Intel Core i7-12700K vs Intel Core i5-13600K
Processor comparison for gaming — benchmark scores, estimated framerates and which one suits your build.
The short answer
Too close to matter. The Intel Core i7-12700K and Intel Core i5-13600K land within 3.4% of each other for gaming. Buy whichever is cheaper, or whichever fits the platform you already own.
Gap measured from estimated framerates across 5 games, not from benchmark scores — scores rank parts reliably but compress the distance between two fast ones.
Benchmark scores
| Measure | Intel Core i7-12700K | Intel Core i5-13600K |
|---|---|---|
| Gaming score ours | 32,749 | 33,638 |
| CPU Mark | 34,722 | 38,272 |
| Faster than | 95% of 1,255 | 96% of 1,255 |
| Geekbench 6 single | 2,555 | 2,718 |
| Geekbench 6 multi | 14,500 | 16,101 |
| Cinebench R23 single | 1,927 | 2,006 |
| Cinebench R23 multi | 22,219 | 23,960 |
| Cores / threads | 12 / 20 | 14 / 20 |
| L3 cache | 25 MB | 24 MB |
| TDP | 125 W | 125 W |
| Socket | LGA 1700 | LGA 1700 |
| Architecture | Alder Lake | Raptor Lake |
| Process | 10 nm | 10 nm |
| Memory support | DDR5/DDR4 | DDR5/DDR4 |
| Integrated graphics | UHD Graphics 770 | UHD Graphics 770 |
| Launch price | $409 | $319 |
| Released | 2021-11-04 | 2022-10-20 |
The gaming score damps high-core-count parts, which win productivity benchmarks without delivering proportional framerate, and credits 3D V-Cache where it is measurably earned.
Estimated FPS at 1080p
Both measured against the same reference graphics card — a GeForce RTX 4090 — so the processor is the variable being tested. Ranges, not exact figures.
| Game | Intel Core i7-12700K | Intel Core i5-13600K |
|---|---|---|
| Fortnite | 278–464 CPU | 292–486 CPU |
| CS:GO | 356–593 CPU | 367–612 CPU |
| GTA V | 121–202 CPU | 125–209 CPU |
| Apex Legends | 131–218 CPU | 133–222 CPU |
| PUBG | 142–237 CPU | 146–244 CPU |
Which suits your build?
These figures hold a reference graphics card constant. Your own build decides the answer: a faster processor buys nothing if the other component is already the limit. Run both against your actual hardware:
Check both in the bottleneck calculator