Startech Socket A Heatsink

SUMMARY: Another in a line of so-so heatsinks.

The kind folks at Everything Connected sent us a sample of StarTech’s Duron/Socket A heatsink.

This is a typical, low cost extrusion unit that you see all over because it’s cheap to make and low cost to vendors so that they can make a decent margin on it. Performance of such units are OK but nothing to get excited about. For “normal” use (spec voltage and speed), a heatsink like this might be an OK solution. Anything beyond that is questionable.

If you come across something like this and wonder how well it will do, there’s a couple of things you can look at that will give you a good indication of how it will stack up to other units:

• THE FAN: All things being equal, a higher speed fan will do more for performance than a lower speed fan, at least up until you push the heatsink beyond its design limit. This unit comes with a 60 x 15 mm fan which typically maxes out at about 15 cfm. There may be higher cfm units, but at this price-point, you’re not going to see one.
  
• THE BASE: Smooth is good, finish marks are bad. This unit might perform better with some lapping, but as shipped it is not well finished. If you run your finger over the base and can feel the “ripples”, performance will be compromised. I could feel ripples on this one.
  
• FLATNESS: If you happen to have a razor blade, you can use it to see how flat the base is. Just place the edge perpendicular to the base and hold it up to the light; if you see lots of light, the base is not particularly flat. Grease is an absolute, but it’s no substitute for a flat base. The Startech was good on this score.

• SURFACE AREA: The more fins you see, the more likely the heatsink will perform better than a unit sporting less fins. If you put two units side-by-side, all things being equal, more surface area will win out.

So after the “eyeball” test, I would not expect the Startech to be a standout performer. Average – Yes. Alpha class – No.

The Startech is very easy to mount – it was extremely easy to put on and get off. I can’t tell you that the clip force was up to AMD’s spec. I mounted it on my ASUS A7V with a Duron 700 @ 900 MHz, 1.84 volts, using Arctic Silver. It took about 3-5 minutes for temps to stabilize, and I ran Prime95 for another 15 minutes and then recorded temps using the ASUS temp probe. The probe is placed next to the CPU’s edge and gave the following results:

About what I expected. It is not inconceivable that, under certain conditions, the Startech might perform on par with the Alpha. Everything Connected ran their own test (HERE) and found it to cool as well as the Alpha. I did not, under the test I ran.

In summary, this is an OK heatsink for the price and for non-critical use.

Thanks again to Everything Connected for sending us this unit to test out.

Email Joe