35 W
Minisforum MS-R1	CPU	llama2:13b	3.91 Tokens/s	37.2 W
Minisforum MS-R1	CPU	llama3.1:70b	0.77 Tokens/s	38.2 W
Minisforum MS-R1 (Nvidia RTX A2000)	GPU	llama3.2:3b	71.36 Tokens/s	94.3 W
Minisforum MS-R1 (Nvidia RTX A2000)	GPU	llama3.1:8b	34.52 Tokens/s	97.3 W

More results: geerlingguy/ai-benchmarks#32

Ollama was run on the CPU for these tests, as I could not yet get Vulkan support working, and could not get either Llama.cpp or Ollama to use Vulkan on the iGPU.

Memory

tinymembench results:

Click to expand memory benchmark result

tinymembench v0.4.10 (simple benchmark for memory throughput and latency)

==========================================================================
== Memory bandwidth tests                                               ==
==                                                                      ==
== Note 1: 1MB = 1000000 bytes                                          ==
== Note 2: Results for 'copy' tests show how many bytes can be          ==
==         copied per second (adding together read and writen           ==
==         bytes would have provided twice higher numbers)              ==
== Note 3: 2-pass copy means that we are using a small temporary buffer ==
==         to first fetch data into it, and only then write it to the   ==
==         destination (source -> L1 cache, L1 cache -> destination)    ==
== Note 4: If sample standard deviation exceeds 0.1%, it is shown in    ==
==         brackets                                                     ==
==========================================================================

 C copy backwards                                     :  15141.6 MB/s (3.3%)
 C copy backwards (32 byte blocks)                    :  15462.1 MB/s (0.7%)
 C copy backwards (64 byte blocks)                    :  15509.4 MB/s (0.4%)
 C copy                                               :  15559.3 MB/s (2.7%)
 C copy prefetched (32 bytes step)                    :  14131.9 MB/s (1.2%)
 C copy prefetched (64 bytes step)                    :  14256.6 MB/s (1.9%)
 C 2-pass copy                                        :  14304.2 MB/s (0.4%)
 C 2-pass copy prefetched (32 bytes step)             :  13702.7 MB/s
 C 2-pass copy prefetched (64 bytes step)             :  14443.0 MB/s
 C fill                                               :  35112.0 MB/s (2.1%)
 C fill (shuffle within 16 byte blocks)               :  35279.8 MB/s (0.3%)
 C fill (shuffle within 32 byte blocks)               :  35368.0 MB/s
 C fill (shuffle within 64 byte blocks)               :  35707.0 MB/s (0.6%)
 NEON 64x2 COPY                                       :  16145.3 MB/s (0.3%)
 NEON 64x2x4 COPY                                     :  16148.5 MB/s (0.2%)
 NEON 64x1x4_x2 COPY                                  :  15632.7 MB/s (0.2%)
 NEON 64x2 COPY prefetch x2                           :  15784.8 MB/s
 NEON 64x2x4 COPY prefetch x1                         :  16216.8 MB/s (0.4%)
 NEON 64x2 COPY prefetch x1                           :  16329.7 MB/s
 NEON 64x2x4 COPY prefetch x1                         :  16288.4 MB/s (0.3%)
 ---
 standard memcpy                                      :  16556.4 MB/s (1.7%)
 standard memset                                      :  41295.1 MB/s (1.9%)
 ---
 NEON LDP/STP copy                                    :  16698.8 MB/s (2.0%)
 NEON LDP/STP copy pldl2strm (32 bytes step)          :  15435.9 MB/s (2.1%)
 NEON LDP/STP copy pldl2strm (64 bytes step)          :  15425.3 MB/s (1.1%)
 NEON LDP/STP copy pldl1keep (32 bytes step)          :  15966.3 MB/s (1.0%)
 NEON LDP/STP copy pldl1keep (64 bytes step)          :  16154.9 MB/s (0.4%)
 NEON LD1/ST1 copy                                    :  15877.5 MB/s
 NEON STP fill                                        :  36616.6 MB/s (2.6%)
 NEON STNP fill                                       :  34990.0 MB/s (2.8%)
 ARM LDP/STP copy                                     :  16254.1 MB/s (1.7%)
 ARM STP fill                                         :  39698.1 MB/s (3.4%)
 ARM STNP fill                                        :  35023.4 MB/s (2.8%)

==========================================================================
== Framebuffer read tests.                                              ==
==                                                                      ==
== Many ARM devices use a part of the system memory as the framebuffer, ==
== typically mapped as uncached but with write-combining enabled.       ==
== Writes to such framebuffers are quite fast, but reads are much       ==
== slower and very sensitive to the alignment and the selection of      ==
== CPU instructions which are used for accessing memory.                ==
==                                                                      ==
== Many x86 systems allocate the framebuffer in the GPU memory,         ==
== accessible for the CPU via a relatively slow PCI-E bus. Moreover,    ==
== PCI-E is asymmetric and handles reads a lot worse than writes.       ==
==                                                                      ==
== If uncached framebuffer reads are reasonably fast (at least 100 MB/s ==
== or preferably >300 MB/s), then using the shadow framebuffer layer    ==
== is not necessary in Xorg DDX drivers, resulting in a nice overall    ==
== performance improvement. For example, the xf86-video-fbturbo DDX     ==
== uses this trick.                                                     ==
==========================================================================

 NEON LDP/STP copy (from framebuffer)                 :   2156.6 MB/s
 NEON LDP/STP 2-pass copy (from framebuffer)          :   1524.4 MB/s
 NEON LD1/ST1 copy (from framebuffer)                 :   2156.4 MB/s
 NEON LD1/ST1 2-pass copy (from framebuffer)          :   1529.3 MB/s
 ARM LDP/STP copy (from framebuffer)                  :   1934.8 MB/s
 ARM LDP/STP 2-pass copy (from framebuffer)           :   1211.3 MB/s

==========================================================================
== Memory latency test                                                  ==
==                                                                      ==
== Average time is measured for random memory accesses in the buffers   ==
== of different sizes. The larger is the buffer, the more significant   ==
== are relative contributions of TLB, L1/L2 cache misses and SDRAM      ==
== accesses. For extremely large buffer sizes we are expecting to see   ==
== page table walk with several requests to SDRAM for almost every      ==
== memory access (though 64MiB is not nearly large enough to experience ==
== this effect to its fullest).                                         ==
==                                                                      ==
== Note 1: All the numbers are representing extra time, which needs to  ==
==         be added to L1 cache latency. The cycle timings for L1 cache ==
==         latency can be usually found in the processor documentation. ==
== Note 2: Dual random read means that we are simultaneously performing ==
==         two independent memory accesses at a time. In the case if    ==
==         the memory subsystem can't handle multiple outstanding       ==
==         requests, dual random read has the same timings as two       ==
==         single reads performed one after another.                    ==
==========================================================================

block size : single random read / dual random read, [MADV_NOHUGEPAGE]
      1024 :    0.0 ns          /     0.0 ns 
      2048 :    0.0 ns          /     0.0 ns 
      4096 :    0.0 ns          /     0.0 ns 
      8192 :    0.0 ns          /     0.0 ns 
     16384 :    0.0 ns          /     0.0 ns 
     32768 :    0.0 ns          /     0.0 ns 
     65536 :    0.0 ns          /     0.0 ns 
    131072 :    1.0 ns          /     1.5 ns 
    262144 :    1.5 ns          /     2.1 ns 
    524288 :    2.1 ns          /     2.4 ns 
   1048576 :   18.4 ns          /    26.7 ns 
   2097152 :   26.6 ns          /    33.6 ns 
   4194304 :   31.0 ns          /    35.9 ns 
   8388608 :   33.7 ns          /    36.8 ns 
  16777216 :   41.2 ns          /    48.0 ns 
  33554432 :  109.9 ns          /   153.1 ns 
  67108864 :  173.4 ns          /   217.5 ns 

block size : single random read / dual random read, [MADV_HUGEPAGE]
      1024 :    0.0 ns          /     0.0 ns 
      2048 :    0.0 ns          /     0.0 ns 
      4096 :    0.0 ns          /     0.0 ns 
      8192 :    0.0 ns          /     0.0 ns 
     16384 :    0.0 ns          /     0.0 ns 
     32768 :    0.0 ns          /     0.0 ns 
     65536 :    0.0 ns          /     0.0 ns 
    131072 :    1.0 ns          /     1.5 ns 
    262144 :    1.5 ns          /     2.1 ns 
    524288 :    1.8 ns          /     2.4 ns 
   1048576 :   17.5 ns          /    26.1 ns 
   2097152 :   25.3 ns          /    32.3 ns 
   4194304 :   29.1 ns          /    34.4 ns 
   8388608 :   33.5 ns          /    35.0 ns 
  16777216 :   39.1 ns          /    46.1 ns 
  33554432 :  107.6 ns          /   151.1 ns 
  67108864 :  159.1 ns          /   202.4 ns 

Core to Core Memory Latency

Note: According to Minisforum's documentation, "cores "0-1,10-11" are big cores; "6-9" are middle cores; "2-5" are small cores."

sbc-bench results

Run sbc-bench and paste a link to the results here: ThomasKaiser/sbc-bench#126

Phoronix Test Suite

Results from pi-general-benchmark.sh:

• pts/encode-mp3: 8.762 sec
• pts/x264 1080p: 53.43 fps
• pts/x264 4K: 12.93 fps
• pts/phpbench: 649067
• pts/build-linux-kernel (defconfig): 896.561 sec

[geerlingguy]

I tried upgrading the GPU, by installing an Nvidia A2000 discrete GPU, specifically this one, the Abovetop RTX A2000 single-slot card, which cost $251 and took a couple weeks to ship.

It fit snugly into the chassis, and the fan even takes air in through the existing vent holes in the cover.

On first boot with it in place, I didn't get any monitor output, which is not unexpected, since I figured the drivers might need to be installed.

lspci shows:

mini@ms-r1:~$ lspci
00:00.0 PCI bridge: CIX Technology Group Co., Ltd. CIX P1 CD8180 PCI Express Root Port
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. Device 8127 (rev 05)
30:00.0 PCI bridge: CIX Technology Group Co., Ltd. CIX P1 CD8180 PCI Express Root Port
31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. Device 8127 (rev 05)
60:00.0 PCI bridge: CIX Technology Group Co., Ltd. CIX P1 CD8180 PCI Express Root Port
61:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter
90:00.0 PCI bridge: CIX Technology Group Co., Ltd. CIX P1 CD8180 PCI Express Root Port
91:00.0 Non-Volatile memory controller: Kingston Technology Company, Inc. Device 502d
c0:00.0 PCI bridge: CIX Technology Group Co., Ltd. CIX P1 CD8180 PCI Express Root Port
c1:00.0 VGA compatible controller: NVIDIA Corporation GA107GLM [RTX A2000 Mobile] (rev a1)
c1:00.1 Audio device: NVIDIA Corporation GA107 High Definition Audio Controller (rev a1)

So the card is at least being seen. And indeed, nvidia-detect (sudo apt install -y nvidia-detect) sees it:

mini@ms-r1:~$ nvidia-detect 
Detected NVIDIA GPUs:
c1:00.0 VGA compatible controller [0300]: NVIDIA Corporation GA107GLM [RTX A2000 Mobile] [10de:25b8] (rev a1)

Checking card:  NVIDIA Corporation GA107GLM [RTX A2000 Mobile] (rev a1)
Your card is supported by all driver versions.
Your card is also supported by the Tesla 470 drivers series.
It is recommended to install the
    nvidia-driver
package.

I tried installing nvidia-driver, but that failed:

mini@ms-r1:~$ sudo apt install -y nvidia-driver
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 nvidia-driver : Depends: nvidia-kernel-dkms (= 535.247.01-1~deb12u1) but it is not installable or
                          nvidia-kernel-535.247.01 or
                          nvidia-open-kernel-535.247.01
E: Unable to correct problems, you have held broken packages.

So next I tried installing nvidia-tesla-470-driver... and that almost worked, but failed:

...
Processing triggers for initramfs-tools (0.142+deb12u3) ...
update-initramfs: Generating /boot/initrd.img-6.6.10-cix-build-generic
W: Possible missing firmware /lib/firmware/xc3028L-v36.fw for built-in driver xc2028
W: Possible missing firmware /lib/firmware/xc3028-v27.fw for built-in driver xc2028
W: Possible missing firmware /lib/firmware/dvb-fe-xc4000-1.4.fw for built-in driver xc4000
W: plymouth: The plugin label.so is missing, the selected theme might not work as expected.
W: plymouth: You might want to install the plymouth-themes package to fix this.
/etc/initramfs-tools/scripts/init-top/growroot.sh: line 33: 953.87 * 512 / 1000 / 1000 / 1000: syntax error: invalid arithmetic operator (error token is ".87 * 512 / 1000 / 1000 / 1000")
Errors were encountered while processing:
 nvidia-tesla-470-kernel-dkms
 nvidia-tesla-470-driver
E: Sub-process /usr/bin/dpkg returned an error code (1)

So then I tried uninstalling it with apt remove, but also got the error again. I had to uninstall the dkms kernel module first, then uninstall the driver. Then run apt autoremove.

Anywho...

[geerlingguy]

Switching gears to the Nvidia driver package download, I grabbed the latest Data Center Driver for Linux aarch64, version 580.105.08.

When I tried the installer, I got a message about CC version mismatches:

Makefile:63: WARNING: Unable to locate the compiler aarch64-none-linux-gnu-gcc from CONFIG_CC_VERSION_TEXT in the kernel configuration.
make[1]: Entering directory '/usr/src/linux-headers-6.6.10-cix-build-generic'
warning: the compiler differs from the one used to build the kernel
 The kernel was built by: aarch64-none-linux-gnu-gcc (Arm GNU Toolchain 12.3.Rel1 (Build arm-12.35)) 12.3.1 20230626
 You are using:           cc (Debian 12.2.0-14+deb12u1) 12.2.0

Warning: Compiler version check failed:

The major and minor number of the compiler used to
compile the kernel:

aarch64-none-linux-gnu-gcc (Arm GNU Toolchain 12.3.Rel1 (Build arm-12.35)) 12.3.1 20230626, GNU ld (Arm GNU Toolchain 12.3.Rel1 (Build arm-12.35)) 2.40.0.20230627

does not match the compiler used here:

cc (Debian 12.2.0-14+deb12u1) 12.2.0
Copyright (C) 2022 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.


It is recommended to set the CC environment variable
to the compiler that was used to compile the kernel.

To skip the test and silence this warning message, set
the IGNORE_CC_MISMATCH environment variable to "1".
However, mixing compiler versions between the kernel
and kernel modules can result in subtle bugs that are
difficult to diagnose.

So throwing caution to the wind, I tried setting IGNORE_CC_MISMATCH=1 and compiling again. That didn't work either.

[geerlingguy]

So I went spelunking and found that Package 'firmware-nvidia-gsp' has no installation candidate, seems to be the main issue. Maybe not available in the repos?

I edited sudo nano /etc/apt/sources.list, and added non-free-firmware to the list for the line that was there:

deb http://mirrors.ustc.edu.cn/debian bookworm main contrib non-free non-free-firmware

Then ran sudo apt update and tried again...

$ sudo apt install -y nvidia-driver firmware-misc-nonfree
...
W: plymouth: The plugin label.so is missing, the selected theme might not work as expected.
W: plymouth: You might want to install the plymouth-themes package to fix this.
/etc/initramfs-tools/scripts/init-top/growroot.sh: line 33: 953.87 * 512 / 1000 / 1000 / 1000: syntax error: invalid arithmetic operator (error token is ".87 * 512 / 1000 / 1000 / 1000")
Errors were encountered while processing:
 nvidia-kernel-dkms
 nvidia-driver
E: Sub-process /usr/bin/dpkg returned an error code (1)

It installed... but when building the kernel module, it still failed.

mini@ms-r1:~/Downloads$ sudo apt --fix-broken install
...
Setting up nvidia-kernel-dkms (535.247.01-1~deb12u1) ...
Removing old nvidia-current-535.247.01 DKMS files...
Deleting module nvidia-current-535.247.01 completely from the DKMS tree.
Loading new nvidia-current-535.247.01 DKMS files...
Building for 6.6.10-cix-build-generic
Building initial module for 6.6.10-cix-build-generic
Error! Bad return status for module build on kernel: 6.6.10-cix-build-generic (aarch64)
Consult /var/lib/dkms/nvidia-current/535.247.01/build/make.log for more information.
...

In that make.log, I have:

make[2]: *** [/usr/src/linux-headers-6.6.10-cix-build-generic/Makefile:1919: /var/lib/dkms/nvidia-current/535.247.01/build] Error 2
make[1]: *** [Makefile:235: __sub-make] Error 2
make[1]: Leaving directory '/usr/src/linux-headers-6.6.10-cix-build-generic'
make: *** [Makefile:137: modules] Error 2

[geerlingguy]

I tried re-flashing the entire OS from the download link provided by Minisforum, but that didn't seem to work, it just keep booting me back into the BIOS.

Apparently I formatted my backup drive and flashed the OS to it instead of to the NVMe SSD I thought I was flashing. Oops. Just re-tested and their image works fine, to restore the system to a clean state.

So next up, I installed my other modern 1-slot GPU, an Intel A310 ECO. Now, this card required a ReDriver when I was testing it on a Raspberry Pi, so I didn't know if it'd work on here anyway, but I booted it up, and sure enough. It's not there.

So upgrading the GPU, it should be possible, but right now I can't get it going on the default Debian 12 image.

[geerlingguy]

Switching gears, I put in a new SSD, plugged in a USB stick flashed with Ubuntu 25.10 Arm Desktop, and booted into that.

The installer splash screen froze after a while, but I heard a welcome sound through the speaker I have plugged in. That meant something was happening... so I unplugged from the built-in HDMI port, and into the Nvidia A2000 that I'd re-installed. And what do you know, I had Ubuntu's installer in 4K!

Idle power with the A2000 active hovers around 29W (versus the 17W without), so it's a substantial load on the system.

I completed the Ubuntu install, ran updates, and it loaded the Nvidia drivers automatically. I then installed CUDA:

sudo apt install -y nvidia-cuda-toolkit

And here's the info on the card:

jgeerling@ms-r1:~$ nvidia-smi
Fri Nov  7 16:55:30 2025       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.95.05              Driver Version: 580.95.05      CUDA Version: 13.0     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA RTX A2000 Laptop GPU    Off |   00000000:C1:00.0 Off |                  N/A |
|  0%   48C    P8             15W /   60W |     386MiB /   8192MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                  sdf fad     GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            1836      G   /usr/bin/gnome-shell                    331MiB |
|    0   N/A  N/A            2133      G   /usr/bin/Xwayland                         2MiB |
+-----------------------------------------------------------------------------------------+

jgeerling@ms-r1:~$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Tue_Aug_15_22:10:07_PDT_2023
Cuda compilation tools, release 12.2, V12.2.140
Build cuda_12.2.r12.2/compiler.33191640_0

vkmark returns a nice score just over 10,000:

=======================================================
    vkmark 2025.01
=======================================================
    Vendor ID:      0x10DE
    Device ID:      0x25B8
    Device Name:    NVIDIA RTX A2000 Laptop GPU
    Driver Version: 2434253120
    Device UUID:    cd2e17d6cf5750afe258140922447ac6
=======================================================
[vertex] device-local=true: FPS: 11610 FrameTime: 0.086 ms
[vertex] device-local=false: FPS: 6145 FrameTime: 0.163 ms
[texture] anisotropy=0: FPS: 11797 FrameTime: 0.085 ms
[texture] anisotropy=16: FPS: 11626 FrameTime: 0.086 ms
[shading] shading=gouraud: FPS: 11437 FrameTime: 0.087 ms
[shading] shading=blinn-phong-inf: FPS: 11393 FrameTime: 0.088 ms
[shading] shading=phong: FPS: 11220 FrameTime: 0.089 ms
[shading] shading=cel: FPS: 11251 FrameTime: 0.089 ms
[effect2d] kernel=edge: FPS: 11026 FrameTime: 0.091 ms
[effect2d] kernel=blur: FPS: 8323 FrameTime: 0.120 ms
[desktop] <default>: FPS: 9285 FrameTime: 0.108 ms
[cube] <default>: FPS: 12659 FrameTime: 0.079 ms
[clear] <default>: FPS: 8979 FrameTime: 0.111 ms
=======================================================
                                   vkmark Score: 10519
=======================================================

Running at 90W max, GravityMark gave me:

GravityMark result: 16,679

I've posted some AI benchmarks in the related issue in my ai-benchmarks repository: geerlingguy/ai-benchmarks#32

[geerlingguy]

Realizing my earlier mistake of flashing Debian 12 to my backup drive (thus erasing it) instead of the NVMe SSD I was trying in the MS-R1, I tried again, this time flashing Debian 12 to the NVMe SSD I put in the MS-R1, and not my backup drive.

That worked.

So now trying again:

$ sudo nano /etc/apt/sources.list
# ...add 'non-free-firmware' to the end of the source inside, save the change

$ sudo apt update
$ sudo apt install -y nvidia-driver firmware-misc-nonfree
...
Processing triggers for nvidia-alternative (535.247.01-1~deb12u1) ...
update-alternatives: updating alternative /usr/lib/nvidia/current because link group nvidia has changed slave links
Setting up nvidia-kernel-dkms (535.247.01-1~deb12u1) ...
Loading new nvidia-current-535.247.01 DKMS files...
Building for 6.6.10-cix-build-generic
Building initial module for 6.6.10-cix-build-generic
Error! Bad return status for module build on kernel: 6.6.10-cix-build-generic (aarch64)
Consult /var/lib/dkms/nvidia-current/535.247.01/build/make.log for more information.
dpkg: error processing package nvidia-kernel-dkms (--configure):
 installed nvidia-kernel-dkms package post-installation script subprocess returned error exit status 10
dpkg: dependency problems prevent configuration of nvidia-driver:
 nvidia-driver depends on nvidia-kernel-dkms (= 535.247.01-1~deb12u1) | nvidia-kernel-535.247.01 | nvidia-open-kernel-535.247.01; however:
  Package nvidia-kernel-dkms is not configured yet.
  Package nvidia-kernel-535.247.01 is not installed.
  Package nvidia-kernel-dkms which provides nvidia-kernel-535.247.01 is not configured yet.
  Package nvidia-open-kernel-535.247.01 is not installed.

dpkg: error processing package nvidia-driver (--configure):
 dependency problems - leaving unconfigured
Processing triggers for libc-bin (2.36-9+deb12u10) ...
Processing triggers for initramfs-tools (0.142+deb12u3) ...
update-initramfs: Generating /boot/initrd.img-6.6.10-cix-build-generic

So that install is just broken on the Debian 12 image Minisforum is shipping. I can't get a kernel module to compile in, because it seems like they may have cross-compiled their kernel, and the proper sources/compiler aren't present. It's a bit of a tangle to unsort that, much easier if you just have a clean install...

I've asked Minisforum about this and am awaiting a response. This problem will probably just go away once if you install vanilla Debian, even without other hardware support enablement.

[geerlingguy]

I also asked Minisforum about power consumption, and here's what they said:

• ASPM is currently disabled for stability (it can save about 2W of power when enabled)
• The 10G NICs (RTL8127) they used, plus the Mediatek WiFi 6E chip, account for the 2-3W difference between the MS-R1 and the Orion O6 (which doesn't have WiFi built-in and uses 5G NICs (RTL8126))
• ASPM isn't displayed in the BIOS, so to enable it (if you want to experiment with it at your own risk):
  • Mount the EFI disk
  • Edit GRUB/GRUB.CFG
  • Change pcie_aspm=off to pcie_aspm=on
  • After reboot, run echo powersave > /sys/module/pcie_aspm/parameters/policy (as root)

It sounds like, similar to Radxa, the Cix SoC has some stability issues trying to conserve power, so the current solution is to feed it a lot of power all the time.

[HeyMeco]

I know that radxa has ASPM disabled in their EDK2 sources for Orion due to stability issues and I noticed that the OrangePi 6 Plus crashes if ASPM isn’t disabled in the kernel command line (during my armbian cix-acpi image bring up). I would keep it disabled until we get the green light from Cix that ASPM is stable.

Btw I can provide you an early Armbian image with Mesa Panthor GPU support if you want to try it out (it’s no where near the performance of the Mali driver but upstream and fairly straight forward). Status is here: armbian/build#8363

[HeyMeco]

Also can confirm that vanilla Fedora 42 works with the DKMS equivalent Nvidia driver package akmod-nvidia. Should apply to all recent upstream distro‘s with arm64 support

[HeyMeco]

Also for running Intel GPU‘s there is a PCIe Bar size patch in radxa‘s edk2 fork so applying that on Minisforum‘s repo and recompiling should fix that until it is merged into cix‘s repo they provide to vendors.

[volyrique]

@geerlingguy Did you use the cortexa57 Blis configuration during the HPL run or just defaults?