ARM CPU Stress Test
#11
(08-31-2018, 08:35 AM)jzhang18 Wrote: sudo ./sbc-bench.sh -c  
result - http://ix.io/1lC4
sudo ./sbc-bench.sh NEON
result - http://ix.io/1lD5
Not sure there is much difference or not.

It seems the 0.6 version you grabbed is pretty deaf to parameters, neither of the results you post are running cpuminer! Try delete your copy of sbc-bench and re-run the wget. Parameter should be either neon (lower case, no -) or -c

@Luke - yeah but confused why he fits a shim - will ask him next time I see him on IRC! My heatsink fits fine without a shim.
  • ROCKPro64 v2.1 2GB, 16Gb eMMC for rootfs, SX8200Pro 512GB NVMe for /home, HDMI video & sound, Bluetooth keyboard & mouse. Arch (6.2 kernel, Openbox desktop) for general purpose daily PC.
  • PinePhone Pro Explorer Edition, daily driver, rk2aw & U-boot on SPI, Arch/SXMO & Arch/phosh on eMMC
  • PinePhone BraveHeart now v1.2b 3/32Gb, Tow-boot with Arch/SXMO on eMMC
#12
sudo /bin/bash ./sbc-bench.sh -T 70
Run 4 hours, 6 cores 100%, max temp: 62 , never reach 70

ls /usr/local/src/
================
cpuminer-multi
mhz
tinymembench
================

Noctua A-Series Cooling Fan Blades with AAO Frame, SSO2 Bearing (NF-A4X10-FLX 5V) on my PINE64 heatsink


I set CPU Governor to "performance" during boot
====================================
Hostname: rockpro64
CPU Frequency:
Core0: 1416Mhz
Core1: 1416Mhz
Core2: 1416Mhz
Core3: 1416Mhz
Core4: 1800Mhz
Core5: 1800Mhz
CPU Governor: performance
Temperature: 41°C
#13
(08-31-2018, 12:03 PM)dukla2000 Wrote:
(08-31-2018, 08:35 AM)jzhang18 Wrote: sudo ./sbc-bench.sh -c  
result - http://ix.io/1lC4
sudo ./sbc-bench.sh NEON
result - http://ix.io/1lD5
Not sure there is much difference or not.

It seems the 0.6 version you grabbed is pretty deaf to parameters, neither of the results you post are running cpuminer! Try delete your copy of sbc-bench and re-run the wget. Parameter should be either neon (lower case, no -) or -c

@Luke - yeah but confused why he fits a shim - will ask him next time I see him on IRC! My heatsink fits fine without a shim.

Well because copper + thermal paste = better thermal efficiency than the thermal pad. But I think the performance is good enough for most applications with the thermal pad anyways.
You can find me on IRC, Discord and Twitter


#14
(08-31-2018, 12:31 PM)Luke Wrote:
(08-31-2018, 12:03 PM)dukla2000 Wrote: @Luke - yeah but confused why he fits a shim - will ask him next time I see him on IRC! My heatsink fits fine without a shim.

Well because copper + thermal paste = better thermal efficiency than the thermal pad. But I think the performance is good enough for most applications with the thermal pad anyways.

Yeah, but with a shim you have 2 junctions between the chip heatspreader and the heatsink. Which is unnecesary and thermally inefficient (compared to just 1 if you put the heatsink straight on the heatspreader. Of course thermal pads tend to be inefficient compared to thermal paste ...)
  • ROCKPro64 v2.1 2GB, 16Gb eMMC for rootfs, SX8200Pro 512GB NVMe for /home, HDMI video & sound, Bluetooth keyboard & mouse. Arch (6.2 kernel, Openbox desktop) for general purpose daily PC.
  • PinePhone Pro Explorer Edition, daily driver, rk2aw & U-boot on SPI, Arch/SXMO & Arch/phosh on eMMC
  • PinePhone BraveHeart now v1.2b 3/32Gb, Tow-boot with Arch/SXMO on eMMC
#15
To my experience - thermal pads on PIN64 heatsink is efficient enough to pass the heat from CPU and RAM.
The more important thing is to keep the heatsink temperature low with a good cooling fan.


Possibly Related Threads…
Thread Author Replies Views Last Post
  Call for test: Clonezilla live for Rockpro64 stevenshiau 1 4,680 04-17-2020, 12:38 AM
Last Post: tllim
  Gentoo, Current state, Test Phase khaosgrille 12 14,929 06-23-2019, 07:14 PM
Last Post: necrose99

Forum Jump:


Users browsing this thread: 1 Guest(s)