Page 1 of 6 123 ... LastLast
Results 1 to 10 of 146

Thread: New tool for undervolt/overclock AMD K8L and K10 processors

Hybrid View

  1. #1
    Join Date
    Oct 2009
    Posts
    200

    Default New tool for undervolt/overclock AMD K8L and K10 processors

    Hi guys, I made a new tool for undervolting/overclocking Family 11h (K8L) and Family 10h (K10) AMD processors, available both in Linux and Windows, for 32 bit and 64 bit flavours.

    The program name is TurionPowerControl, since it has born as a tool to control power usage of Turion ZM/RM and Athlon QL processors, then it has evolved to support also Phenom, Phenom II, Athlon II desktop processors and mobile Athlon II, Turion II and Phenom II.

    I put here because I need some feedback from users: K8L support is established and working well, but K10 support is still in a early state and needs testing and feedback since I have no continous access to K10 processors.

    You can grab the program here:

    TurionPowerControl 0.29 pre-alpha

    The package contains compiled win x86 and AMD64 binaries, compiled Fedora 13 AMD64 binary and full source code with compile script. Last but not least, there's a full documentation with some usage and compiling examples.

  2. #2
    Join Date
    Dec 2009
    Location
    Italy
    Posts
    174

    Default

    Quote Originally Posted by blackshard View Post
    Hi guys, I made a new tool for undervolting/overclocking Family 11h (K8L) and Family 10h (K10) AMD processors, available both in Linux and Windows, for 32 bit and 64 bit flavours.

    The program name is TurionPowerControl, since it has born as a tool to control power usage of Turion ZM/RM and Athlon QL processors, then it has evolved to support also Phenom, Phenom II, Athlon II desktop processors and mobile Athlon II, Turion II and Phenom II.

    I put here because I need some feedback from users: K8L support is established and working well, but K10 support is still in a early state and needs testing and feedback since I have no continous access to K10 processors.

    You can grab the program here:

    TurionPowerControl 0.29 pre-alpha

    The package contains compiled win x86 and AMD64 binaries, compiled Fedora 13 AMD64 binary and full source code with compile script. Last but not least, there's a full documentation with some usage and compiling examples.
    Hi blackshard, thanks for this tool. It seems great work, but unfortunately I am unable to get it working. I'm trying it on a Phenom II X4 940 BE (10h).
    Code:
    sudo ./TurionPowerControl -l
    gives
    Code:
    Turion Power States Optimization and Control - by blackshard - v0.29.1 (alpha)
    Detected CPU:
    Family: 0xf             Model: 0x4              Stepping: 0x2
    Extended Family: 0x10   Extended Model: 0x4
    Package Type: 0x1       BrandId: 0x1a86
    Detected Physical Cores: 4
    Detected processor: AMD Family 10h Processor
    Processor has 4 cores
    Processor has 5 p-states
    
    Power States table:
    -- Core 0
    core 0 pstate 0 - En:1 VID:16 FID:20 DID:0 Freq:3600 VCore: 1.1500
    core 0 pstate 1 - En:1 VID:24 FID:7 DID:0 Freq:2300 VCore: 0.9500
    core 0 pstate 2 - En:1 VID:32 FID:2 DID:0 Freq:1800 VCore: 0.7625
    core 0 pstate 3 - En:1 VID:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    core 0 pstate 4 - En:0 VID:0 FID:0 DID:0 Freq:1600 VCore: 1.5500
    -- Core 1
    core 1 pstate 0 - En:1 VID:16 FID:20 DID:0 Freq:3600 VCore: 1.1500
    core 1 pstate 1 - En:1 VID:24 FID:7 DID:0 Freq:2300 VCore: 0.9500
    core 1 pstate 2 - En:1 VID:32 FID:2 DID:0 Freq:1800 VCore: 0.7625
    core 1 pstate 3 - En:1 VID:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    core 1 pstate 4 - En:0 VID:0 FID:0 DID:0 Freq:1600 VCore: 1.5500
    -- Core 2
    core 2 pstate 0 - En:1 VID:16 FID:20 DID:0 Freq:3600 VCore: 1.1500
    core 2 pstate 1 - En:1 VID:24 FID:7 DID:0 Freq:2300 VCore: 0.9500
    core 2 pstate 2 - En:1 VID:32 FID:2 DID:0 Freq:1800 VCore: 0.7625
    core 2 pstate 3 - En:1 VID:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    core 2 pstate 4 - En:0 VID:0 FID:0 DID:0 Freq:1600 VCore: 1.5500
    -- Core 3
    core 3 pstate 0 - En:1 VID:16 FID:20 DID:0 Freq:3600 VCore: 1.1500
    core 3 pstate 1 - En:1 VID:24 FID:7 DID:0 Freq:2300 VCore: 0.9500
    core 3 pstate 2 - En:1 VID:32 FID:2 DID:0 Freq:1800 VCore: 0.7625
    core 3 pstate 3 - En:1 VID:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    core 3 pstate 4 - En:0 VID:0 FID:0 DID:0 Freq:1600 VCore: 1.5500
    Processor Maximum PState: 3
    Processor Startup PState: 3
    Processor Maximum Operating Frequency: No maximum defined. Unlocked multiplier.
    
    Minimum allowed VID: 123 (-0.375v) - Maximum allowed VID 0 (1.550v)
    Processor AltVID: 34 (0.738v)
    Core 0 C1E CMP halt bit is disabled
    Core 1 C1E CMP halt bit is disabled
    Core 2 C1E CMP halt bit is disabled
    Core 3 C1E CMP halt bit is disabled
    
    Voltage Regulator Slamming time register: 4
    Voltage Regulator Step Up Ramp Time: 8
    Voltage Regulator Step Down Ramp Time: 8
    Processor is using Parallel VID Interface (probably Single Plane mode)
    Processor PState Identifier: 0x2
    PSI_L bit not enabled
    If I try for example
    Code:
    sudo ./TurionPowerControl -set core all pstate 3 frequency 700
    I get
    Code:
    Turion Power States Optimization and Control - by blackshard - v0.29.1 (alpha)
    Detected CPU:
    Family: 0xf             Model: 0x4              Stepping: 0x2
    Extended Family: 0x10   Extended Model: 0x4
    Package Type: 0x1       BrandId: 0x1a86
    Detected Physical Cores: 4
    set frequency 700 Mhz on all cores pstate 3
    
    For frequency 700, FID is 12.000000, DID 2.000000
    
    
    For frequency 700, FID is 12.000000, DID 2.000000
    
    
    For frequency 700, FID is 12.000000, DID 2.000000
    
    
    For frequency 700, FID is 12.000000, DID 2.000000
    
    *** -set parsing terminated
    Unfortunately /proc/cpuinfo and KDE System Monitor still report the CPU at 800 Mhz when idle. I've also tried the manual way with
    Code:
    sudo ./TurionPowerControl -pf 0 3 2 -pd 0 3 0
    (for example) but the result is the same. Please note though that launching
    Code:
    sudo ./TurionPowerControl -l
    after one of these commands shows that pstates have been correctly altered.

    Could you help me? Thanks again

  3. #3
    Join Date
    Oct 2009
    Posts
    200

    Default

    Hi kbios.

    Actually cpuinfo (and don't know kde system monitor too) don't update in real time the information about the cpu.
    Also many utilities (like cpufreq-info) state that they do hardware calls to report their data, but actually they don't!

    Tpc instead reads all its info directly from the processor, it can't be wrong.
    With Linux, the only way I know to see if cpu frequency has changed is to do a benchmark. Prime95 is my favourite benchmark: it has a builtin benchmark that takes seconds to say if your processor has changed frequency.

    Also, are you running your processor on an AM2 motherboard? In that case probably you can change voltage at various pstates, but it won't have effect.

  4. #4
    Join Date
    Jun 2006
    Location
    Portugal
    Posts
    524

    Default

    Quote Originally Posted by blackshard View Post
    Hi kbios.

    Actually cpuinfo (and don't know kde system monitor too) don't update in real time the information about the cpu.
    Also many utilities (like cpufreq-info) state that they do hardware calls to report their data, but actually they don't!

    Tpc instead reads all its info directly from the processor, it can't be wrong.
    With Linux, the only way I know to see if cpu frequency has changed is to do a benchmark. Prime95 is my favourite benchmark: it has a builtin benchmark that takes seconds to say if your processor has changed frequency.

    Also, are you running your processor on an AM2 motherboard? In that case probably you can change voltage at various pstates, but it won't have effect.
    Exactly. I opened a bug report about this some time ago on the kernel and it was rejected. The cpu speed shown everywhere normally is wrong when you over- or underclock.

  5. #5
    Join Date
    Mar 2010
    Posts
    19

    Default

    Doesn't work after compilation. My system is ArchLinux 64-bit, processor AMD Phenom II 905e

    Code:
    # tpc -l
    Turion Power States Optimization and Control - by blackshard - v0.29 (pre-alpha)
    cpuid:open: No such file or directory
    Cpuid_Fn0 Instruction failed
    Error: unable to get processor specifications

  6. #6
    Join Date
    Oct 2009
    Posts
    200

    Default

    First of all, grab newer version 0.29.1 alpha below. I'm sorry, but I can't edit first post to put updated version on top.

    TurionPowerControl 0.29.1 alpha

    Then all people having problems with cpuid/msr can take a look at the manual paragraph 1.2

  7. #7
    Join Date
    Dec 2009
    Location
    Italy
    Posts
    174

    Default

    Quote Originally Posted by blackshard View Post
    Hi kbios.

    Actually cpuinfo (and don't know kde system monitor too) don't update in real time the information about the cpu.
    Also many utilities (like cpufreq-info) state that they do hardware calls to report their data, but actually they don't!

    Tpc instead reads all its info directly from the processor, it can't be wrong.
    With Linux, the only way I know to see if cpu frequency has changed is to do a benchmark. Prime95 is my favourite benchmark: it has a builtin benchmark that takes seconds to say if your processor has changed frequency.

    Also, are you running your processor on an AM2 motherboard? In that case probably you can change voltage at various pstates, but it won't have effect.
    Hi, thanks for your answer. I tried setting all pstates to 100 mhz and I felt the result , so a big thanks for this tool. Yes, I'm using an AM2 motherboard. Apart from changing VID, do you think the reportd one is correct? Because 1.15V for 3600 mhz seems a bit too low to me (and the bios says otherwise).

  8. #8
    Join Date
    Oct 2009
    Posts
    200

    Default

    Quote Originally Posted by kbios View Post
    Hi, thanks for your answer. I tried setting all pstates to 100 mhz and I felt the result , so a big thanks for this tool. Yes, I'm using an AM2 motherboard. Apart from changing VID, do you think the reportd one is correct? Because 1.15V for 3600 mhz seems a bit too low to me (and the bios says otherwise).
    Most probably it is reported correctly, but there is an issue with AM2+ CPUs on AM2 boards: since there is just one single power plane, the processor and the northbridge (i.e. the memory controller) share the power source. So the processor is feed with northbridge voltage, which is usually higher than processor voltage, so those VID numbers means just nothing in such situation. That's the reason why AMD raise the TDP of AM2+ CPUs when put in an AM2 boards.

    You may try to change those VIDs but shouldn't happen anything at all. To be precise, I guess that because I never had the chance to experiment it by myself, but just read it on AMD documentation.

    At the moment the program can't change the Northbridge VID on K10 processors because I have to test and experiment by myself since the matter is a bit delicate.

  9. #9
    Join Date
    Oct 2009
    Posts
    200

    Default

    try to rerun tpc, maybe some other program (or perhaps the kernel) is manipulating the processor counters.

    Also it would be nice if you could post here the output of the program launched with -pcgetinfo switch

  10. #10
    Join Date
    Dec 2009
    Location
    Italy
    Posts
    174

    Default

    Quote Originally Posted by blackshard View Post
    Most probably it is reported correctly, but there is an issue with AM2+ CPUs on AM2 boards: since there is just one single power plane, the processor and the northbridge (i.e. the memory controller) share the power source. So the processor is feed with northbridge voltage, which is usually higher than processor voltage, so those VID numbers means just nothing in such situation. That's the reason why AMD raise the TDP of AM2+ CPUs when put in an AM2 boards.

    You may try to change those VIDs but shouldn't happen anything at all. To be precise, I guess that because I never had the chance to experiment it by myself, but just read it on AMD documentation.

    At the moment the program can't change the Northbridge VID on K10 processors because I have to test and experiment by myself since the matter is a bit delicate.
    You're right: changing the VID doesn't actually affect processor performance. The only way I have found to change it is to use a tool that changes northbridge vid, so the two voltages seem correlated indeed.
    Thanks for your clarification

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •