Product: CB5-5G12 heatsink & fan

Manufacturer: Coolermaster

Homepage: www.Coolermaster.com.tw

I spoke to Coolermaster about taking a look at some of their range, and within 48hrs the courier was at the door. We received a box full of goodies to review but we are looking at them one at a time rather than all at once.

Let’s take a look first at the CB5-5G12. This is a copper-based heatsink rated by Coolermaster for PIII or Socket A up to 1.3ghz.

The CB5-5G12 in all its glory:

First impressions after taking it out of the box are good. The unit is small, light and extremely attractive aesthetically. The overall build quality is very high, as you would expect from Coolermaster. The first problem that we noticed was there appeared to be bent fins on the heatsink. On further inspection, we realised that the malleability was so great that the slightest pressure on them would bend them and we must have bent them when we pulled the unit out of the box. This was easily rectified with a finger.

The CB5 comes with a large green thermal pad attached, so we initially tested with this in place and afterwards using basic thermal paste. Coolermaster informed us that they supply this on the coolers because AMD requires it (along with 16-22psi for the clips).

Test conditions

The system that the cooler was tested on:

Abit KT7A-RAID

Duron 700mhz

256mb PC100 generic

Voodoo 3 3000 PCI

Maxtor 6 Gig HD

Our aim is to provide a real world test, i.e. to simulate the experience that you would have if you had bought this heatsink. Therefore, we will take measurements from the onboard temp gauge under the cpu and decibel readings but not under lab conditions.

First we took off our generic fan and heatsink that had been supplied with the Duron 700. The ambient room temperature was kept at 22c, if this changed we stopped testing until it had returned to this base level.

Fitting

We only needed one attempt to put on the heatsink because its diminutive size allows for easy positioning and the clip was remarkably soft but still able to hold the unit in rock solid position. A far cry from other clips.

Can it perform?

Once we fired up the machine we were immediately struck by the lack of noise. Whether it is the fan’s speed (which ran at around the 5k mark) or the frequency, at any rate it is very quiet.

Our first reading is for the noise level, which we feel, is very important to a cooler. The readings are taken at 15cm from the cooler and also 1.5 metres away which is the rough distance that you would sit from the base unit.

As you can see, the figures bear out our initial impressions. This unit is remarkably quiet for one rated at 1.3ghz but can it actually cool at a high level?

Heat dissipation

Our generic hsf was idling at 31c and under Seti@home it was topping 44c. The Coolermaster immediately showed these figures up with an impressive improvement.

A 2c difference at idle and 5c under load is a great performance from this tiny unit. We next decided to loop Mad Onion’s 3D Mark 2000 with Seti. The result was none too surprising, the temp rose to 44c.

So, as a stock cooler at a level way below it’s top rating it is over performing, so let’s take it up a level. Previously our Duron could only reach a measly 840mhz! I know that I should be happy to overclock at all but…. The constraint before was the temp (at around 60c) and our PC100.

After the lengthy process of multiplier and fsb changing then re-boot and test, we found that 945mhz @1.85v was our new limit! Success!

The temp under full load was 53c and the system ran quite well but it was unable to complete 3D Mark 2000. We took it down to 900mhz @ 1.775v and the result was rock solid. The idle was 30c and under load it hit 48c, frankly these are well within the range that we would happily operate.

Conclusion

I am extremely impressed by this cooler’s performance with reference to my present set-up. It easily outperformed my generic hsf at all levels and by a large margin. The noise levels are very low and the psu probably makes more noise. The construction is first class and the look is great. It allowed us to increase our top speed by nearly 200mhz and this is not to be sniffed at. A solid all rounder for the stock cooler or overclockers.

The final ChillBlast rating is *******

We tested again with Arctic Silver II and the core fully cleaned with Isopropyl Alcohol. The results were astounding with a reduction of 2c on all readings. This is due to the inefficiency of thermal pads and cores that are already dirty from other pads. We will replace the fan with a more powerful delta and post the results later

Coolermaster site can be found at www.Coolermaster.com.tw

Many thanks to Coolermaster for the supply of this and many more of the company’s range that will be coming soon to the ChillBlast review section.

HOME