Gigabyte H55N-USB3 : Mini-ITX done the Gigabyte way...
by Rajinder Gill on June 15, 2010 7:49 AM EST- Posted in
- Motherboards
- Intel
- Gigabyte
- Mini ITX
We start with the same old "swings and roundabouts" in performance figures that you’ll never notice in real world usage:
Power Consumption
Our power consumption testing utilizes the same batch of components under similar circumstances in a bid to monitor variances between idle and CPU load conditions. We install the vendor supplied power saving utilities on each board and enable power saving modes that don't involve any kind of underclocking or CPU core frequency modulation in order to run an apples to apples comparison.
ATX PSU switching losses are absent from our figures because we monitor power consumption directly at the DC rails of the PSU. These figures measure only the CPU, motherboard and memory DC power draw and exclude any other peripherals, such as cooling fans and hard drives etc. Actual AC power consumption for the motherboard will be anywhere from 15~40% higher than these figures depending upon the efficiency of your power supply.
The H55N-USB3's idle power consumption is excellent, although you lose out on efficiency under load compared to boards from Intel and ECS.
Overclocking
Using the stock Intel cooler with our 661 sample, we managed achieve around 4.2GHz Linpack stable:
The Clarkdale IMC isn’t ideally suited to high memory bandwidth, so we settled for a perfectly respectable DDR3-1640 with CAS 6-7-6-18 timings using our high-end Corsair Dominator kit whilst keeping QPI and CPU frequency in sync. This overclock was held stable by using 1.31Vcore (load voltage is likely a lot less as we kept Vdroop enabled), a set VTT of 1.31V and VDIMM of 1.65V.
Pushing higher than this would be possible with improved processor cooling, but as a side-effect requires more VTT if QPI and memory frequency is not relaxed - and we’re already setting 1.31V which seems to be the upper limit of what the VTT VRM can handle. Setting a higher voltage results in the board shutting down during Linpack, which confirms OCP does work on this particular rail.
We also checked out possibilities with a Lynnfield, using our 875-K sample to see what the board can handle:
3.66GHz is the upper level of what is achievable during Linpack with a HT enabled i7 processor. It’s not the greatest overclock, but is better than what some of the other H55 mini-TX boards can manage and is competitive with DFI’s MI-T36, aside from it being easier to cool the processor on the latter due to a better layout. Out of interest, we did attempt multiplier only overclocking too, but found that same overall CPU clock speed limit – any higher and the board shuts down during Linpack.
Should you opt for an i5-750 Lynnfield, CPU speeds up to 3.8GHz should be sustainable, provided you can find a way to keep the CPU cool enough under heavy loads.
Facebook
Twitter
RSS
77 Comments
View All Comments
ReaM - Thursday, June 17, 2010 - link
For Gods SAke do not overclock a mainboard that doesnt have MOSFET cooling. It will die after 20 minutes prime or lynx.Put a fat, bold DO NOT OVERCLOCK into overclock section.
Rajinder Gill - Thursday, June 17, 2010 - link
Heat is NOT the absolute killer of mosfet's on motherboards- it's current. Most mosfets will handle junction temperatures of 150~170 Celsius before they suffer a breakdown. What boards need are properly implemented OCP which on the face of it this board has.Later
Raja
play2learn - Thursday, June 17, 2010 - link
Did anyone find any information about Sound through HDMI ?Rajinder Gill - Thursday, June 17, 2010 - link
Sound through HDMI works fine into my Sony TV. As part of the power consumption test I run 1080P video and I did not get any glitches.just4U - Thursday, June 17, 2010 - link
It's a pricey board.. I was thinking it would be under $70 not over a $100.Faxfane - Friday, June 18, 2010 - link
Under 'Gaming and 3D Performance', section: AutoCAD, there is this part: 'ECS holds and advantage'. 'And' should be 'an', something easily auto-corrected by the brain, but is still unbecoming of such a professional site. Otherwise, this was a very informative and pleasurable article.shady28 - Saturday, June 19, 2010 - link
Lets get real. The main problem with ITX is the confined cases with poor cooling and small power supplies. This review is done without a case on a test bench, a high power high hea GTX 275, and with a 950W full size power supply.
What we really need to do know is what you can practically do with this ITX board. Some poor slob is going to run out and build an ITX box and high end GPU only to find his P/S keeps tripping or worse, something gets damaged and he's out a few hundred bucks.
Rajinder Gill - Sunday, June 20, 2010 - link
The Lian Li case listed in this review (PCQ08B) can hold a full sized ATX PSU and a full length graphics card. The PSU is only going to trip if you decide to buy an under-powered one.Regards
Raja
paihuaizhe - Sunday, June 20, 2010 - link
(nike-alliance).(com)=>is a leading worldwide wholesaler company (or u can sayorganization)
staryoshi - Tuesday, June 29, 2010 - link
I am currently using this board with an i7 860 and a Corsair H50. I simply clipped off the bulk of the retention bracket such that only the studs and their surrounding plastic remain. It does not contact any of the vulnerable ICs and I get my precious H50 performance. It fits beautifully in the Sugo 05.I am superbly satisfied overall, but I have run into a problem:
I can't get any changes to memory timings (G.Skill ECO DDR3-1333 7-7-7-21) to stick. It tells me that their has been an overclocking failure and it defaults back to 9-9-9-24. It acknowledges the XMP profile but will not use the timings. Is there some setting I'm overlooking?