12gen N-series Nas motherboard (topton, cwwk, ... )
- Thread starter Steffe
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Not better. Couldn't get mine (n305) to boot with C state lower than 6 nor with the power saving option for the gpu enabled. For some reason turbostat still reports it being in c7 most of the time and OS requests c10.
Last edited:
I feel like the CPU could be entering the C-states just fine as there's a service I use for streaming (MediaMTX) that starts dropping UDP packets when the CPU enters C8. This happens on the ODROID H4, and also in the BKHD board, which makes me think the CPU is indeed entering the states. Why it doesn't affect the power draw much, I have no idea.
Thanks. Do you have states for the Package C-State? It does seem that Cores have no issue entering C-states but it's the package that doesn't. Also do you have a power meter that would let you measure from the wall?
No package C-State are reported.
Measurement at the wall is 20-23W with 3 active HDD's+2 SSD BUT the PSU is a 500W gold power supply, i doubt it's able to achieve lower power draw ( peak is 35W or so).
S-tui report package power at "idle" around 2W and 10.9W in stress. System is never fully idle as i run around 40 dockers on it.
P.S. never had problems booting it and never had to "tune" C-States in bios. I rune clean debian ( with openmediavault on top) and works perfectly, never tried proxmox where i see most people have problems going lower than C6
Last edited:
I have a hard time to decide between this one and a CWWK AMD-8845HS - since it will be a simple roon / logitech media server i dont need that many powerfull cores but would be nice to have ecc for zfs.
does anyone had success with ecc ram on 8845 motherboard? i am afraid that it was only supported on 7735hs as per amd docs: https://www.amd.com/en/products/processors/laptop/ryzen/8000-series/amd-ryzen-7-8845hs.html
it had a link to the other mobo i am considering: CWWK eight-slot 10G NAS motherboard N150/N305/N355 / dual 2.5G network
so the conclusion is that both of them does not have ecc.
Just another update: I tried the BKHD router and with ASPM enabled at idle it uses 6W (!). It doesn't have SATA ports so it doesn't have a JMB585 on board. Pretty impressive, sadly I need SFP and adding the SFP card (X520) it jumps up to 11W. This makes me think that the NAS BKHD motherboard *should* also use around 6W with all PCIe devices disabled, we must be missing something. Maybe just having the JMB585 on the board makes it draw power, even though it's disabled? This also happens with the X520 - it uses power even with the root port disabled. The only way to make the router go down to 6W is to physically unplug the card.
Last edited:
I think you're seeing two effects making the wattage worse. 1) Both the JMB and the older ethernet chips are well known for drinking a lot of power and ignoring most attempts to get them to power down. So they are thirsty in use and don't really go to sleep. 2) both chipsets are capable of hot plugging and things like wake-on-lan. This keeps them in a higher activity state than we might hope - potentially, even in a board that doesn't support the function! Also support for those spontaneous events appears to keep the motherboard chipsets active even when embedded in the cpu.
So you did well to find even a minimal setup where the motherboard would enter a true idle state. It also suggests that the motherboards' underlying power handling isn't as bad as we feared. The problem is the combination of parts and settings leaves everything more thirsty than we'd like.
One reason that cwwk were nagged into making the purple board was to get a chance at low power sata idle using the ASM1166 chip.
Just got a Topton N150 green board (actual label: BKHD-1264-NAS25030388) as I figure it's a little hard to justify paying double for the newer purple boards. OMV installed and running just fine, only a few annoyances I've noticed with this board so far:
1. System fan connector (4 pin) seems to only run the fan at full speed. It's a bit annoying as I have the Jonsbo N2 case of which the 120mm is very loud at full speed. Doesn't appear to be a fix for this so I guess I'll have to do a DIY resistor mod.
2. The motherboard always powers on when power is applied. The BIOS setting for power loss = stay off, but that doesn't seem to do anything.
1. System fan connector (4 pin) seems to only run the fan at full speed. It's a bit annoying as I have the Jonsbo N2 case of which the 120mm is very loud at full speed. Doesn't appear to be a fix for this so I guess I'll have to do a DIY resistor mod.
2. The motherboard always powers on when power is applied. The BIOS setting for power loss = stay off, but that doesn't seem to do anything.
Last edited:
I bought this motherboard, it does not support ASPM (due to the 10G chipset firmware).
Hi everyone,
I'm encountering an issue with Unraid where the system fails to boot if CPU C-states are enabled in the BIOS. When this happens, I get USB-related errors during boot, and Unraid doesn't start properly.
After some testing and research, I discovered that the system only boots correctly when a USB keyboard is physically connected. With a USB keyboard plugged in, everything works fine—even with C-states enabled.
I came across a suggestion on the Unraid forum that switching the BIOS from UEFI mode to Legacy (i.e., selecting USB:device instead of UEFI:device) can solve this problem.
Has anyone with a CWWK motherboard tried this? Do these boards support legacy boot mode in the BIOS?
I'm encountering an issue with Unraid where the system fails to boot if CPU C-states are enabled in the BIOS. When this happens, I get USB-related errors during boot, and Unraid doesn't start properly.
After some testing and research, I discovered that the system only boots correctly when a USB keyboard is physically connected. With a USB keyboard plugged in, everything works fine—even with C-states enabled.
I came across a suggestion on the Unraid forum that switching the BIOS from UEFI mode to Legacy (i.e., selecting USB:device instead of UEFI:device) can solve this problem.
Has anyone with a CWWK motherboard tried this? Do these boards support legacy boot mode in the BIOS?
I noticed the system fan thing as well. But I noticed that on Windows I could control it using the LibreHardwareMonitor software, so there must be a software way to do it on Linux. I couldn't find it exposed in the /sys filesystem but there might be a module that has to be loaded first to control it through PWM.
Do we have the same board? I have the BKHD-1264NAS and it has 2 fans, a CPU_FAN and SYS_FAN. In my case the CPU_FAN is bios controlled and the SYS_FAN can be software controlled on Windows, but not on Linux.