merlin wrote:
...I'm testing out with the stock Core I3-2100 heatsink/fan first, which seems acceptable noise at idle. I think the first thing I need to do is replace the 80mm fan in the Antec ISK300-65 (which btw is a totally awesome case for a htpc, I'm surprised how well this turned out). [...] Update: Tested running 4 thread prime95 + furmark. Only hits about 68w at wall. Wow and now it's showing 22w at idle.
I added the emphasis above. According to
the SPCR article on the ISK300-65, the PSU efficiency should be comparable to the picoPSU at these low voltages (85% or better according to Antec's specs).
bwanaaa wrote:
anyone have experience comparing this board to the intel DH67CFB3?
I'm putting together a system using this board. I just got it last night, and didn't have time to do a lot of research, but I did some quick testing with the following components:
CPU: Intel i5-2500k w/stock heatsink+fan
MB: Intel BOXDH67CFB3
PSU: picoPSU-160-XT w/Edac 150W AC-DC Brick
RAM: Kingston KVR1333D3N9K2/8G (dual 4GB sticks)
Other fan: 140mm Case fan from Lian-Li PC-Q11B (LI121425BE-4-A)
HDD: Kingston 16GB SSD
My idle power was also 22 Watts (using a Kill-A-Watt).
The Intel motherboard is actually a replacement for a Biostar TH61. I originally went with the Biostar because it was the cheapest LGA1155 mini-ITX motherboard with an optical SPDIF port. But I sent it back because its power efficiency was dramatically lower: with the same config as above, I was getting 28 Watts at idle.
One thing to note about the Intel DH67CF: when you first boot it with a CPU that has greater than 65 Watts TDP, you get a big scary warning and legal disclaimer that you have to accept by pressing "Y" to continue. I intend to use a Thermalright AXP-140 with a downward-blowing (i.e. towards the motherboard) 140mm fan, so I should be OK. But for me, that warning dimmed the shine of the DH67CF a bit.
I'm tempted to buy an Asrock H67M-ITX just to compare. josephclemente was getting 26 Watts idle with a Seasonic X-560 PSU.
SPCR covered the X-650, which presumably has a similar efficiency curve to the 560---probably around 65% efficiency (or less) at these low idle power loads. Assuming 65% efficiency, 26 Watts AC input is about 17 Watts DC output.
If you assume the ISK300-65 PSU has 85% efficiency, then Merlin's 22 Watts of AC input is about 19 Watts DC output. But I would expect Merlin's DC requirement to be the same or slightly lower than josephclemente's (lower spec'ed CPU, fewer overall components). If you assume the DC requirements are the same for the two systems, then the ISK00-65 PSU's efficiency works out to be about 78% (at this level).
Hmm... If I could get the same or better idle power consumption out of the Asrock, then it's the better board all around: it's cheaper, and I'm assuming no >65W TDP warning. Just by going off pictures of the two boards, I'm
guessing that the Asrock has a more robust VRM section (but that's pure conjecture on my part).