There’s been a lot of talk lately about Intel’s potential degradation issues with the high-end 13th and 14th generation CPUs, so I thought I’d use Powenetics and test an i9-13900K under three different power settings: Standard, Performance and Extreme. Then there’s the “Insane” profile, where you set the 4096 watts as an infinity factor and leave the CPU to its own self-protection.
Since I have a (self-designed and manufactured) high-end solution (Powenetics v2) that allows me to perform precise CPU power measurements, I decided to use it for exactly this purpose. These are my settings in detail:
| Intel Core i9-13900K (14900K) | Basic setting (“Standard”) | Performance | Extreme | Insane |
| Basic performance of the processor | 125 | 125 | 125 | 125 |
| PL1 | 125 | 125 | 250 | 4096 |
| PL2 | 188 | 253 | 250 | 4096 |
| Iccmax | 249 | 307 | 511 | 511 |
Intel explicitly does not recommend the baseline power profile as it severely affects the performance of the CPU. It is recommended to implement the highest power profile that is compatible with each individual motherboard design. As you will see from my test results, the baseline profile does indeed affect performance in Cinebench, even though this is only about rendering. There will be a second article where Igor will run these profiles with real workstation applications in a very time-consuming way, with some very interesting nuances and performance differences. But I don’t want to get ahead of myself. Intel only sees the usual suspects such as CB23 and, of course, games. So let’s start today with CB23 and rendering.
Test system and methodology
| Environment/laboratory | Ambient temperature: 26°C ±1°C Humidity: 40% ±4%RH |
| Motherboard | Asus Prime Z690-A |
| CPU | Intel 13900K |
| GPU | NVIDIA 1070 Ti |
| NVMe | XPG GAMMIX S50 Lite 1TB |
| MEMORY | XPG Lancer DDR5 (2 x 16GB) 6000MHz |
| Power supply | Seasonic Vertex 1200W (Cybenetics Platinum) |
| CPU cooler | Arctic Liquid Freezer III 420 A-RGB AIO (fan fully open) |
| Case | DimasTech bench |
I used a strong cooling solution to test how high the Intel CPU can run when all restrictions are lifted.
The load application is none other than the infamous Prime95 (Small FFTs) program, which puts a huge strain on the processor. To test the CPU’s performance under the different power profile settings, I use Cinebench R23. I also have the Powenetics system to continuously measure the CPU’s power consumption, which is much more accurate and faster than using a software solution like HWinfo.
36 Antworten
Kommentar
Lade neue Kommentare
Urgestein
Urgestein
Neuling
Urgestein
Urgestein
Veteran
Urgestein
1
Veteran
Urgestein
Urgestein
Urgestein
1
Urgestein
Mitglied
Neuling
Mitglied
Urgestein
1
Alle Kommentare lesen unter igor´sLAB Community →