Product: Assorted

Manufacturer: Assorted

Homepage:N/A

In a vain attempt to fill a few hours, we have decided to honour our recently departed review monkey. Therefore, we are proud to announce the latest review candidates hail from the land of thermal paste. In the world of overclocking, there is no greater bounty than a reduction in heat reading. We are constantly told about the miracle pastes available on the market but are they all that?

This, out of all the reviews we have done so far, was the most labour-intensive. Due to a large number of glaring anomalies, we were forced to re-test every paste twice. You'd better appreciate the effort or I'll.....not put out and then Bubba'll be really pissed. Yeah, now you are worried.

The contestants contain only one big name and that is Arctic Silver II. This is the market's leading brand and it's name is synonymous with high performance. If you want specs then here are the ChillBlast specs - it looks silvery. If you are still Oliver Twist about these things, then check out their site here.

Next comes the offering from Shinizu. This comes in typical OEM form - sachet, but unlike the other sacheted (if not then I have made it a verb) pastes it is not a white silicon base but grey. This makes me believe that it may contain silver or another metal. As ever, this is wild conjecture, but why else are you here?

Numbers two, three and four appear to be alike. They are all white silicon based pastes which came with the ThermalTake Mini Super Orb, Generic HSF and the syringe from Zalman. This last container shape is a particularly irresponsible. During a heavy night's "work" I nearly jacked up silicon.

We ran out of batteries taking some "medical" shots for a.....friend. So we were forced to trade the only thing we have of worth in return for batteries. The things that we do for you people. Marvel at the fruits of our labours (quite literally)

Now you can see the irresponsible shaping by Zalman.

Test Rig

Duron 700

Abit KT7A-RAID

2*128 mb PC133 CL2

Voodoo 3 3000 PCI

Maxtor 6 Gig HD

Enermax EG 365P-VE

The PC is the same one used for l the cooler reviews. We have chosen to use two coolers with different bases. The first is the Coolermaster CH5-5K12 with it's smooth copper bottom.

It is smooth, honest. Next, the drum rolls and in stumbles our in-bred friend the generic hsf with it's heavily-milled aluminium base.

This one is a real minefield of pits and craters. Just the sort of thing that the thermal pastes are designed to overcome.

As usual, we will take measurements from the onboard temp gauge under the cpu. We accept that this method may not be that reliable but it is the only way 99% of people measure their temps. The exact temps are unimportant, the relative figures are what matter.

The ambient room temperature was kept at 22C +/- 1C, if this changed we stopped testing until it had returned to this base level.

Before we applied any test paste, we cleaned the core and the bottom of the heatsink thoroughly with Isopropyl alcohol. Then the relevant paste was blobbed onto one end of the core, a business card or other suitable roach material was used to spread it across the core. The aim being to get as thin a film as possible covering the core with any paste left over was disposed of.

Grease? Paste?

I always refer to these substances as pastes but after this review I know better. There appears to be two main groups which are paste or grease. The difference is obvious. The only paste that we have here is the Shinizu and it is very stodgy indeed with a sheen on it that would indicate the opposite. If you work it for more than 10 secs it dries to a relatively hard state.

Disconcerting but it comes highly recommended so challengers are you ready?

The rest of the boys come in far more liquid states with differing degrees of viscosity.

Zalman was easy to dispense and the grease was quite shiny. It was relatively thick and sticky and could easily hold its peaks.

Arctic Silver II was matt but with similar physical properties to the Zalman.

TT's offering was close to the Zalman but it was more liquid and so returned to a spherical shape after being dispensed.

Generic grease had by this time half separated and was clear in patches. It was very close to water in fluidity. Not good.

The finger test is a basic aid to help you see how each thermal "substance" reacts to pressure. The Shinizu and generic both come out badly. The Shinizu appears to have an inability to spread thinly and evenly whereas the generic just runs away. Zalman and ThermalTake's greases are very similar but the the Arctic Silver II comes out at the head of the pack with the most fine uniform film.

Out of the frying pan...

The next test I like to refer to as spontaneous thermal immersion or as most people would say, putting the boys in the oven. Let's see how they react to being baked for 30 mins at 200F. This was to see if there was any change in their consistancy with the application of heat.

Left to right: Zalman, Arctic Silver II, Shinizu, ThermalTake and generic.

Not much obvious visual change except from the generic which has a greasy stain around it.

The finger test shows how the oven has exacerbated the Shinizu compound with the rest all appearing to fall within acceptable "spread" standards.

Results

As with the cpu cooler tests, we will take both idle and full load temperatures. Idle after 30 mins and full load after 30 mins of Seti@home. The cpu was set at 900 mhz with a core voltage of 1.775v. This is purely superfluous info as the only important factors are the relative temps.

Immediately, Arctic Silver II and the Shinizu take the lead. Surprisingly, the generic is not last with the Zalman skulking in the rear. Next we try the substances on the far rougher bottomed generic heatsink. This is where, we are told, thermal pastes come into their own. Filling the gaps created and encouraging optimal heat transfer.

The Arctic Silver II again puts in a good performance with the Shinizu dropping into the pack.

During testing, we noted some very large discrepancies at idle. These were around 34C and would not reduce despite being left for over an hour! It affected the Zalman most often, then the Arctic Silver and Shinizu. I can only think that it was as a result of the application's being too thick despite every attempt made to make sure it was paper thin.

Conclusion

The results were very interesting. Our margin for error is +/- 1C and this would feasibly mean that all the contenders may have posted identical results. The Shinizu was a good heat transfer substance when it has a smooth base to conduct the heat to, but when the rougher generic heatsink was used, it fell back slightly. This is probably due to its stodgy texture which was unable to create a perfect fit between core and sink.

The worst performance was put in by the Zalman which was a surprise since it seemed very similar in appearance to the other silicon greases. These other two make up the rabble but that is not to belittle their status. They are free and are more than up to most jobs in the short term but the separation that we witnessed was worrying. If this happens after a few hours, what will it be like after a few weeks?

I cannot call it but going on the numbers above, there is only one substance that was able to produce the lowest results on both heatsink surfaces and that was the Arctic Silver II. It came across as a well designed package, with a good applicator, excellent texture and premium performance.

The bottom line is if you are not bothered too much by a degree or two then go with the generic option. If, on the other you are an avid overclocker and want to be sure that you are reducing your core temp as much as possible then pay for the Arctic Silver II. You will be guaranteed a high quality of product. If you can save a degree then you are on the way to increasing your cpu lifetime.

Therefore the final Chillblast recommendation is Arctic Silver II.

Arctic Silver's website can be found here.

Zalman's website is here.

ThermalTake's site is here.

HOME