Testbed Setup


We used 2x2GB 1600MHz memory kits from A-Data and OCZ to test the memory compatibility of ZOTAC’s newest motherboard. The operating system was installed on a 320GB WD hard drive while the applications were stored by a 1TB WD hard drive. A fresh installation of Windows 7 64-bit was applied to every motherboard before testing.

Intel’s factory heatsink handled the processor cooling duties at the factory frequencies. During overclocking we used a Corsair H50 Hydro cooler to cool the processor although it was hardly needed.

When testing the ZOTAC 8200-ITX we relied on the nVidia chipset for graphics. The H55-ITX used an overclocked XFX 5770 for all tests except where noted. We used the same XFX HD 5770 for all of the DFI MI-P55 tests as well.

While testing the Intel-based motherboards we tried to keep the DDR3 timings as similar as possible. Unfortunately, due to a bug in the H55-ITX BIOS we had to run the CAS latency at 10. Memory frequency on these boards was set at 1333MHz due to fact that the i3 530 does not natively support 1600MHz RAM. Interestingly, the BIOS on the H55-ITX has an option for 1600MHz but choosing it prevents the system from POSTing. However, when the i3 530 was overclocked the memory frequency was bumped up to 1500MHz while using the same timings.

We used two ZOTAC motherboards for our wireless performance test. Our review motherboard, the H55-ITX, has wireless-n capabilities while the other board, the ZOTAC 8200-ITX, uses the older wireless-g standard. We measured the time it took both test systems to transfer a 1GB movie file. In order to simulate a typical HTPC environment the test systems were separated from the wireless-n router by thirty feet and two insulated walls.

For our lone AMD board all benchmarks were run using 4GB of 800MHz DDR2. The main RAM timings were set at 5-5-5-15 while any additional sub timings were left on auto.

We ran two sets of gaming benchmarks on the H55-ITX motherboard. One set uses Intel’s HD graphics while the other uses a HD 5770 from XFX. We have included these to demonstrate the viability of using i3 530’s integrated GPU for graphics.


Power Consumption

For our system power consumption tests we measure the watts being pulled out from the wall socket by the entire system. Keep in mind that power consumption at the wall can vary for a specific system based on the efficiency of the power supply being used. For these tests every power saving option in the BIOSes were turned on.

System Power Consumption - Idle Power

System Power Consumption - Load Power


The Zotac H55-ITX put down some of the lowest idle power consumption numbers that I have ever recorded on a desktop. System consumption under 100% CPU load was very mild as well. Numbers this low would make the H55-ITX ideal for a system that is rarely shutdown, like a HTPC.

Board Layout Gaming Results
Comments Locked

69 Comments

View All Comments

  • Duwelon - Sunday, February 28, 2010 - link

    Wrong.
  • GeorgeH - Sunday, February 28, 2010 - link

    Actually, very not wrong.

    After a quick search, it looks like PCP&C 610W is only about 50% efficient at 30-40W (AC power) and about 70% at 70-60W.

    This means the DC idle/load wattage numbers on the IGP are ~20W/50W. With a power supply rated at ~85% in that region, the measured power would be ~25W/60W, which is hugely different from the stated 35/76W - 30% "error" on the low end is pretty crappy.
  • michal1980 - Sunday, February 28, 2010 - link

    really? tomshard was able to get idle power down to ~25watts

    http://www.tomshardware.com/reviews/25w-performanc...">http://www.tomshardware.com/reviews/25w-performanc...

    now whos wrong?

    have an address I have some crows pie to ship you.
  • mariush - Monday, March 1, 2010 - link

    I'd rather use a bigger power supply (possible a Zen 400W passive one) and have the system use 40W instead of using that 220W power supply tom's hardware uses, with a 40 mm NOISY fan, just to get 22W.

    The difference in watts is so small my cell phone charger eats it when I leave it plugged in over night to charge my phone.
  • Duwelon - Sunday, February 28, 2010 - link

    So efficiency does or doesn't matter with you? Reality shows efficiency is all that matters as long as you have a stable and fairly priced PSU...
  • michal1980 - Sunday, February 28, 2010 - link

    efficeny matters. But I doubt that a 610w power supply is very efficent @ 30watts.

    have you ever heard of the right tool for the job?
  • mindless1 - Sunday, February 28, 2010 - link

    Have you ever heard of gathering more evidence before jumping to conclusions?

    The purpose is a relative, not necessarily absolute power measurement. Think about it, they are not trying to match every possible system configuration a user could have.

    For that, a 610W PSU is fine.

    Further, you don't factually know what the right tool for the job is because neither Tom's, nor most reviewers, test long term with minimally sized PSU (that tend to fail prematurely when ran at a higher % of total load rating).

    The right tool for the job is not really the most efficient without considering anything else. Think about it, if you really cared about efficiency you would turn off the computer you are using instead to post on the internet!
  • michal1980 - Sunday, February 28, 2010 - link

    what false conclusion? Did you even look at this review?

    except for the 860 cpu, @ load the system used less then 110watts of power. Using a PSU thats rated 6x that is a waste, on both ends, idle, and peak.

    PSU's certified for 80+, have tragets at 20, 50, and 80% of loads. The efficency of most, if not all falls significantly below a 20% load.

    a quailty 200W psu would be a much better fit for most systems builds then any 600w psu.

  • Duwelon - Sunday, February 28, 2010 - link

    Have you ever heard of admitting your original statement was wrong to just move on? Thanks for admitting efficiency does matter after your first blunder and then acting like a turd to defend it. No doubt you can find a better PSU, but at least now we know you're smart enough to realize efficiency does matter and you don't think it's pulling 610watts out of the socket or something. Have a nice day.
  • michal1980 - Sunday, February 28, 2010 - link

    wrong about what?

    a 610 watt psu is overkill for a system running ~30-150w.

    If you cannot understand that, you are too stupid to even continue this disscussion

Log in

Don't have an account? Sign up now