After checking out the hardware with an unboxing and a teardown of the GEEKOM A5 Pro 2026 Edition mini PC, and testing it with the pre-installed Windows 11 Pro OS in the second part of the review, we will now report our experience with Ubuntu 25.10 to show how the AMD Ryzen 5 7530U mini PC performs under Linux.
The Ubuntu 25.10 review includes an overview of the system, benchmarks, storage and networking (2.5GbE and WiFi 6) performance testing, and measurements of CPU temperature under stress, fan noise levels, and power consumption of the mini PC under various workloads.
Ubuntu 25.10 installation
We decided to install Ubuntu 25.10 in dual boot configuration alongside Windows 11. To do so, we first shrank the Windows partition (C:) to roughly half its original size, before inserting a USB drive with the Ubuntu 25.10 ISO to carry on with the installation.
The installation proceeded smoothly on this mini PC, and we didn’t have to enter the BIOS after installation to change the boot priorities, like we sometimes have to do.
GEEKOM A5 Pro 2026 Edition system information on Ubuntu 25.10
Goingto Settings → About, we can confirm we have a GEEKOM A5 Pro model powered by a 12-thread AMD Ryzen 5 7530U processor with Radeon Graphics, 16GB memory, and a 1 TB drive. The system runs Ubuntu 25.10 64-bit with GNOME 46, Wayland compositor, and Linux 6.17 kernel.
We can get a few more details in the terminal:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
aey@A5-Pro-CNX:~$ cat /etc/lsb-release DISTRIB_ID=Ubuntu DISTRIB_RELEASE=25.10 DISTRIB_CODENAME=questing DISTRIB_DESCRIPTION="Ubuntu 25.10" aey@A5-Pro-CNX:~$ uname -a Linux A5-Pro-CNX 6.17.0-14-generic #14-Ubuntu SMP PREEMPT_DYNAMIC Fri Jan 9 17:01:16 UTC 2026 x86_64 GNU/Linux aey@A5-Pro-CNX:~$ free -mh total used free shared buff/cache available Mem: 14Gi 4.8Gi 6.6Gi 245Mi 3.6Gi 9.7Gi Swap: 4.0Gi 0B 4.0Gi aey@A5-Pro-CNX:~$ df -mh Filesystem Size Used Avail Use% Mounted on tmpfs 3.0G 2.4M 3.0G 1% /run /dev/nvme0n1p5 466G 14G 429G 3% / tmpfs 7.3G 93M 7.2G 2% /dev/shm efivarfs 128K 15K 109K 12% /sys/firmware/efi/efivars tmpfs 5.0M 12K 5.0M 1% /run/lock tmpfs 1.0M 0 1.0M 0% /run/credentials/systemd-journald.service tmpfs 1.0M 0 1.0M 0% /run/credentials/systemd-resolved.service tmpfs 7.3G 25M 7.3G 1% /tmp /dev/nvme0n1p1 296M 69M 228M 24% /boot/efi tmpfs 1.5G 116K 1.5G 1% /run/user/1000 |
And even more using the inxi utility:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 |
aey@A5-Pro-CNX:~$ inxi -Fc0 System: Host: A5-Pro-CNX Kernel: 6.17.0-14-generic arch: x86_64 bits: 64 Desktop: GNOME v: 49.0 Distro: Ubuntu 25.10 (Questing Quokka) Machine: Type: Desktop System: GEEKOM product: A5 Pro v: N/A serial: <superuser required> Mobo: N/A model: A5 Pro serial: <superuser required> UEFI: American Megatrends LLC. v: 2.42 date: 11/24/2025 CPU: Info: 6-core model: AMD Ryzen 5 7530U with Radeon Graphics bits: 64 type: MT MCP cache: L2: 3 MiB Speed (MHz): avg: 1096 min/max: 411/4548 cores: 1: 1096 2: 1096 3: 1096 4: 1096 5: 1096 6: 1096 7: 1096 8: 1096 9: 1096 10: 1096 11: 1096 12: 1096 Graphics: Device-1: Advanced Micro Devices [AMD/ATI] Barcelo driver: amdgpu v: kernel Display: wayland server: X.Org v: 24.1.6 with: Xwayland v: 24.1.6 compositor: gnome-shell driver: dri: radeonsi gpu: amdgpu resolution: 1: 3840x2160~60Hz 2: 6144x3456~60Hz API: EGL v: 1.5 drivers: kms_swrast,radeonsi,swrast platforms: gbm,wayland,x11,surfaceless,device API: OpenGL v: 4.6 compat-v: 4.5 vendor: amd mesa v: 25.2.8-0ubuntu0.25.10.1 renderer: AMD Radeon Graphics (radeonsi renoir ACO DRM 3.64 6.17.0-14-generic) Info: Tools: api: eglinfo,glxinfo x11: xdriinfo, xdpyinfo, xprop, xrandr Audio: Device-1: Advanced Micro Devices [AMD/ATI] Renoir/Cezanne HDMI/DP Audio driver: snd_hda_intel Device-2: Advanced Micro Devices [AMD] Audio Coprocessor driver: N/A Device-3: Advanced Micro Devices [AMD] Family 17h/19h/1ah HD Audio driver: snd_hda_intel API: ALSA v: k6.17.0-14-generic status: kernel-api Server-1: PipeWire v: 1.4.7 status: active Network: Device-1: Realtek RTL8125 2.5GbE driver: r8169 IF: enp2s0 state: down mac: 38:f7:cd:d8:a6:3d Device-2: Realtek RTL8852BE PCIe 802.11ax Wireless Network driver: rtw89_8852be IF: wlp3s0 state: up mac: 50:fe:0c:35:19:58 Bluetooth: Device-1: IMC Networks Bluetooth Radio driver: btusb type: USB Report: hciconfig ID: hci0 state: up address: 50:FE:0C:35:19:59 bt-v: 5.2 Drives: Local Storage: total: 953.87 GiB used: 13.1 GiB (1.4%) ID-1: /dev/nvme0n1 model: WPBSN4M8-1TGP size: 953.87 GiB Partition: ID-1: / size: 465.13 GiB used: 13.03 GiB (2.8%) fs: ext4 dev: /dev/nvme0n1p5 ID-2: /boot/efi size: 296 MiB used: 68.8 MiB (23.2%) fs: vfat dev: /dev/nvme0n1p1 Swap: ID-1: swap-1 type: file size: 4 GiB used: 0 KiB (0.0%) file: /swap.img Sensors: System Temperatures: cpu: 50.1 C mobo: N/A gpu: amdgpu temp: 40.0 C Fan Speeds (rpm): N/A Info: Memory: total: 16 GiB note: est. available: 14.52 GiB used: 4.79 GiB (33.0%) Processes: 362 Uptime: 41m Shell: Bash inxi: 3.3.39 |
Everything looks properly detected. An AMD Ryzen 5 7530U 6-core/12-thread processor clocked up to 4548 MHz, 16GB RAM, a 465.13 GB rootfs on the WPBSN4N8-1TGP SSD (953.97 GB), 2.5GbE using Realtek RTL8125, and Wi-Fi 6 and Bluetooth 5.2 using a Realtek RTL8852BE wireless module. The idle CPU temperature is reported to be 50.1°C, but we will test this in more detail later on.
Ubuntu 25.10 Benchmarks on GEEKOM A5 Pro 2026 Edition mini PC
We’ll start Ubuntu 25.10 benchmarks with Thomas Kaiser’s sbc-bench script:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 |
aey@A5-Pro-CNX:~/Downloads/sbc-bench-master$ sudo ./sbc-bench.sh -r [sudo: authenticate] Password: Starting to examine hardware/software for review purposes... sbc-bench v0.9.72 Installing needed tools: distro packages already installed. Done. Checking cpufreq OPP. Done. Executing tinymembench. Done. Executing RAM latency tester. Done. Executing OpenSSL benchmark. Done. Executing 7-zip benchmark. Done. Throttling test: heating up the device, 5 more minutes to wait. Done. Checking cpufreq OPP again. Done (10 minutes elapsed). Results validation: * Measured clockspeed not lower than advertised max CPU clockspeed * No swapping * Background activity (%system) OK # GEEKOM A5 Pro / Ryzen 5 7530U Tested with sbc-bench v0.9.72 on Wed, 04 Mar 2026 14:03:50 +0700. ### General information: Information courtesy of cpufetch: Name: AMD Ryzen 5 7530U with Radeon Graphics Microarchitecture: Zen 3 Technology: 7nm Max Frequency: 4.547 GHz Cores: 6 cores (12 threads) AVX: AVX,AVX2 FMA: FMA3 L1i Size: 32KB (192KB Total) L1d Size: 32KB (192KB Total) L2 Size: 512KB (3MB Total) L3 Size: 16MB Ryzen 5 7530U, Kernel: x86_64, Userland: amd64 CPU sysfs topology (clusters, cpufreq members, clockspeeds) cpufreq min max CPU cluster policy speed speed core type 0 0 0 411 4548 Zen 3 1 0 1 411 4548 Zen 3 2 0 2 411 4548 Zen 3 3 0 3 411 4548 Zen 3 4 0 4 411 4548 Zen 3 5 0 5 411 4548 Zen 3 6 0 6 411 4548 Zen 3 7 0 7 411 4548 Zen 3 8 0 8 411 4548 Zen 3 9 0 9 411 4548 Zen 3 10 0 10 411 4548 Zen 3 11 0 11 411 4548 Zen 3 14867 KB available RAM ### Policies (performance vs. idle consumption): Status of performance related policies found below /sys: /sys/module/pcie_aspm/parameters/policy: default [performance] powersave powersupersave ### Clockspeeds (idle vs. heated up): Before at 58.0°C: cpu0: OPP: 4547, Measured: 4523 After at 65.1°C: cpu0: OPP: 4547, Measured: 4504 ### Performance baseline * memcpy: 16443.7 MB/s, memchr: 33325.1 MB/s, memset: 15861.3 MB/s * 16M latency: 39.45 24.61 30.74 23.56 30.60 36.27 42.26 46.69 * 128M latency: 103.3 97.28 100.7 97.13 100.6 98.61 101.5 100.5 * 7-zip MIPS (3 consecutive runs): 38503, 35443, 35160 (36370 avg), single-threaded: 5329 * `aes-256-cbc 1029812.76k 1175425.77k 1214248.87k 1222859.43k 1225555.97k 1224955.22k` * `aes-256-cbc 1045362.31k 1177312.64k 1214937.26k 1223130.79k 1225108.14k 1225299.29k` ### PCIe and storage devices: * O2 OZ777 SD/MMC Card Reader: Speed 2.5GT/s, Width x1, driver in use: sdhci-pci, * Realtek RTL8125 2.5GbE: Speed 5GT/s, Width x1, driver in use: r8169, * Realtek RTL8852BE PCIe 802.11ax Wireless Network: Speed 2.5GT/s, Width x1, driver in use: rtw89_8852be, * AMD Barcelo: Speed 8GT/s, Width x16, driver in use: amdgpu, ASPM Disabled * AMD Renoir/Cezanne USB 3.1: Speed 8GT/s, Width x16, driver in use: xhci_hcd, ASPM Disabled * AMD Renoir/Cezanne USB 3.1: Speed 8GT/s, Width x16, driver in use: xhci_hcd, ASPM Disabled * AMD Audio Coprocessor: Speed 8GT/s, Width x16, driver in use: , ASPM Disabled * 953.9GB "WPBSN4M8-1TGP" SSD as /dev/nvme0: Speed 8GT/s (downgraded), Width x4, 0% worn out, drive temp: 42°C, ASPM Disabled ### Challenging filesystems: The following partitions are NTFS: nvme0n1p3,nvme0n1p4 -> https://tinyurl.com/mv7wvzct ### Swap configuration: * /swap.img on /dev/nvme0n1p5: 4.0G (0K used) ### Software versions: * Ubuntu 25.10 (questing) * Compiler: /usr/bin/gcc (Ubuntu 15.2.0-4ubuntu4) 15.2.0 / x86_64-linux-gnu * OpenSSL 3.5.3, built on 16 Sep 2025 (Library: OpenSSL 3.5.3 16 Sep 2025) ### Kernel info: * `/proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.17.0-14-generic root=UUID=11f8b60c-e81d-4a0e-bff6-a7fa7c3f9081 ro quiet splash crashkernel=2G-4G:320M,4G-32G:512M,32G-64G:1024M,64G-128G:2048M,128G-:4096M vt.handoff=7` * Vulnerability Spec rstack overflow: Mitigation; Safe RET * Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl * Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization * Vulnerability Tsa: Mitigation; Clear CPU buffers * Vulnerability Vmscape: Mitigation; IBPB before exit to userspace * Kernel 6.17.0-14-generic / CONFIG_HZ=1000 Waiting for the device to cool down...................................... 44.6°C^C |
The 7-zip benchmark started at 38,503 MIPS for the first run, before dropping to 35,443 MIPS for the second, and 35,160 for the third, with an average of 36,370 MIPS. This seems to indicate power limits were reached rather than thermal limits, since the CPU temperature never exceeded 75.9°C as reported in the full log. It’s interesting to note that single-core workloads result in higher CPU temperatures…
|
1 2 3 4 5 6 7 8 9 |
System health while running 7-zip single core benchmark: Time CPU load %cpu %sys %usr %nice %io %irq Temp 13:11:25: 4392MHz 1.09 1% 0% 0% 0% 0% 0% 70.4°C 13:11:30: 4400MHz 1.08 8% 0% 8% 0% 0% 0% 75.8°C 13:11:35: 4417MHz 1.07 8% 0% 8% 0% 0% 0% 75.9°C 13:11:40: 4392MHz 1.07 8% 0% 8% 0% 0% 0% 75.8°C 13:11:45: 4417MHz 1.06 8% 0% 8% 0% 0% 0% 73.8°C 13:11:50: 4392MHz 1.06 8% 0% 8% 0% 0% 0% 74.9°C |
… compared to multi-core workloads due to (I assume) power limits kicking in:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 |
System health while running 7-zip multi core benchmark: Time CPU load %cpu %sys %usr %nice %io %irq Temp 13:11:53: 3504MHz 1.13 1% 0% 1% 0% 0% 0% 74.0°C 13:12:04: 3194MHz 2.80 96% 0% 95% 0% 0% 0% 69.4°C 13:12:15: 3527MHz 3.41 90% 1% 88% 0% 0% 0% 70.5°C 13:12:26: 2995MHz 5.02 90% 1% 88% 0% 0% 0% 69.2°C 13:12:37: 2895MHz 6.39 94% 0% 93% 0% 0% 0% 67.6°C 13:12:48: 2909MHz 7.03 94% 0% 93% 0% 0% 0% 66.9°C 13:12:59: 2895MHz 8.34 89% 1% 87% 0% 0% 0% 67.5°C 13:13:10: 2887MHz 8.91 97% 0% 96% 0% 0% 0% 65.4°C 13:13:20: 2920MHz 9.46 92% 1% 91% 0% 0% 0% 68.2°C 13:13:31: 2945MHz 9.85 90% 1% 88% 0% 0% 0% 69.4°C |
Talking about power limits, we installed the ryzenadj script to check those:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
aey@A5-Pro-CNX:~/RyzenAdj/build$ sudo ./ryzenadj -i no compatible ryzen_smu kernel module found, fallback to /dev/mem CPU Family: Cezanne SMU BIOS Interface Version: 18 Version: v0.18.0 PM Table Version: 400005 | Name | Value | Parameter | |---------------------|-----------|--------------------| | STAPM LIMIT | 15.001 | stapm-limit | | STAPM VALUE | 4.194 | | | PPT LIMIT FAST | 25.001 | fast-limit | | PPT VALUE FAST | 10.409 | | | PPT LIMIT SLOW | 20.001 | slow-limit | | PPT VALUE SLOW | 3.824 | | | StapmTimeConst | 60.000 | stapm-time | | SlowPPTTimeConst | 5.000 | slow-time | | PPT LIMIT APU | 20.001 | apu-slow-limit | | PPT VALUE APU | 3.824 | | | TDC LIMIT VDD | 44.000 | vrm-current | | TDC VALUE VDD | 0.576 | | | TDC LIMIT SOC | 13.000 | vrmsoc-current | | TDC VALUE SOC | 0.835 | | | EDC LIMIT VDD | 95.000 | vrmmax-current | | EDC VALUE VDD | 95.000 | | | EDC LIMIT SOC | 17.000 | vrmsocmax-current | | EDC VALUE SOC | 0.000 | | | THM LIMIT CORE | 100.000 | tctl-temp | | THM VALUE CORE | 46.849 | | | STT LIMIT APU | 0.000 | apu-skin-temp | | STT VALUE APU | 0.000 | | | STT LIMIT dGPU | 0.000 | dgpu-skin-temp | | STT VALUE dGPU | 0.000 | | | CCLK Boost SETPOINT | 50.000 | power-saving / | | CCLK BUSY VALUE | 40.199 | max-performance | |
Here are the key values:
- STAPM LIMIT – 15 Watts (TDP)
- PPT LIMIT FAST – 25 Watts, equivalent to MTP/PL2
- PPT LIMIT SLOW – 20 Watts, equivalent to PBP/PL1
For reference, the power limits in Windows were the same at 20W (PBP/PL1) and 25W (MTP/PL2).
Geekbench 6.6.0 was used to test the single-core and multi-core performance of the AMD Ryzen 5 7530U CPU.
The computer achieved a single-core score of 1,980 points and a multi-core score of 7,024 points.
The GPU was tested with Unigine Heaven Benchmark 4.0. The AMD Ryzen 5 7530U mini PC could render the benchmark at 27.2 FPS and a score of 686 points at a standard 1920×1080 resolution.
Time to check YouTube 4K and 8K video playback in Firefox and Chrome web browsers.
A YouTube 4K 60FPS video could play smoothly in Firefox, albeit with some dropped frames: 744 out of 17,761.
It was easier at 4K 30FPS with almost no dropped frames (only 3 over 5 minutes).
Google Chrome performed a little better with the YouTube 4K 60FPS video since there were 82 frames dropped out of 17,533. It’s still hard to visually tell the difference between Firefox and Chrome.
When switching to YouTube 8K 30 FPS, both Firefox (350 frames dropped out of 749)…
… and Chrome (136 frames dropped out of 988) struggled, and the video was choppy and not watchable.
It’s the same experience as in Windows, and means the GEEKOM A5 Pro 2026 Edition mini PC works well with YouTube 4K videos, but is not suitable for 8K YouTube video streaming, even at 30 FPS.
We’ll now evaluate web browsing performance with Speedometer 2.0 on Firefox. The mini PC managed 352 runs per minute.
Since Speedometer 2.0 is deprecated, we repeated the test with Speedometer 3.0: 22.8 points.
GEEKOM A5 Pro 2026 Edition Ubuntu 25.10 benchmark results compared to other mid-range mini PCs
Now that we have some benchmark results on Ubuntu 25.10 for the GEEKOM A5 Pro 2026 Edition, we’ll compare it to other mid-range mini PCs we tested with Linux (Ubuntu 22.04/24.04), namely the Beelink EQi13 Pro (Intel Core i5-13500H), GEEKOM A6 (AMD Ryzen 7 5800H), and GEEKOM A5 (AMD Ryzen 7 5800H).
Here’s a summary of the key specifications for the four systems.
| GEEKOM A5 Pro 2026 Edition | Beelink EQi13 Pro | GEEKOM A5 | GEEKOM A6 | |
|---|---|---|---|---|
| SoC | AMD Ryzen 5 7530U | Intel Core i5-13500H | AMD Ryzen 7 5800H | AMD Ryzen 7 6800H |
| CPU | 6 cores, 12 threads up to 4.5GHz | 12 cores, 16 threads, up to 4.70 GHz | 8 cores, 16 threads, up to 4.4 GHz | 8 cores, 16 threads, up to 4.7 GHz |
| GPU | AMD Radeon Vega Graphics | 80 EU Intel Iris Xe Graphics | AMD Radeon Vega 8 Graphics | AMD Radeon 680M Graphics |
| Memory | 16GB DDR4-3200 | 32GB DDR4-3200 | 32GB DDR4-3200 | 32GB DDR5-5600 |
| Storage | 1TB M.2 SSD | 1TB NVMe SSD | 512GB NVMe SSD | 1TB NVMe SSD |
| Default OS | Windows 11 Pro | Windows 11 Pro | Windows 11 Pro | Windows 11 Pro |
| Price (March 1, 2026 ) | $549 | $629 | $529 | $649 |
All four devices are in a similar $550 – $650 price range. The GEEKOM A5 Pro 2026 Edition features a hexa-core processor, so it’s expected to struggle in multi-core benchmarks against octa-core and 12-core processors. Let’s see if this is confirmed by the benchmark results.
| GEEKOM A5 Pro 2026 Edition | Beelink EQi13 Pro | GEEKOM A5 | GEEKOM A6 | |
|---|---|---|---|---|
| sbc-bench.sh | ||||
| - memcpy | 16,443.7 MB/s | 19,332.7 MB/s | 18,717 MB/s | 19,761.4 MB/s |
| - memset | 15,861.3 MB/s | 21,189.4 MB/s | 43,837 MB/s | 18,944.4 MB/s |
| - 7-zip (average) | 36,370 | 49,150 | 53,610 | 58,990 |
| - 7-zip (top result) | 38,503 | 49,237 | 54,850 | 59,599 |
| - OpenSSL AES-256 16K | 1,224,955.22k | 1,663,926.27k | 1,202,869.59k | 1,271,949.99k |
| Geekbench 6 Single | 1,980 | 2,545 | 2,002 | 2,111 |
| Geekbench 6 Multi | 7,024 | 9,625 | 9,347 | 10,573 |
| Unigine Heaven score | 686 | 920 | 890 | 1,698 |
| Speedometer 2.0 (Firefox) | 352 | 329 | 218 | 262 |
Multi-core benchmarks (7-Zip and GeekBench 6.6 multi-core) are indeed impacted, but we can also see weaknesses in terms of memory bandwidth and 3D graphics performance. The single-core performance of the GEEKOM A5 Pro 2026 Edition is similar to that of GEEKOM A5, and the only real positive is the improved web browsing score with Speedometer 2.0, but that could be due to web browser performance improvements over time rather than the hardware. Our conclusion at this stage is the same as in Windows 11: “the earlier GEEKOM A5 looks like a better deal if you can live with a smaller capacity 512GB NVMe SSD”.
Storage and USB performance
We tested the 1TB NVMe SSD using iozone3:
|
1 2 3 4 5 6 7 8 9 10 11 |
aey@A5-Pro-CNX:~$ sudo iozone -e -I -a -s 1000M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 1024000 4 146209 196101 262765 246238 77908 152699 1024000 16 493950 540077 759877 749838 260197 500503 1024000 512 2616002 2711260 2705624 2706009 2269644 2495291 1024000 1024 2855444 2903498 2966680 2971362 2703430 2664494 1024000 16384 3272670 3196435 3520948 3545959 3504452 3287415 iozone test complete. |
Results were 3,520 MB/s for sequential reads and 3,272MB/s for sequential writes, which compares to 3,720 MB/s and 3,434 MB/s, respectively, using CrystalDiskMark on Windows 11.
We also tested the USB 3.2 Gen 2 Type-A/C ports on the mini PC with an ORICO M234C3-U4 M.2 NVMe SSD enclosure and the single USB 2.0 Type-A port with a Seagate USB HDD, as well as lsusb and iozone utilities.
For reference, here are the results for the front left USB 3.2 Type-A ports:
|
1 2 3 4 5 6 7 8 9 |
aey@A5-Pro-CNX:~$ lsusb -t | grep uas |__ Port 002: Dev 003, If 0, Class=Mass Storage, Driver=uas, 10000M aey@A5-Pro-CNX:/media/aey/TB3-EXT4$ sudo iozone -e -I -a -s 1000M -r 16384k -i 0 -i 1 random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 1024000 16384 771076 781302 870208 872782 iozone test complete. |
and the USB 2.0 port:
|
1 2 3 4 5 6 7 8 9 |
aey@A5-Pro-CNX:~$ lsusb -t | grep uas |__ Port 003: Dev 005, If 0, Class=Mass Storage, Driver=uas, 480M aey@A5-Pro-CNX:/media/aey/USB3_EXT4$ sudo iozone -e -I -a -s 1000M -r 16384k -i 0 -i 1 random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 1024000 16384 31763 31236 42185 39833 iozone test complete. |
Summary for all six ports from left to right:
- Front panel
- USB-A #1 – USB 3.2 – 10 Gbps – Read speed: 870 MB/s; write speed: 771 MB/s
- USB-A #2 – USB 3.2 – 10 Gbps – Read speed: 866 MB/s; write speed: 773 MB/s
- Rear panel
- USB-C #1 – USB 3.2 – 10 Gbps – Read speed: 900 MB/s; write speed: 798 MB/s
- USB-A #1 (top) – USB 3.2 – 10 Gbps – Read speed: 910 MB/s; write speed: 794 MB/s
- USB-A #2 (bottom) – USB 3.0 – 480 Mbps – Read speed: 42 MB/s; write speed: 31 MB/s
- USB-C #2 – USB 3.2 – 10 Gbps – Read speed: 911 MB/s; write speed: 792 MB/s
The results are mostly as expected/advertised. It looks like the USB 3.2 Type-A ports on the front panels are slightly slower than the ones on the rear panel.
2.5GbE, WiFi 6, and Bluetooth 5.2
We’ll now test 2.5GbE and 5GHz WiFi 6 networking performance on the GEEKOM A5 Pro 2026 Edition running Ubuntu 25.10 using iperf3. We’ll start with the 2.5GbE port using an UP Xtreme i11 Edge mini PC on the other side.
- Download
1234567891011121314151617aey@A5-Pro-CNX:~$ iperf3 -t 60 -c 192.168.31.12 -i 10 -RConnecting to host 192.168.31.12, port 5201Reverse mode, remote host 192.168.31.12 is sending[ 5] local 192.168.31.227 port 47890 connected to 192.168.31.12 port 5201[ ID] Interval Transfer Bitrate[ 5] 0.00-10.01 sec 2.74 GBytes 2.35 Gbits/sec[ 5] 10.01-20.01 sec 2.74 GBytes 2.35 Gbits/sec[ 5] 20.01-30.01 sec 2.74 GBytes 2.35 Gbits/sec[ 5] 30.01-40.01 sec 2.74 GBytes 2.35 Gbits/sec[ 5] 40.01-50.01 sec 2.74 GBytes 2.35 Gbits/sec[ 5] 50.01-60.01 sec 2.74 GBytes 2.35 Gbits/sec- - - - - - - - - - - - - - - - - - - - - - - - -[ ID] Interval Transfer Bitrate Retr[ 5] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec 0 sender[ 5] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec receiveriperf Done. - Upload
12345678910111213141516aey@A5-Pro-CNX:~$ iperf3 -t 60 -c 192.168.31.12 -i 10Connecting to host 192.168.31.12, port 5201[ 5] local 192.168.31.227 port 54312 connected to 192.168.31.12 port 5201[ ID] Interval Transfer Bitrate Retr Cwnd[ 5] 0.00-10.01 sec 2.74 GBytes 2.36 Gbits/sec 0 932 KBytes[ 5] 10.01-20.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.36 MBytes[ 5] 20.01-30.01 sec 2.69 GBytes 2.31 Gbits/sec 0 1.36 MBytes[ 5] 30.01-40.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.36 MBytes[ 5] 40.01-50.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.36 MBytes[ 5] 50.01-60.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.36 MBytes- - - - - - - - - - - - - - - - - - - - - - - - -[ ID] Interval Transfer Bitrate Retr[ 5] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec 0 sender[ 5] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec receiveriperf Done. - Full-duplex (bidrectional)
12345678910111213141516171819202122232425aey@A5-Pro-CNX:~$ iperf3 -t 60 -c 192.168.31.12 -i 10 --bidirConnecting to host 192.168.31.12, port 5201[ 5] local 192.168.31.227 port 34916 connected to 192.168.31.12 port 5201[ 7] local 192.168.31.227 port 34930 connected to 192.168.31.12 port 5201[ ID][Role] Interval Transfer Bitrate Retr Cwnd[ 5][TX-C] 0.00-10.01 sec 2.74 GBytes 2.35 Gbits/sec 0 805 KBytes[ 7][RX-C] 0.00-10.01 sec 2.74 GBytes 2.35 Gbits/sec[ 5][TX-C] 10.01-20.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.12 MBytes[ 7][RX-C] 10.01-20.01 sec 2.73 GBytes 2.35 Gbits/sec[ 5][TX-C] 20.01-30.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.12 MBytes[ 7][RX-C] 20.01-30.01 sec 2.73 GBytes 2.35 Gbits/sec[ 5][TX-C] 30.01-40.01 sec 2.73 GBytes 2.35 Gbits/sec 0 1.69 MBytes[ 7][RX-C] 30.01-40.01 sec 2.73 GBytes 2.35 Gbits/sec[ 5][TX-C] 40.01-50.01 sec 2.73 GBytes 2.35 Gbits/sec 0 1.69 MBytes[ 7][RX-C] 40.01-50.01 sec 2.73 GBytes 2.35 Gbits/sec[ 5][TX-C] 50.01-60.01 sec 2.74 GBytes 2.35 Gbits/sec 0 1.69 MBytes[ 7][RX-C] 50.01-60.01 sec 2.73 GBytes 2.35 Gbits/sec- - - - - - - - - - - - - - - - - - - - - - - - -[ ID][Role] Interval Transfer Bitrate Retr[ 5][TX-C] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec 0 sender[ 5][TX-C] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec receiver[ 7][RX-C] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec 12 sender[ 7][RX-C] 0.00-60.01 sec 16.4 GBytes 2.35 Gbits/sec receiveriperf Done.
The results are perfect at 2.35 Gbps in all conditions.
Let’s now switch to 5 GHz WiFi 6 testing, adding a Xiaomi Mi Router AX6000 router to the mix.
- Download
1234567891011121314151617aey@A5-Pro-CNX:~$ iperf3 -t 60 -c 192.168.31.12 -i 10 -RConnecting to host 192.168.31.12, port 5201Reverse mode, remote host 192.168.31.12 is sending[ 5] local 192.168.31.111 port 53898 connected to 192.168.31.12 port 5201[ ID] Interval Transfer Bitrate[ 5] 0.00-10.01 sec 970 MBytes 812 Mbits/sec[ 5] 10.01-20.01 sec 972 MBytes 815 Mbits/sec[ 5] 20.01-30.01 sec 966 MBytes 811 Mbits/sec[ 5] 30.01-40.01 sec 935 MBytes 785 Mbits/sec[ 5] 40.01-50.01 sec 974 MBytes 817 Mbits/sec[ 5] 50.01-60.01 sec 1006 MBytes 844 Mbits/sec- - - - - - - - - - - - - - - - - - - - - - - - -[ ID] Interval Transfer Bitrate Retr[ 5] 0.00-60.02 sec 5.69 GBytes 814 Mbits/sec 2 sender[ 5] 0.00-60.01 sec 5.69 GBytes 814 Mbits/sec receiveriperf Done. - Upload
12345678910111213141516aey@A5-Pro-CNX:~$ iperf3 -t 60 -c 192.168.31.12 -i 10Connecting to host 192.168.31.12, port 5201[ 5] local 192.168.31.111 port 41836 connected to 192.168.31.12 port 5201[ ID] Interval Transfer Bitrate Retr Cwnd[ 5] 0.00-10.01 sec 546 MBytes 457 Mbits/sec 1 2.32 MBytes[ 5] 10.01-20.01 sec 571 MBytes 479 Mbits/sec 0 2.44 MBytes[ 5] 20.01-30.01 sec 570 MBytes 478 Mbits/sec 0 3.93 MBytes[ 5] 30.01-40.01 sec 574 MBytes 481 Mbits/sec 1 3.93 MBytes[ 5] 40.01-50.01 sec 564 MBytes 473 Mbits/sec 0 3.93 MBytes[ 5] 50.01-60.01 sec 572 MBytes 480 Mbits/sec 0 3.93 MBytes- - - - - - - - - - - - - - - - - - - - - - - - -[ ID] Interval Transfer Bitrate Retr[ 5] 0.00-60.01 sec 3.32 GBytes 475 Mbits/sec 2 sender[ 5] 0.00-60.03 sec 3.31 GBytes 474 Mbits/sec receiveriperf Done.
814 Mbps for downloads and 474 Mbps for uploads is similar to what we got in Windows. Not outstanding, but it should do for most people, especially considering this is a mid-range mini PC.
We also quickly tested Bluetooth 5.2 by successfully transferring a file between the GEEKOM A5 Pro and a Redmi Note 14 Pro 5G smartphone.
Thermal performance of the GEEKOM A5 Pro 2026 Edition
In order to evaluate the thermal performance/cooling efficiency of the GEEKOM A5 Pro 2026 Edition mini PC, we ran a stress test on the 6-core/12-thread AMD Ryzen 5 7530U SoC while monitoring CPU temperature and frequency with Psensor temperature monitor and sbc-bench.sh script.
The test results show that the CPU maintains a consistent clock speed of around 2.6 GHz under full load, with the CPU temperature remaining in a 64–66°C range in a room at about 28°C. There’s plenty of legroom, and people wanting to extract a bit more performance may consider increasing PL2, although this should increase power consumption.
Fan noise
Since the GEEKOM A5 Pro 2026 Edition is one of the quieter actively cooled mini PCs I’ve tested. We measured the fan noise with a sound level meter placed around 5 centimeters from the top of the device:
- Idle – 37.7 – 38 dBA (fan off)
- Stress test on all 12 threads – 47.7 – 48.1 dBA for the initial ~10 seconds, after which the noise level drops and stabilizes at around 39.3 – 40.5 dBA for ~30 seconds, repeating in cycles.
For reference, the meter measures around 36.6 – 37.0 dBA in a quiet room. During typical everyday use, the mini PC’s fan is practically inaudible.
GEEKOM A5 Pro’s power consumption on Ubuntu 25.10
We finally measured the power consumption with a wall power meter:
- Power off – 0.7 Watts
- Idle – 3.8 – 4.4 Watts
- Video playback – 28.0 – 29.1 Watts (YouTube 4K 60FPS in Firefox)
- CPU stress test (stress -c 12)
- First few seconds – 35.9 – 36.7 Watts
- Longer runs – 24.0 – 24.3 Watts
Note: During the testing, the mini PC was connected to Wi-Fi 6, a USB RF dongle for a wireless keyboard and mouse combo, and a GAOMON PD2200 Pen Display 21.5-inch drawing tablet via one of the HDMI ports.
Conclusion
The GEEKOM A5 Pro 2026 Edition mini PC worked reasonably well on Ubuntu 25.10, and we didn’t encounter any major issues. In other words, it performs about the same in Linux as in Windows 11 Pro, supports multi-monitor setups up to 4 displays via HDMI and USB-C (DisplayPort) ports, handles YouTube 4K video playback well up to 60 FPS, and delivers good network performance through 2.5GbE and WiFi 6.
The mini PC never throttles due to excessive temperature, consumes less power than most other mini PCs we’ve reviewed, and the fan is pretty quiet, even under load. Coolness and quietness are nice, but the mini PC’s performance is on the low side at this price point, especially for multi-core workloads (partially due to the 6-core/12-thread CPU), 3D graphics performance, and memory bandwidth. The system can not handle YouTube 8K videos either, even at 30 FPS, but 4Kp60 YouTube videos play just fine.
The summary for Linux is the same as in Windows. The GEEKOM A5 Pro 2026 Edition is suitable for people wanting a fairly quiet mini PC (bearing in mind it’s not fanless) and care about power consumption and want to avoid thermal throttling even in hotter climates. If those metrics are not important, and you prefer higher performance at a similar price, the GEEKOM A5 looks to be a better option, and we’ve previously seen that it also works well on Ubuntu. (We had an issue with WiFi in Ubuntu 22.04, but it should now be resolved with a more recent Linux kernel).
We’d like to thank GEEKOM for sending the A5 Pro 2026 Edition mini PC for review. It can be purchased for $569 on the GEEKOM US store, or $549 on Amazon ,or 619 Euros on the GEEKOM DE store. You can also use CNXA5PRO coupon to get a 7% discount in the US stores.
CNXSoft: This article is a translation – with a few additional insights – of the original review on CNX Software Thailand by Suthinee Kerdkaew.
Jean-Luc started CNX Software in 2010 as a part-time endeavor, before quitting his job as a software engineering manager, and starting to write daily news, and reviews full time later in 2011.
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress. We also use affiliate links in articles to earn commissions if you make a purchase after clicking on those links.
