Testing & Performance
For the testing section of this review, we will run the ML240P Mirage RGB through 2 different tests. The first will test the stability of the CPU’s clock as well as putting it under a sustained load for a period of time. The second will take place to give us what we could expect max temperatures to hit during intense processes like video encoding. All tests fans were running at max RPM of 2000RPM and the pump was running at full speed as well. Even with these settings maxed, the pump was inaudible and the fans were still extremely quiet.
For the first set of tests, we decided to go with a well-known program to test CPU stability called “OCCT”. This program goes through various types of calculations which require massive amounts of CPU computations and in turn increases its temperature and tests its stability. For each configuration of the CPU, we first let the system idle for 5 minutes to get a fairly accurate idle temperature then proceeded to run OCCT’s Small Dataset for 20 minutes to record the average and max load temperatures. If at any point during the test the computer crashes, we adjust the OC setting until it is stable for the entire 20-minute run. You may notice we are not using Prime95 (version 26.6); this was because no matter what cooling we had on this specific 9900K it would always start hitting over 100C with an overclock. So, we decided to move to OCCT which still does an excellent job testing stability but produces less heat doing so. Once we get the CPU stable, we move to our second test which is running Cinebench R15 for 5 consecutive runs. This test will give us an idea of what the max temperature we could encounter is during an intensive task in a real-world scenario.
For our first configuration, we run the 9900K at stock speeds and voltage which pegs all the cores to 4.7GHz during our test. After 20 minutes of OCCT, we get a max temperature of a toasty 89C and an average of 79C. This is to be expected as running the 9900K at stock tends to overvolt the core when and my 9900K itself always tends to run hot (especially Core #2). In Cinibench R15, we see a nice drop in temperature with the max hitting only 79C and an average of 72C. Seems we have some room to OC so we will move to our next profile for 4.8GHz.
OCCT – Small Dataset – Stock
Cinebench R15 – Stock
Moving to our 4.8GHz profile with a voltage of 1.295v we see an improvement in our OCCT temperatures with the max being 5C lower at 84C and the average also dropping 4C to 75C. In Cinebench, we see both our max and average temperatures increase with a new max of 85C and an average of 79C. You will notice that the max temperature was from our Core #2 with the other core only hitting 76-82C. These temperatures seem promising and we should be able to get a little bit more out of these 9900K.
OCCT – Small Dataset – 4.8GHz @ 1.295v
Cinebench R15 – 4.8GHz @ 1.295v
Moving up to our next overclock of 4.8GHz, we had to increase our voltage 0.05v which could definitely be seen in the temperatures. OCCT temps largely remained the same with only a 2C increase in max temperature at 86C and the average maintaining 75C. In our Cinebench run, our max temperature rose 8C capping out at 93C and our average hitting a new high of 86C.
OCCT – Small Dataset – 4.9GHz @ 1.345v
Cinebench R15 – 4.9GHz @ 1.345v
When overclocking, I like to keep the average temperature below 90C keeping it about 10C from tjMax and at 4.9GHz we are pretty close to that. I did try to go for 5.0GHz on this chip but no matter what voltage I gave it, OCCT would still keep crashing. Overall, the ML240P Mirage did an excellent job on keeping my 9900K in check and average temps below 90c with a 4.9GHz overclock. Though I’m disappointed I didn’t hit 5.0GHz, it wasn’t the fault of the ML240P but more so losing the silicon lottery. There are many chips out there that can hit 5.0GHz with much less voltage than my 4.9GHz overclock and the ML240P Mirage would have no issues cooling it.