Testing

To test, I’m using my normal rig with the following specs:

First of all, you need to have the CPU mounting bracket on which comes with your motherboard, this thing looks like a claw and the Jetart mounting system uses it to clip to.

Before the AK2000 I had the Noctua NH-U12 on my CPU which doesn’t require the P4 mounting bracket so I had to remove it. To test the AK2000’s water block I had to reinstall it. However, I managed to lose all the pictures of the installed water block due to a horrendous HDD failure. Suffice to say I wasn’t too impressed. To add further insult to injury, when I was taking the mounting bracket off to remount the NH-U12, I managed to snap one of the clips. This means that I can’t use the mounting bracket anymore, so there aren’t any pictures of the CPU water block installed.

The CPU water block problem was entirely my fault (well, Maxtor’s fault but we won’t go into that… :P) but the VGA block is due to another issue. Once I installed the block and took pictures – which I subsequently lost – I removed the block so that I could take a picture of it on its own. When trying to reinstall the block, the thread on one of the holes for the screws was badly worn meaning the screw won’t hole the block in place. I now can’t mount the block in place so there won’t be an in-action shot of this either. However, I managed to take a picture of it not on the motherboard to give you an idea of how it’s mounted.

People say a picture is worth a thousand words, so I’ve got a hell of a lot of typing in-lieu of these two pictures…. The CPU water block did look pretty good with the shiny metal plate which the clips push on to hold the block in place. The transparent blue plastic and the copper went well and overall the kit looked professional. Another bonus was that the kit used a LOT less space in my case than my humungous NH-U12 meaning system ventilation was increased dramatically. I noticed that the PWM area and the Northbridge where noticeably cooler as a result. My RAM also seemed to be cooler.

On my graphics card, the block looked good, and was equal to my previous passive heatsink in the way of noise.

I have to say, the colour coded tubing looks quite cheap and a little tacky. I prefer clear tubing as it looks more professional and clean. I personally think that the tubing isn’t meant to be seen.

After start-up my rig was left for 30 minutes, and then the temperature was recorded to represent an idle reading. Full load was recorded at the end of 30 minutes of CPU-burn in test. Ambient was 28C throughout. To test the water cooling ability, I used a burn in test and had the AK2000 set at 3 different RPM speeds.

It must be noted that my PC runs a lot hotter than most due to the position that it is in (under my desk near a radiator).

It should be noted that on the top RPM the fan is very noisy and isn’t really acceptable in a market where silent cooling products are becoming a necessity. When the RPM is set lower, the noise is more tolerable.

First I had the CPU and VGA on the same loop which yielded less than impressive results. In fact, I was too scared to run the AK2000 on the low RPM at the temperature was going upwards of 60C.

The high RPM reading was more acceptable, but nothing to write home about. Once you factor in the amount of noise that the unit was making, it wasn’t really all that incredible.

Once the VGA loop was removed, the results were more acceptable. In fact, it was pretty much on par with my NH-U12 which I count as the best air cooler out there. However, the amount of noise was again a lot higher than the NH-U12.

Finally, I used the AK2000 just on my graphics card which was much more effective. In fact I would say that it’s the best cooling I’ve had on it. However, it’s rather bulky for just cooling a graphics card. On the plus side, I managed to eek an extra few MHz of overclock out of the card.

I move my PC around a lot (LAN parties etc.) and the AK2000 is simply not designed for this. It’s really annoying having to move the actual computer box and the water cooling box at the same time. It also means that there is more chance of a leakage. In short, water cooling isn’t really right for my setup.

After using the patrolling unit for a while, the temperature probe stopped working and would read something completely ridiculous like -15 or 100.