Articles: Cases/PSU

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 ]

Measurement Resuts

Let’s first check out the results for each system case and see how the cooling of the HDDs depends on their positions.

The HDDs are numbered by their positions: the HDD1 and HDD2 are in the top cage, the others in the bottom cage. Obviously, the top cage calls for an additional fan because the HDDs are considerably hotter in it. The temperature is not dangerous, but high indeed.

As for the noise factor, the case is all right. You can only hear the fans working at minimum speed if you put your ear right next to the case. The sound of the HDDs’ heads could be heard during the IOMeter test yet it was much muffled. And you should note that the Raptor series drives are targeted at workstations and entry-level servers and do not work quietly. When the fan speed was set at 1600rpm, you could hear the fans. At maximum speed they produced an audible hum.

The HDDs are numbered from top to bottom again. There is a smaller difference between them and it is the higher HDDs that are cooled better now. They are closer to the back-panel fan whereas the bottom HDDs are farther from the fan and in a nook where the airflows can barely reach.

The noise is somewhat higher than with the P182. I could hear the HDDs more distinctly than in the previous case.

The HDDs are numbered from top to bottom. The system case is superb in terms of cooling. It is going to be a good choice for an advanced gaming configuration.

There is more noise, however. The fans are all right. They are very quiet while the 200mm fan at 400rpm is completely silent. The problem is that the HDDs are only separated from the outer world by the metallic mesh in the brackets and the cages, so their sound reaches the user in full. If the HDDs are not accessed, the case is but slightly worse than the Sonata Plus 550 in terms of noisiness.

The HDDs are numbered from top to bottom. The components are somewhat hotter than in the Nine Hundred, but there are fewer system fans. To remind you, the Three Hundred comes with the top fan and the back-panel 120mm fan preinstalled. It doesn’t have a fan on the front panel. Despite that, the case cools the HDDs and mainboard better than the above-discussed P182 and Sonata do. The Nine Hundred and Three Hundred also keep the graphics card’s temperature lower as the hot air coming from the CPU is efficiently removed from the reverse side of the graphics card. People who have top-end graphics cards or cards with passive coolers are going to appreciate this. The temperature of the HDDs in the Three Hundred can be as low as in the Nine Hundred if you install fans on the front panel (a fan is especially desirable in the bottom corner).

The case is about as noisy as the Nine Hundred. The fans are almost silent in the quiet mode, hum audibly at medium speed, and become irritating at high speed. The main problem is the sound of the HDDs that are only separated from the user with a mesh.

The HDD1 is on the bottom of the case, the HDD2 is in the top compartment under the optical drive. Cramped in the limited room with weak airflows, the HDD1 is somewhat hotter. The HDD2 feels better as it is surrounded with air that is refreshing by the PSU fan. Yes, it would be good to install a system fan in the bottom seat on the front panel. A short graphics card is better for it, too. Otherwise, the system case coped quite well with cooling my rather advanced test configuration.

The level of noise is about as high as that of the Sonata Plus 550, perhaps a bit higher. I guess it is acceptable for such a small system case.

The HDD1 is placed flat under the optical drive. The HDD2 is vertically to the side of the latter. The difference in the temperatures can be easily explained. The HDD1 has more space around it which is important for such a dense component placement as in this system case.

The case coped with cooling the HDDs, though. The problem was about the CPU cooler I could fit into it. When the CPU load was high, the cooler would instantly reach its maximum speed of 2500rpm yet could not keep the temperature low. The CPU was steadily getting hotter and hotter until began to skip clock cycles at 80°C.

I thought about the issue for a while. I didn’t have a Scythe Shuriken then but had a few different coolers similar to the Floston. I first tried a cooler named Tornado: it had the same fan as the Floston but a taller heatsink. The temperature lowered by a few degrees. Interestingly, the Floston had a copper core while the Tornado’s heatsink was all aluminum. It means that it is the heatsink surface area rather than the material that is most important for modern CPUs that have a heat-spreading cap.

Next I turned to more respectable manufacturers and took a GlacialTech Igloo 5057 PWM PP. Unfortunately, it didn’t fit into the NSK1380 by a couple millimeters. Then I took the fan from the GlacialTech cooler (a rated speed of 3600rpm) and installed it on the heatsink of the mentioned Tornado (2500rpm) hoping to keep the CPU temperature below 70°C.

I was surprised to see the cooler do much better with the new fan. The CPU temperature had been over 75°C at 2500rpm but now it was 69°C at 1928rpm after half an hour of the Prime95 test. It is with this handmade cooler that I performed the tests.

Why did I tell you all that if there exists the Scythe Shuriken, developed specifically for compact system cases? Well, I just wanted to illustrate the fact that brand-name coolers, although obviously similar to no-name ones, prove to be better in practical applications. You don’t pay for the brand only, you pay for quality!

From the noise aspect, the CPU cooler and the blower produced most noise in the NSK1380. The blower doesn’t have a tachometer or a speed regulator. You can only make it quieter by connecting to 7V power. On the whole, I made three modifications (replaced the CPU cooler with the Scythe Shuriken, switched the blower to 7V power, and stuck a plastic bar to the PSU as written above) that transformed the NSK1380 into a quiet system case. It was quite comfortable even in a living room at night. Of course, it is all more difficult with small system cases because compact coolers are not popular and you have to pick up your CPU from economical series, but that’s the tradeoff of the compactness. But I want to note again that I assembled a really advanced gaming system in the NSK1380, with a superb CPU and a very fast graphics card.

Generally speaking, the main cooling-related problem of the NSK1380 is that the airflows are different. The fans of the components drive the air in different directions. The scheme that Intel suggested in its BTX standard might work here: the CPU at the front edge of the mainboard, its fan driving the air along the latter. Alas, BTX mainboard are so rare that the NSK1380 would become yet another barebone kit if it were designed in BTX form-factor.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 ]


Comments currently: 0

Add your Comment