VCore discrepancy

nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
edited July 2007 in Hardware
Hi everyone,

I'm overclocking a socket 754 Athlon64 3200+ (2,2GHz, Venice) with an Asus K8N motherboard and I have noticed that when I raise the CPU voltage in the BIOS, it will not register the new value after I save the changes and the VCore voltage remains the same when I reboot.

I just want to increase the VCore from 1,45 to 1,50 or 1,55 if necessary, and the BIOS setting for the CPU voltage will show the higher value but the VCore reading in the BIOS is unchanged and CPU-Z also doesn't show the new voltage setting. I don't know why that happens and I'd like to know if this is a normal finding or not and whether there's a different way to apply the changes properly.

Thanks in advance for your time and any suggestions you wish to provide.

Regards,
n301

Comments

  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    Just a small update on this matter because I discovered that when I use 1,6V for the CPU voltage in the BIOS, the VCore is now shown to be 1,5V both in the BIOS VCore monitor and the CPU-Z voltage reading.

    I don't know why it happens like this and I'm assuming that CPU voltage and VCore mean the same thing so I expected an increase to 1,5V for the CPU voltage would also lead to a VCore of 1,5V
  • lemonlimelemonlime Canada Member
    edited July 2007
    That is odd, but I'd be less surprised if the reading didn't increase at all. What BIOS revision do you currently have on this board? It could be a BIOS bug of some sort.
  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    Hi lemonline,

    I'm using the latest BIOS edition (1011.005) for this motherboard.

    I don't know if it's a bug or not but at the moment I have reduced the CPU voltage value slightly from 1.6 to 1.575, and the VCore monitor in the BIOS continues to vary between 1.52 and 1.536 as it did previously when I had the CPU voltage set at 1.6.

    If the CPU voltage is set below 1.55, the VCore monitor reading goes down to 1.42 - 1.44V.
    With the CPU voltage set at 1.45, the VCore monitor will be between 1.42 and 1.44 as expected.
    To make things even more confusing, when I set the CPU voltage at 1.4, the VCore monitor will still vary between 1.42 and 1.44.

    I can't make much sense of the voltage settings but at least the VCore is currently around the 1.5V mark, and this allows me to venture into 2.6 GHz at a reasonable 40°C :)
  • edited July 2007
    It sounds to me like your board is quite finicky on what vcore it will actually deliver that Venice socket 754 chip you have. But at least it will let you change vcore so you should be able to do some overclocking with it. BTW, what psu are you using on your system? If you have an inadequate psu it could possibly affect how well your board overclocks.
  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    Hi muddocktor,

    Thanks for your suggestion with respect to the PSU. It could be the culprit since it's just a 300W supply but only the basics are connected to the motherboard (1 hard drive, 1 DVD drive, 1 memory stick (512Mb), 64 Mb graphics card) and I'm not planning to push the VCore any higher than 1.55V. I just wanted to see if I could get there at all first and then check what clock speed this CPU is capable of before I consider a more powerful unit.

    It's not too shabby and currently it can hit 2,7 GHz (11X multiplier) with the memory kept at 1:1, and the temperatures are below 40 °C.
  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    It seems to be the low output of the 300W PSU in this case that isn't allowing uniform VCore values. It will not allow a VCore in the 1,55V territory at all unless I push the CPU voltage all the way up to 1.675 and in that instance the VCore goes right up to 1,6.

    For the little time I had it running at VCore 1,6, I did get a clock speed of 2,77GHz and it completed the 32M Super PI successfully and also showed no problems with over 3 hours of FAH but the temperatures did soar under the load and were constantly hovering around 50°C. Taking this into consideration along with the added stress on the PSU, I think it's more sensible to just stick to the default VCore until I get a better PSU since this processor appears to have good potential to reach a solid 2,8 GHz.
  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    Taking the PSU matter a little further, I'd like to ask whether a 450W unit will be suitable for overclocking purposes with the computer I'm using (Athlon64 socket 754 2,2GHz, 1x 512Mb RAM, 1x SATA hard disk, 1x IDE DVD burner, 1x 64 Mb AGP 4X Graphics Card), and then what PSU parameter should I look out for in order to get a more uniform VCore rise when I increase the CPU voltage.

    Thanks in advance for your replies and I appreciate any suggestions you wish to offer.

    Regards,
    n301
  • edited July 2007
    A 450 watt quality psu should be able to power your computer. For a recommendation, something like the Antec Earthwatts 500 would be a great choice according to the review and ZipZoomFly has it for $79 and they also have a $20 rebate that expires today. But even without the rebate that still isn't a bad price for a good psu.

    If you want to save a little more money and live slightly more dangerously (because of the name brand), you could also buy the Rosewill RP500-2, which Newegg has here for $49.99 plus shipping. According to the test done by madmat in the article I linked to, this cheaper psu is a pleasant surprise in it's performance/cost ratio. But the Antec is definitely the more efficient psu and will produce less heat because of it's higher efficiency.
  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    Thanks for your suggestions muddocktor and that Antec PSU is certainly one I'd consider and fortunately the 500W edition is available here in the UK as well.

    I came across a couple of other ones as well as I was looking for the Antec and since I know very little about PSU models and specifications, I have listed them below in case you feel that any of those are just as good or have any other benefits compared to the Antec.

    FSP BlueStorm AX500-A (500W)
    FSP SAGA 450W
    Jeantech JN120F-450-APSL (450W)
    Vcool 480W
    Tagan TG480-U01 (480W)

    Thanks again for your time and the useful information you provide :)
  • edited July 2007
    Fortron Source generally has decent power supplies and the FSB Bluestorm 2 500 watt unit was tested by HardOCP and found to be a very good psu also. On the others I'm not sure. Since you are in the UK, I imagine all the power supplies available to you have active pfc, which is good. But between the Bluestorm 2 and the Antec Earthwatts, I think the nod has to go to the Antec due to it's slightly better efficiency if the price on them is close.
  • nonstop301nonstop301 51° 27' 24.87" N // 0° 11' 38.91" W Member
    edited July 2007
    Thanks again for your input muddockor and I decided to order the Earthwatts in the end :)
    I'm not sure what active PFC is but the Earthwatts does have that feature. Perhaps surprisingly, the Earthwatts also costs slightly less than the BlueStorm at the retailers I looked at here in the UK.
    For future reference I'd just like to ask whether the fact that there are two 12V output rails, allows the PSU to support rises in the CPU voltage better than the PSUs that don't have this feature.
Sign In or Register to comment.