Can the big brother do everything better? – Corsair iCUE XC7 RGB Elite LCD water block review including teardown and test loop 2.0

1 - Intro and test loop 2.0 2 - Unboxing Corsair XC7 RGB Elite LCD 3 - Teardown and comparison with XC7 Pro 4 - iCUE software options 5 - Performance LGA1700 – Intel Core i9-13900KF 6 - Performance AM5 – AMD Ryzen 9 7950X 7 - WLP-Abdrücke und Montage-Erfahrungen 8 - Lessons learned, summary and conclusion

After our previous water block retest, which hopefully dispelled some untruths and rumors, here is the second part with a few improvements and implemented change suggestions from the community. Corsair has also asked us to test the successor to the XC7 Pro, the XC7 Elite LCD – and we’re happy to oblige. As a reference, I would recommend part 1 again, as it explains the calculation of the performance metrics and the test parameters of the CPUs in detail:

10 CPU water blocks tested on socket LGA1700 and AM5 in a big roundup including a necessary reappraisal

Test loop 2.0

As you can see from the title, I have revised my test setup, including the water cooling loop, in order to be able to record even more metrics or existing metrics at a higher resolution. In the comments to the last post, there was a request for a diagram of the loop, which I would like to fulfill today. It’s a bit cobbled together with draw.io, but after all its the content that matters. Anyone who knows of better software for drawing out water cooling loops is welcome to comment again.

The design has essentially remained the same, except that the quick-disconnect fittings on the water block side have been replaced by Alphacool ES, as these generate significantly less back pressure and therefore allow for far higher flow. In addition, another Aquacomputer highflow NEXT temperature and flow sensor has been integrated so that data can now be recorded before and after the water block.

Furthermore, a Peaktech P5144 differential pressure measuring device is now connected between the inlet and outlet of the water block, allowing the pressure difference or loss through the water block to be determined directly. Accordingly, T-pieces, additional quick-disconnects(QDCs) and adapters for converting to 8 mm hose diameter for the measuring device are now used here. I use a dedicated “zero piece” for calibration, which effectively only consists of fittings and a piece of hose. This is used to simulate a water block with a differential pressure of 0 mBar and a temperature difference of 0 K and to calibrate the sensors to this.

Finally, I set the two D5 pumps differently so that, together with the new QDCs, I could map a higher resolution for the flow rates. With the zero piece, the 2 pumps together now deliver 230 l/h and the single pump (top center in the diagram) still delivers 100 l/h. We therefore test the water blocks in the < 100 l/h range, where most normal water cooling systems with a single pump can be found in reality, and in the uncompromising enthusiast variant at > 200 l/h. The measurable performance difference between the two configurations should therefore also become larger or more high-resolution.

The water temperature at the outlet side continues to serve as the reference temperature for the delta core temperatures, just as in the previous test. Unfortunately, this time it has become even more difficult to collect a constant number of data sets for each 15-minute test run. Sometimes you get 300 samples, sometimes 500, with the exact same settings and real-time priority of the HWInfo process. Accordingly, I am forced to simply use all data sets that were logged, but a maximum of 419. If anyone knows a trick how to get HWinfo to a constant number of log entries, please let me know in the forum thread.