![[H]ardOCP](/images/blank.gif){kind=small}
- Date:
- Wednesday, October 02, 2002
- Author:
- Steve Lynch
- Editor:
- Sean Quinn
- Google +1
Water Block Round-Up
19 of the biggest baddest water blocks rounded up and tested, just for you guys out there that like your action wet and heavily overclocked.
About a month ago, we broke out with a little water block action during our review of the BTMS Water Cooling System and warned you there would be much more water-cooling coverage to follow here at [H]ard|OCP. Well, true to our word, we've rounded up 19 of the biggest and baddest water blocks on the planet. We decided to use every water block we had on hand for this, which includes a few that have been on the market for as long as two years, as well as the latest cutting edge blocks. What we hope to do is offer a little information on a wide variety of blocks in order to help you make a more informed decision when you contemplate what block is right for you.
The scope of this review will cover basic water block performance on both the AMD & Intel platforms under heavy overclocking conditions. Testing will also be conducted with the use of a 122W TEC (Peltier). Hopefully the results we provide here will assist water cooling enthusiasts, as well as those who have never used an H20 set-up, in understanding what they need from a water block.
We intentionally decided to review just water blocks at this time, leaving the other individual components of a water-cooling set up for a separate review. To level the playing field as much as possible, we used the same pump and radiator set up with all the blocks in this round-up (adapters were used where necessary). We understand that various aspects of a complete system all effect the overall cooling ability of any given water block, but we will not be including data like ”temperature in and out of the radiator” etc., because ultimately what really matters is the final temperatures achieved by each block at the CPU in a controlled environment.
What We Used:
We will be testing with the following:
Pentium 4 2.8GHz processor running at 3.4GHz (162 x 21 = 3402) and 1.7v.
AMD Thoroughbred 2200+ (Core A Rev.) (1.8GHz) running at 2.05GHz (152 x 13.5 = 2052) and at 1.8v.
Both CPUs were chosen because of the internal temperature sensors, which provide extremely accurate core temperature readings. The AMD TBred 2200+ generates a great deal of heat even at default settings, let alone when it's clocked at 2.05GHz at 1.80v. The same can be said for the Northwood at 3.4GHz and 1.7v; things get very warm quick.
Eheim 1250:
The Eheim 1250 has a flow rate of 317gph, giving it a high enough flow rate to effectively remove as much heat as these blocks can absorb and dissipate. Anything more than this in a single processor configuration, in my opinion, is overkill.
Black Ice Pro Radiator:
The Black Ice Xtreme LE Chrome is one of the best radiators out there, although an automotive heater core can be just as effective, with minimal effort converting them for use in your PC and work extremely well with medium flow 120mm fans like the 90cfm Sunon we used.
Custom Lexan Hold Down Plates:
Custom hold down plates are easy to make from Lexan using tools almost everyone owns (hacksaw blade, drill). These can be made to adapt AMD blocks to work on a Pentium4 or AMD Athlon. We made several of these in order to use older blocks on both platforms.
Copper Cold Plate:
Cold plates are used between TECs (Peltier) and the actual processor core. This particular copper cold plate came as part of one the Swiftech water blocks featured in this review, and I liked it so much. I used it for all the blocks.
122W TEC (Thermo-Electric Cooler):
While not the most powerful TEC out there, the 122W is more than enough to bring CPU temperatures down close to zero when coupled with a water block that can effectively remove the heat generated by the hot side of the TEC.
ASUS Probe v2.16.09:
The ASUS Utility was used to monitor temperatures throughout testing. Using a program such as this gives you the ability to actively monitor the rest of your system, like voltages and fan speeds, while recording temperature readings.
TEC Power Supply:
Using a stand-alone TEC power supply eliminates the strain that's put on the power circuit by using high watt TECs.
Our Testing Methodology:
All units were tested with the Pentium 4 2.8GHz CPU at 3.4GHz (162x21) at 1.7volts. The Thoroughbred 2200+ (1.8GHz) was overclocked to 2.05GHz (152x13.5) at 1.8v.
Temperature readings are taken from the on-die temperature sensor that both the Pentium 4 as well as the new Thoroughbred core CPUs now feature. The Asus Probe software was used to monitor and record those temperatures.
Prime95 was run simultaneously with a UT2K3 CTF map running in the background with 16 bots going at it to get our temperatures as high as possible for a sustained period of time for testing. The ambient temperature in the test room was kept at 75°F for the duration of the testing in order to make our test results as accurate as possible.
With the noted two applications running, we allowed the CPU to stay at 100% load for 30 minutes and then observed the temperatures and kept an eye on them from there. What we're looking for is where they peaked and leveled off. This is the temperature we recorded.