CPU profiles in the BIOS:
If you’ve been following our tests of Z890 motherboards, you’ll know that there are three profiles inserted directly into the BIOS. The idea is to avoid the instability concerns of Intel processors. As you can see from the BIOS screenshots, there are three profiles: Intel Default Mode, Intel Baseline Mode and ASRock Extreme Mode. Logically, these three profiles should make a difference in terms of power consumption and therefore performance. To find out, we’ll load each profile in turn and run a series of multithreaded benchmarks under Cinebench 23.
Cinebench R23 with the different profiles :
Here we go, with these three profiles: Intel Default Mode, Intel Baseline Mode and ASRock Extreme Mode.
Intel Baseline Mode is the lowest-performance profile, but the one that consumes the least power and maintains the lowest temperatures. We obtain a score of 37796 points (178 watts). Now, in our opinion, it’s a shame to miss out on the performance of the Core Ultra 9 285K with this profile.
The Intel Default Mode profile is closer to what we’ve come to expect in terms of performance, with a score of 42318 points (238 watts). The ASRock Extreme Mode profile consumes the most power, at 253 watts according to Cinebench R23, but with a very slightly higher score of 42622 points.
Two overclocking profiles in the BIOS :
This is the first of the Z890 motherboards we’re testing, featuring two CPU overclocking profiles. To use them, first select the ASRock Extreme Mode profile. We have Gaming OC Preset Stage 1 and Gaming OC Preset Stage 2. Let’s load them and check the gains obtained. As you can see from the screenshot above, the benefits of these two stages are clear to see.
Let’s get to testing with Cinebench R23. We can see that Stage 1 has modified the multiplier coefficients of the P and E cores. In terms of performance, we go from 42622 points to 43062 points, a gain of just one percent. Power consumption rises from 253 to 286 watts. Unfortunately, Stage 2 offers nothing more, with an identical score.
