Test Setup
Please read our test setup information carefully below, we are using Windows 11, and each testbed is customized for each CPU being compared. Each CPU required a different motherboard, but the AIO/cooling/PSU/SSD and RAM are all equal among them.
Integrated Graphics Drivers
For the Intel Core i9-12900K, we navigated to Intel’s webpage and downloaded the official latest drivers for it, labeled Intel Graphics – Windows DCH Drivers for 12th Gen Alder Lake CPUs. The driver version was 30.0.101.1340 and dated 2/7/2022, this was the absolute latest driver at the time of testing.
For the AMD Ryzen 7 5700G, we navigated to AMD’s webpage for processors with integrated graphics and downloaded the latest drivers for it. The version was Adrenalin 22.2.3 Optional dated 2/24/2022, this was the absolute latest driver at the time of testing.
DDR4 Memory
With the Intel Core i9-12900K and AMD Ryzen 7 5700G both of the integrated graphics rely on the system memory. They are bound to the performance of the system memory. We wanted to make this as equal as we could in our testing so that it is the same and only the GPU differs.
Since Alder Lake supports DDR4, we decided to run it on an Intel Z690 Chipset DDR4 motherboard, the GIGABYTE Z690 GAMING X DDR4 Motherboard that we have used multiple times for testing, instead of DDR5. In this way, we could use the exact same RAM modules on both systems, at the exact same frequency and RAM timings.
The RAM we used is a Crucial Ballistix 16GB Kit (2 x 8GB) DDR4-3600 BL2K8G36C16U4B. We ran it at the timings of 3600MHz @ 16-18-18-38 1T on BOTH CPUs. Therefore both iGPUs have the exact same memory bandwidth and capacity.
Why Compare a 12900K to a 5700G?
You may be asking, why are we comparing a $589 CPU to a $359 CPU, wouldn’t the 12900K whoop the 5700G? Well, if we were just testing CPU performance, yes. However, we are testing the integrated graphics performance, and therefore we will be 100% GPU bound, not CPU bound in performance. We are ultimately held back by both iGPUs, the CPU won’t be fully realized or utilized while we game because the GPUs are holding them back. If we were using a fast dedicated GPU, it would be relevant.
In addition, the 12600K, 12700K, and 12900K all use the same Intel UHD 770 iGPU, but the 12900K has it clocked at the highest boost clock. Therefore, if we test the 12900K, we will know the full potential of the Intel UHD 770 in gaming performance. We might as well because we’d learn nothing relevant testing the slower UHD 770 in the 12600K and 12700K. In addition, the 5700G represents AMD’s highest-end APU of the Vega 8 in 5000 series CPUs. Therefore, we are testing top-end vs top-end integrated graphics performance on the desktop, and the test is relevant against each other from both camps.