Silence!
No, that wasn’t a command that was the war cry of a computing movement! Gamers and modders around the world want three things from their computers; Performance, cooling, and silence. Personally, I don’t care how loud my fans are. I have one case that’s so quiet that I sometimes don’t even realize it’s on. I have another that sounds like a jet airliner getting ready to land. But, preference is preference, and the majority seems to opt for the sweet sounds of silence.
Evercool is well known for their fans. I have several of them myself. I’ve always been impressed with the force at which they move air across my motherboards. But their CPU coolers? I’ve not had much intimacy with those. So when I got the opportunity to review their new Silent Shark CPU Cooler, I was all over it. And as tempted as I was to watch reruns of Shark Week in HD to load the CPU for testing, I fought my urges and ran Prime95 instead.
Before we go any further, here’s a little info on the Shark:
|
- The design of the six heat pipes giant should provide adequate cooling which enhance heat dissipation effect up to 250W.
- The design of dual tower enhances heat dissipation effect.
- Special fin feature that generate outstanding vortex airflow.
- Copper-based design ensures that the CPU heat can be evenly transmitted to the heat pipes.
- HPO-12025 collaborates with Evercool exclusive Shark Fan and Silent Fan that can increase the speed of heat conduction.
- By different angles installation, our fan enhances the motherboard component cooling performance and overclocking.
- Whole new MPCS(Multi Platform Clip System) provides great installation stability.
Okay, so if it seems like their description is worded a little strangely, you should know that they are based out of Taipei, China. I know, the first thing that springs to mind is cheaply made imitations of electronics we have imported from there anyway, but Silent Shark has built itself up on reputation. Their products work, and they are concrete sturdy.
Okay, so now that we got all that out of the way, we can open this box. The packaging itself is enough to make you want to buy it. There are two freaking sharks right on the front “swimming” to frame the nicely sized window. If the window isn’t enough to let you know what you’re getting, they also put quite a few pictures of their product all over the rest of the box. There’s even a picture under the window of what you’re looking at through the window. I guess they really want you to know what you’re getting.
I popped open the top of the box and pulled out yet another box. This one was a small square white one filled with plastic baggies with all the hardware necessary for installation. There were also some extenders for the Shark’s fan wires. Then I pulled the Shark out.
Okay, I’m getting extremely irritated these days with the size of heatsinks. They simply don’t need to be this big to sufficiently cool. Or even to maximize cooling. These manufacturers need to learn the M.E.D. Minimum Effective Dose. It’s a sports/medicine term. It means you use exactly what’s needed to achieve your goal and not a speck more or less. But, like those goobers holding up political signs on the side of the road, whining will get me nowhere, and if this thing works then I can take solace in that fact.
Now, if you don’t have a rear CPU access panel like I have, then you’re going to be taking your motherboard out of your case for this install. Evercool provides their own custom backplates for their coolers, and you’re going to have to remove the stock one. If you’re new to this, I’m sorry. Most of you are old hats at taking your machines apart. A percentage of you will get a chubby from it. Numbers never lie, folks.
You have to unscrew the top of the Shark to remove the top cap and fan assembly in order to screw down the cooler onto your CPU. This isn’t an issue. The issue for me was that my higher profile RAM was too tall for this cooler to sit on top of it. I ended up having to take my good ram out and replace it with slower, albeit lower, RAM. Okay, so now I have the heatsink screwed down. I go to put the cap back on… and it won’t go. The cooler juts out almost exactly the same distance as the interior width of my case.
Resisting the urge to break out the lube, I worked it in there just enough that it popped into place. After a quick power check to make sure I hadn’t broken anything, I put the side of the case back on. No need to reinstall the retaining screws, the case itself was holding the fans in (Insert eye roll here).
Okay, I should give you some info on my lab. I have one room with not much more than computer hardware in it. The only PC running in this room is my testing unit, which is what I’m using to test the Shark. It has an AMD FX-4170 CPU rated at 4.2Ghz in it, with 8GB of DDR3 (Cut down to 2GB due to this coolers low profile), a 250GB SATA HDD, and an MSI motherboard. I keep my machine uncluttered, with all wires tied back and every available fan mount fitted with a fan. I like to think you all allow such generous airflow for your machines as well. If not, then shame on you. Oh yeah, and the temperature in the lab is a steady 68° F at all times.
The first thing I wanted to do is check the decibel level of the fans. I wanted to let the machine run for a full hour at idle before checking base temperatures anyway, so I took the opportunity to check the noise level of the fans. This heatsink has two fans on it. The middle fan is their Silent Fan, and the outside fan, which can be opened outward up to 30 degrees, is their Silent Shark Fan. It is named so because of the shark fin appearance of the blades on the fan. I tested this one first, as it was easier to get to. Just as advertised, I got right at 16.4 dBA. Not too shabby.
Now I had the issue of getting a reading on that center fan. I had to maneuver it out just enough that I could get my microphone up to it. It didn’t want to let go at first, but I got it out far enough and ran my test. It clocked in at a respectable 25.8 dBA. Just for reference, the stock cooler that came with my CPU could get to well over 30 dBA while playing Skyrim. So this is nice change of pace.
Now for the fun part. I let my computer sit at idle for nearly an hour while I did the fan tests and did some light reading. Now it was time to see what kind of temperatures this baby could suppress. I got an idle reading first to see how she did just sitting pretty. At stock, I was getting 27 C. With the Silent Shark installed, I was getting 26° C. That may not seem like much, but in Fahrenheit, that’s a difference of running at over 80 degrees at stock to down at 78 degrees with the Silent Shark. Not a mind-numbing jump in temps, but this is only the idle test. Now we’ll do the load, and then the Overclock.
I fired up Prime95 and let it run for fifteen minutes. What I got at the end of that time frame was a solid 37° C temperatures. A little higher than the 34° C that the Noctua NH-D14 provided, but certainly better than the 45° C that stock cooling gave me. Now it’s time to do a little overclocking and see what kind of cooling power she provides with higher loads on the CPU.
Left: Silent Shark Idle Temp. Right: Silent Shark Load Temp
I overclocked my FX 4170 from the stock 4.2Ghz up to 4.8Ghz. I’ve seen people get theirs over 5Ghz, and that would have been a little better for heat generation, but I just can’t seem to find that stable sweet spot on the 4170 beyond the 4.8 Ghz mark. But even so, it still generates some heat, so let us see what kind of temp we get at idle on an overclocked CPU. I let her sit for a little over an hour while I did a little editing and surfed the web, then came back to check the idle OC temp. 26° C. Exactly the same as the stock idle temp. Odd, but the load test will show the true jump in temp on an overclocked CPU.
Left: Overclocked idle temp. Right: Overclocked load temp
So I fired up Prime95 once again and let it run for a full half hour. Fifteen minutes would have been sufficient, but I wanted to make absolutely sure. I allowed Core Temp to run the whole time to make sure I didn’t miss a single degree fluctuation. At the end, I was very happy with the results. 39° C on Core Temp. A degree less than the Noctua NH-D14, and I was overclocked a little higher for this test than for the Noctua.
I should point out that when not stressed in any way, the AMD FX-4170 doesn’t like to throw accurate numbers when it comes to temperature. The sensor in the newer AMD CPUs intentionally diminishes accurate idle temps as they are based on the movement of electrons in the channels of the chip transistors. It sucks, but that’s why I run Core Temp constantly, trying to grab that golden idle number. The load temps are completely accurate, however, according to the info I got from an AMD source.
So what does this mean for the Silent Shark? While being outperformed in the standard load test, she was still a meaner creature when it came to snacking on heat on an overclocked CPU. So while the Shark isn’t the absolute best, it’s certainly near the top of its food chain. Got a Shark Week joke in there after all.