Announcement

Collapse
No announcement yet.

New tool for undervolt/overclock AMD K8L and K10 processors

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Hello blackshed, I have a AM2 board (Asus M2NPV-VM) with a Phenom II 925 installed running Ubuntu 10.10. Perhaps we could do some testing on my system? FYI the output of ./TurionPowerControl -l gives:

    Turion Power States Optimization and Control - by blackshard - v0.30
    Detected CPU:
    Family: 0xf Model: 0x4 Stepping: 0x2
    Extended Family: 0x10 Extended Model: 0x4
    Package Type: 0x1 BrandId: 0x1996
    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:18 FID:12 DID:0 Freq:2800 VCore: 1.1000
    core 0 pstate 1 - En:1 VID:26 FID:5 DID:0 Freq:2100 VCore: 0.9000
    core 0 pstate 2 - En:1 VID:32 FID:0 DID:0 Freq:1600 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:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    -- Core 1
    core 1 pstate 0 - En:1 VID:18 FID:12 DID:0 Freq:2800 VCore: 1.1000
    core 1 pstate 1 - En:1 VID:26 FID:5 DID:0 Freq:2100 VCore: 0.9000
    core 1 pstate 2 - En:1 VID:32 FID:0 DID:0 Freq:1600 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:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    -- Core 2
    core 2 pstate 0 - En:1 VID:18 FID:12 DID:0 Freq:2800 VCore: 1.1000
    core 2 pstate 1 - En:1 VID:26 FID:5 DID:0 Freq:2100 VCore: 0.9000
    core 2 pstate 2 - En:1 VID:32 FID:0 DID:0 Freq:1600 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:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    -- Core 3
    core 3 pstate 0 - En:1 VID:18 FID:12 DID:0 Freq:2800 VCore: 1.1000
    core 3 pstate 1 - En:1 VID:26 FID:5 DID:0 Freq:2100 VCore: 0.9000
    core 3 pstate 2 - En:1 VID:32 FID:0 DID:0 Freq:1600 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:32 FID:0 DID:1 Freq:800 VCore: 0.7625
    Processor Maximum PState: 3
    Processor Startup PState: 3
    Processor Maximum Operating Frequency: 2800 MHz

    Minimum allowed VID: 63 (0.375v) - Maximum allowed VID 18 (1.100v)
    Processor AltVID: 34 (0.738v)
    Northbridge Power States table:
    PState 0 - NbVid 18 (1.1000) NbDid 0 NbFid 4
    PState 1 - NbVid 26 (0.9000) NbDid 0 NbFid 4
    PState 2 - NbVid 32 (0.7625) NbDid 0 NbFid 4
    PState 3 - NbVid 32 (0.7625) NbDid 0 NbFid 4
    PState 4 - NbVid 32 (0.7625) NbDid 0 NbFid 4
    Northbridge Maximum frequency: 2000
    * Warning: PVI mode is set. Northbridge voltage is used for processor voltage at given pstates!
    * Changing Northbridge voltage changes core voltage too.

    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
    This is the part (i.e. VCore 0.85 - 1.425V):
    http://products.amd.com/en-us/Deskto...False&f12=True

    Comment


    • (this may be a double post, my first is in the moderation queue as it contained a URL)

      blackshard I have a AM2 board (Asus M2NPV-VM) with Phenom II 925 running Ubuntu 10.10 - perhaps we can use this for testing?

      Comment


      • Thanks very much for the log. Now there are 3 people with AM2 boards and AM2+/AM3 processor with the same wrong voltage values.

        I really convinced myself that I misunderstood the datasheet.
        Stay tuned, maybe I'll push out a testing version for people with AM2 boards with some express modifications, but I can't be sure of the date (I hope in this weekend) because at the moment I'm a bit busy

        Comment


        • Hello guys, I managed to do some modification to the code and removed some. People with AM2 boards and AM2+ processors please try this testing code available here:

          tpc testing version 0.30.1 - only for AM2 boards!

          Maybe try to list the pstates table and check if voltages are coherent. Also try to set some vcores and see if the -l switch reports them correctly. Any test would be really appreciated!

          Comment


          • Originally posted by blackshard View Post
            Hello guys, I managed to do some modification to the code and removed some. People with AM2 boards and AM2+ processors please try this testing code available here:

            tpc testing version 0.30.1 - only for AM2 boards!

            Maybe try to list the pstates table and check if voltages are coherent. Also try to set some vcores and see if the -l switch reports them correctly. Any test would be really appreciated!
            Hi blackshard, thanks for the update. The voltages are reported correctly, except for the line
            Code:
            Minimum allowed VID: 0 (1.550v) - Maximum allowed VID 0 (1.550v)
            where the minimum VID is obviously incorrect. Unfortunately, this makes it impossible to change vids at all, since all reasonable values are below the reported minimum and thus produce the error
            Code:
            Unable to set vcore: 1.403v is below minimum allowed vcore (1.550v)
            Note that the old version with the patch did not exhibit this problem. Thanks again for you work

            Comment


            • Looking at the code, it seems the problem is that my CPU has no minimum vcore. In the old version, at line 650 of K10Processor.cpp there was
              Code:
              if (minVid==0) {
              		if (getPVIMode()) return 0x3f; else return 0x7b;
              	}
              But in the new one it has disappeared, so 0 is returned as the minvid and this is just the same as returning maxvid.

              Comment


              • Thanks very much kbios. I updated the archive with a corrected version, the link is the same.

                Comment


                • Originally posted by blackshard View Post
                  Thanks very much kbios. I updated the archive with a corrected version, the link is the same.
                  Thanks, it works perfectly now!

                  I noticed that there are a few warnings during compilation, most of them are just about different integer types, but two (which are the same) influence the output:

                  Code:
                  Griffin.cpp:1151: warning: spurious trailing % in format
                  K10Processor.cpp:1370: warning: spurious trailing % in format
                  The percent sign of cpu usage is interpreted as a format identifier and thus ignored.
                  I've made a small patch that fixes all the warnings and I would be happy if you could review it.
                  Patch

                  Comment


                  • Thanks for the patch kbios, I will check it out. Unfortunately I have to deal with Visual C++ compatibility too so I need to double check. I'm a little busy lately, but ASAP I will push out a newer version with voltage values patch and your patch finally in place.

                    Comment


                    • Trinidad TL-52

                      Hi, can I use this tool on an AMD Trinidad processor? I have an old Amilo here with a TL-52 don't care much if I burn it

                      Comment

                      Working...
                      X