Raijintek Orcus 240 AIO cooler review

Cooling 189 Page 7 of 8 Published by

teaser

Adding some juice

Cranking up the volts.

It's at this point in the review where we get to the real acid test. How does a late 2017 AIO hold up against an overclocked 8 core CPU from the same year? At this point, 1.315v is being fed to each of the CPU cores, all running at a frequency of 3.8Ghz fixed. 1.15v is also being pumped through the SOC, as oppost to the 1.1v stock. All of this will, naturally, lead to notably increased power consumption, and HWInfo 64 reported a draw of around 115W for the CPU. This should be easily within the capability of a 240mm cooler, to be honest. Whilst most manufacturers do not list a TDP rating for liquid coolers, estimating this to be in the ballpark of 220-250W would not be that wild, in my opinion. Unlike the Leto Pro that we saw the other week, I don't really expect temperatures and noise levels to change that much when testing this cooler against an overclocked CPU.

Naturally, the tests run here are the same as before, with some time left between the stock and overclocked runs to ensure that the liquid in the loop has had some time to return to as close to ambient as it is going to get. The flipside of water taking longer to saturate and reach max. temperature is that it also takes longer to cool down, which is why your system's CPU might run hotter post gaming/torture with a liquid cooler than it would with an air cooler.

42117_untitled-2

Idle

42116_untitled-1

Load



So, how does the cooler perform? Put simply, it's a repeat of the stock tests. Well, to a degree. The Orcus clocks in at 52.5C, one degree higher than the S24. One degree is, with respect, entirely in the margin of error territory, so it is close enough to call the result an effective tie. I think Raijintek can be proud that their cooler is able to essentially match the S24 (definitively one of the best AIOs I have used to date) in terms of load thermals and noise when running a fairly heavily overclocked 1700X. As an aside, I would be interested in re-running this test with a higher end mainboard that allowed my to connect both radiator fans to a OPT header, or even CPU_FAN with the pump connected to a pump header.

What about noise, though? Well, the issue I mentioned on the previous page aside, noise was really not much higher than at stock, with the cooler only coming in at about 1-1.5dBa higher. Was that noticeable? Yes, very slightly. Did it bother me? No, not at all. The fans have a fairly innocuous 'hum' all the way up to about 70%, when they can start to get a little bit whiny. Naturally, as they were being hampered by being connected to SYS_FAN, and therefore thermals being read with a 20C offset in place, it's a good result.

I cannot call this results suite a mixed bag as I feel as though the cooler was certainly hampered, which is why I was so annoyed that there was no included hub, splitter, or that Raijintek hadn't opted for a smarter way to connect the fans. Had they done that, the above result I think would have been very different, certainly from an accoustics perspective.

With all that said and done, shall we get ourselves to a conclusion?

Share this content
Twitter Facebook Reddit WhatsApp Email Print