Question Overclocking the GPU for GPD MicroPC and Chuwi Hi10X (Intel Celeron N4100/N4120)
Question Overclocking the GPU for GPD MicroPC and Chuwi Hi10X (Intel Celeron N4100/N4120)
Hello. I can assist with adjusting the GPU settings for GPD MicroPC and Chuwi Hi10X (Intel Celeron N4100 and N4120). Intel has set a power cap of 8.5 watts per GPU.
Google Translate is used here.
The conversion details are as follows: 8.5 watts equals 700 MHz.
Comparison of GPD MicroPC and GPD Win 2 performance
Is the GPD MicroPC really half as strong as GPD Win 2? Why? Given that both CPUs have a maximum TDP of 15 watts, are they built on the same 14 nm process, and were released just one year apart?
In reality, the Intel Celeron N4100 is equivalent to the Intel Core m3-7Y30, but it reduces the video card performance and prevents undervolting. These chips show nearly identical CPU scores in Cinebench, yet differ significantly in Cinebench OpenGL.
Methods to circumvent power restrictions
Techniques for bypassing GPU power limits
My overclocking commands:
Rw.exe /Min /Nologo /Stdout /Command="W16 0xFED170A8 0000"
Rw.exe /Min /Nologo /Stdout /Command="W16 0xFED170AC 0000"
Rw.exe /Min /Nologo /Stdout /Command="WRMSR 0x610 0x0 0x00000000"
I attempted adjustments from various sources, but only managed to lift the overall processor TDP cap. When trying to bypass the GPU restriction, a blue screen appeared (tried MSR 640 and 638).
GPD Win 2 features a GPU that has been overclocked to 14 watts in FurMark, while its CPU also reaches 14 watts during the AIDA64 stress test.
The GPD MicroPC’s CPU can be overclocked up to 14 watts, but its GPU consistently stays below 8.5 watts regardless of attempts.
I tried ThrottleStop, but it only allowed me to set the processor TDP limit; the GPU remained constrained and underperformed.
Intel XTU is incompatible with Celeron processors.
How can this issue be resolved? Please advise if anyone understands. I aim to surpass the GPD Win 2 performance and remove these restrictions.
I also tested using the IGPU=X option in the ThrottleStop configuration file, but it had no effect.
I evaluated GPU load by running it with zero CPU activity, one at a time.
Hello once more. I searched extensively and discovered two individuals sharing similar goals; one of them shared quite valuable insights. However, I still find myself confused about the details.
I also need assistance.
Because of the increased costs for video cards and electronics, I can't afford a high-performance PC. Additionally, finding GPD MicroPC models in such a compact form with numerous USB ports and a touchpad at an affordable price is difficult. All alternatives are priced above $1000 and significantly larger in size.
Currently, I don't know the correct base address for offset 8068. In the Intel Celeron N4100, only FED170 addresses are used (here TDP and DPTF limits apply). I examined all the bits in this range but didn't find anything.
I added 8068 to several addresses and observed the outcomes:
FED00000 – all surrounding bits are set to FF (empty)
FED08068 – same as above
FED10000 – all surrounding bits are FF (empty)
FED18068 – after adding 8068, the surrounding area remains empty
FED15000 – surrounding bits remain empty
FED1D068 – surrounding bits remain empty
FED17000 – checked this address; adjacent lines have different structures and don't match
FED1F068 – surrounding bits are empty
FED17000 + 8068 – also empty surrounding bits
FEE001B8 – all surrounding bits are empty
FEE08220 – after adding 8068, the area stays empty
Also, I reviewed all addresses from FED00000 to FF300000. I changed the Intel Graphics Power Plan setting and observed whether the bit changed. No changes were detected across any bit that wasn't FF or 00.
This suggests the Intel Graphics Power Plan might be controlling the limit I'm seeking. Or it could be in the MSR area. How can I test all MSR registers without triggering a BSOD? A BSOD occurs when I attempt to read an unused MSR.
I've shared all the details and tests in my previous correspondence.
If you're an expert who can assist, please let me know here and I'll add you to a private chat on this site. Multiple participants can join.
I'm always willing to compensate for the time invested.
I checked all MSRs, but none of them are used to regulate GPU power. In addition, the MSR 0x640 was unavailable, it was missing from the list after being dumped by MSRdump.exe. Either Intel is no longer using this MSR, or have they blocked it in some way?
After that, I was able to catch a process that started every time the Intel Graphics power plan was switched. Process Hacker helped me with this.
This svchost.exe process uses the umpo.dll file, and these system services are also associated with this process:
BrokerInfrastructure
DconLaunch
Power
SystemEventBroker
And all this ultimately refers to the acpipmi.sys file.
This is an ACPI PMI device, it is located along the ACPI\ACPI000D\PMI path, and is responsible for power management of various PC components.
But since I'm new to all this, I can't figure out where this ACPI PMI is in memory. And how to edit this memory.
Here's the information I could find on Google about the UMPO service:
User-Mode Power Service
Overview of the Power Metering and Budgeting Infrastructure
ACPI Power Meter Interface
Overview of the ACPI 4.0 Power Metering Objects
Power metering
pmi.h header
IOCTL_PMI_SET_CONFIGURATION IOCTL (pmi.h)
Power Meter Interface
Ioctl.cpp