MameXP said:
So the Mag3 perform better in yout test? as i see it int the graph the intersections sow that Mag3 is the better choice. Am i right?
Yes, the mag3 will perform better than the MCP600, at least that's an estimate with the information that's available today.
Thing is, we need to look at the size of the difference, and it is very small. Small enough to mean around a 0.1C difference at the CPU. For really low restriction setups the flow-rate difference will be larger, but >8-10LPM falls in the realm of diminishing returns for most waterblocks, and so any performance difference between the two pumps as seen at the waterblock would still be fairly minimal.
Even if one created a setup that totally played to the Mag3's strengths, one would still be unlikely to see more than a 0.5C difference at the CPU, and even then, such low-restriction high-flow waterblocks are generally out-performed by higher-restriction designs that closes the gap between the Mag3 and the MCP600 again.
Getting on to the issue of pump heat. The MCP600 will put about 8-9W of heat in the water. The Mag3 I estimate at around 14W given its design. That extra 5W difference at the pump will typically translate into around a 0.2C difference in the water temperature for many setups.
So the Mag3 gains ~0.1C from the higher flow, but loses ~0.2C from the extra pump heat.
i.e. the pumps will provide cooling performance that for the most part is essentially equal when it comes to the temperatures seen on the devices being cooled (CPU, GPU, etc). The differences are terribly minor (~0.1C).
The decision as to which pump to buy will ultimately boil down to which pump provides the additional features that one desires. Noise, voltage supply, barbs, reliability, EMF interference on monitors, etc. The performance difference between the two pumps is of negligible concern.