600W PSU sufficient ?

Discussion in 'Asus' started by Adam, Apr 9, 2014.

  1. Adam

    Adam Guest

    Is the following power supply ...

    - OCZ ModXStream Pro 600W Modular High Performance Power Supply

    sufficient for the components listed below?

    My components are the following ...

    1) Motherboard:
    ASUS Sabertooth 990FX (pre-R2.0 w/ 1394-FireWire) AM3+ AMD 990FX SATA 6Gb/s
    USB 3.0 ATX AMD Motherboard

    2) CPU:
    AMD Phenom II X4 965 Black Edition Deneb 3.4GHz Socket AM3 125W Quad-Core
    Processor HDZ965FBGMBOX

    3) Display Adapter:
    Sapphire 100315L Radeon HD 6850 Video Card - 1024MB, GDDR5, PCIe,
    mini-Display Port, HDMI, Dual DVI

    4) RAM:
    Crucial Ballistix Tactical 240-Pin DDR3 SDRAM 8 Dual Channel Kit 1866 (PC3
    15000) BLT2CP4G3D1869DT1TX0

    5) HDD:
    Seagate Barracuda ST2000DM001 2TB Serial ATA Hard Drive - 7200RPM, 64MB,
    SATA 6Gb/s

    6) Optical Drives:
    2 x DVD/CD-RW Combo, Lite-On Super AllWrite 24X SATA DVD+/-RW Dual Layer
    Drive (IHAS124-04)

    7) USB:
    keyboard, mouse, scanner

    8) PCI:
    Creative SoundBlaster Live! 5.1 sound card

    9) Fans:
    CPU, chassis, power
    Adam, Apr 9, 2014
    1. Advertisements

  2. Adam

    tb Guest

    Maybe this site can help you:
    tb, Apr 9, 2014
    1. Advertisements

  3. Are you not capable of adding up your consumption rates, and deciding
    for yourself?
    DecadentLinuxUserNumeroUno, Apr 10, 2014
  4. Adam

    Paul Guest

    The track record on hand calculations, isn't all that good :)

    Even some of the power estimation web pages, were off by
    a factor of two.


    The best source for video card power, used to be Xbitlabs.
    But they've stopped doing the measurements, so are no
    longer my first source of info.

    The manufacturer values, makes things a bit complicated.


    Max Power Draw: 127 W

    The estimate should end up being conservative in terms
    of picking a power supply. The card likely doesn't
    draw that much power.

    The rest will be easier to estimate. I allocate
    50W for motherboard and RAM, but YMMV. 300W plus
    a few storage devices. Call the total about 350W
    or so.


    Before buying an OCZ, some part of OCZ went bankrupt, and
    Toshiba bought the company name (wanted the SSD stuff).
    I don't really know what happened to warranty responsibility
    for remaining product lines like PSUs. That could be why
    the supply is available at a lower price.


    +5V @ 25A
    +12V1 @ 25A
    +12V2 @ 25A
    -12V @ 0.3A
    +3.3V @ 25A
    +5VSB @ 2.5A

    The combined 12V1/12V2 limit is 42 amps. The video card
    and CPU would be 250W (rough ballpark) or 21 amps. Plus
    the storage devices use a little bit of that.


    It passes a quick smell test.

    The OCZ power supply section, is the previous PCPowerAndCooling
    company. A company that contracted out construction of supplies,
    but seemed to be on the ball when it came to not shipping a lot
    of crap. Lots of companies in the PSU industry contract out,
    but many of them are willing to ship shoddy goods to make a buck.
    PCPower was better than average (in terms of initial product
    quality, and also in terms of customer support). I don't know
    where the staff and skills will go now though.

    Paul, Apr 10, 2014
  5. Adam

    Paul Guest

    The last time I tried the Asus one, it had a "resolution"
    of 50W. When the Asus calculator ran, it rounded to the
    nearest 50W. That's why your "600W" number is so round.
    You can add an extra 5W electrical load, and the final
    number goes up by 50W (due to the rounding).

    The eXtreme one is still a bit high, but it's giving you
    the go-ahead to buy the 600W supply.

    In terms of "getting good" at calculating these things,
    you can use a clamp-on DC ammeter to validate your
    estimates. I own such a meter here. It uses a Hall probe,
    and no wires have to be cut to use it. Some of my
    "round numbers" come from using that thing. My only
    problem, is I no longer buy motherboards, so have no
    recent information to share.

    As an example, an optical drive (CD/DVD burner) has
    a name plate rating of 12V @ 1.5A. Yet, if I stick a
    CD in the tray, and run the spindle motor at high speed,
    the clamp-on meter reads 1.0 amps. If it was a BlueRay
    drive, it might be slightly higher. (At least, the BlueRay
    nameplate value is higher, for what that's worth.)


    The cheapest "fun" you can have, is spend $20 or so on
    a Kill-O-Watt meter. Build up your computer, using the 600W
    supply. Then use the Kill-O-Watt to measure the actual power.
    You'll find it will be less than the 446W from eXtreme, and
    less than the number I came up with. When measuring the power,
    you want a 3D game running on the computer screen, plus perhaps
    one copy of Prime95 running.

    Depending on where you're measuring, you are also making
    a measurement of the "inefficiency" of the switching power
    converters. The power supply contains a switching power
    converter (85% efficient). The VCore around the CPU socket, might
    be 85 to 90% efficient. When measuring with the Kill-O-Watt,
    a 125W processor might end up using 125W/(0.85*0.85) just
    to give some idea what impact the various conversions
    are having. The value at the wall, will be higher than
    the PSU rating (it takes 600W/0.85 at the wall or 705W
    at the wall, to make 600W inside the computer). As long
    as you're careful about how you interpret the Kill-O-Watt
    result, the Kill-O-Watt makes a no muss no fuss way of
    gaining some actual feedback, about how bad your
    system is on power. The power will be a lot higher while
    gaming, whereas it could be 100W at idle.


    That's for people who want some amount of actual info
    about their computer build, but without the usage of
    "wires" or "probes". And one of those for $20 or $25, is
    a lot less than a clamp-on DC ammeter.

    This is the style of clamp-on DC meter I got a number of years ago.



    I use the 40.00 Amp DC full scale range on that meter, for
    usage inside a computer case. You grab a bundle of similar
    colored wires on the main ATX cable, stick them inside the
    jaws of the meter, and the meter "sums" the current in
    the wires and reads it out. The 3.3V and 5V rails use
    multiple wires, and you just stick all the wires within
    the jaws to get the total value in one measurement. I
    can tell instantly how much CPU VCore is using, just
    by putting the jaws around the two yellow wires on ATX12V
    2x2 connector. Then, I have to correct the reading, for the
    85-90% efficiency of the VCore conversion, to conclude the
    power the CPU is using. As measuring the yellow wires, also
    includes the heat wasted on the VCore components.

    Paul, Apr 10, 2014
  6. Adam

    Paul Guest

    Well, never mind then.

    Paul, Apr 10, 2014
  7. Adam

    Adam Guest

    Adam, Apr 10, 2014
  8. Adam

    Adam Guest

    Sorry, I don't know how, which is why I am asking.
    Adam, Apr 10, 2014
  9. It is simple addition, not "a calculation" (inferring complexity).
    The elements ALL have consumption declarations.
    No need to estimate. A product does not get a CE cert without such a
    declaration, so they ALL have it.
    Hard to trust their "measurement" method as well. Especially if it is
    not clearly defined.

    And no, watching the AC consumption jump does not suffice.
    You must not know much about electronics. All the declarations for
    all the devices are correct.

    The power supply itself is the ne device that does NOT consume what it
    is rated at UNLESS it is fully loaded. Not very complicated at all.

    Sorry... I do not do ATI CRAP.
    You are not very bright. It very likely does, and "at idle" (if a
    video card could ever be such), it very likely consumes a good 85% of
    that full bore number.
    Again... there is NO NEED to estimate anything
    Not too bright. The cpu alone declares 125W.
    AND the power supply one fires such a load with should be nearly twice
    that in capacity.
    snip unrelated crap.
    All considering correct guesstimation AND maths. I have doubts on
    each ability, and the results thereof. A completely retarded crack.

    Modern RoHS lead free PCBs are soldered up HOTTER than the old 63/37
    process. They typically have 'burnt' smells associated with them, right
    off the line. That is before they ever even get any power.
    EVGA are the masters.

    I do not expect an ATI dope to believe it though.

    Too much retarded cross-posting by the OP or others in this thread
    DecadentLinuxUserNumeroUno, Apr 10, 2014
  10. Adam

    Adam Guest

    Thanks (Guru Paul), my PSU was purchased (around June 2012) before
    Toshiba's acquisition of OCZ (around December 2013).
    So, hopefully, the PSU is of good quality.
    Adam, Apr 10, 2014
  11. There are declarations on the products you listed. If you already
    have them installed, neither you nor we can sum up for you, since we do
    not know what your stuff uses either.

    My i7 hex core and X79 Dark Mobo and GTX650Ti all use more than yours.

    I have a 750 Watt gold series supply. Your 600 Watt choice is very
    likely plenty sufficient. Spending less could be risky and getting
    bigger probably not needed (unless you plan to use it again on a future
    build)(though it would likely use less then).
    DecadentLinuxUserNumeroUno, Apr 10, 2014
  12. Adam

    Adam Guest

    Thanks, I was nervous about using the 600W PSU but no longer. :)
    Adam, Apr 10, 2014
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.