Intel Core Ultra 9 290K Plus vs Intel Xeon 6736P Comparison

The Intel Xeon 6736P and Intel Core Ultra 9 290K Plus are both 96th-percentile performers, but they achieve that status through fundamentally different designs. The Xeon 6736P is a 36-core Granite Rapids server processor built for memory bandwidth and sustained throughput, while the Core Ultra 9 290K Plus is a 24-thread Arrow Lake Refresh desktop chip with clock speeds up to 5.80 GHz. Benchmark results show the Core Ultra 9 wins 12 of 17 head-to-head tests, but the Xeon 6736P dominates in specific server-oriented workloads, including a 97% lead in Passmark physics and a 30.1% lead in random string sorting.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Xeon 6736P has 36 cores and 72 threads. The Intel Core Ultra 9 290K Plus has 24 cores and 24 threads, meaning it lacks hyper-threading entirely.

Q: What is the biggest performance gap between the two?

A: The largest single-test margin is in Passmark single-thread performance, where the Core Ultra 9 290K Plus scores 4823 versus 2024 for the Xeon 6736P, a 58% difference. The Xeon's largest win is in Passmark physics at 6531 versus 3315, a 97% advantage.

Q: How do their average benchmark scores compare?

A: The Xeon 6736P has a higher average benchmark score of 87864, while the Core Ultra 9 290K Plus averages 84003. Both sit at the 96th percentile of all CPUs.

Q: Which processor supports more memory bandwidth?

A: The Xeon 6736P supports eight-channel DDR5 memory with 409.6 GB/s bandwidth. The Core Ultra 9 290K Plus uses dual-channel DDR5 with 115.2 GB/s bandwidth.

Q: Do both processors support ECC memory?

A: Yes, both the Intel Xeon 6736P and the Intel Core Ultra 9 290K Plus list ECC memory support as enabled.

Q: What are the closest rivals for each processor?

A: The Xeon 6736P's nearest rival is the Intel Xeon 6731P, with a 0.1% score difference. The Core Ultra 9 290K Plus's closest rival is the Intel Core Ultra 9 285K, with a 0.2% difference.

Architecture Differences

The two processors come from different Intel foundries and process nodes. The Xeon 6736P is built on Intel's 5 nm process with a 598 mm² die, while the Core Ultra 9 290K Plus uses TSMC's 3 nm node with a 243 mm² die and 17,800 million transistors. The Xeon uses the Granite Rapids architecture on Socket 4710, whereas the Core Ultra 9 uses Arrow Lake Refresh on Socket 1851.

Cache hierarchies diverge significantly. The Xeon 6736P features 144 MB of shared L3 cache, while the Core Ultra 9 290K Plus has 36 MB of shared L3. Per-core L1 and L2 caches also differ: the Xeon has 112 KB L1 and 2 MB L2 per core, while the Core Ultra 9 has 192 KB L1 and 3 MB L2 per core. The Xeon's larger total cache reflects its server design for massive working sets.

Memory architecture is a major differentiator. The Xeon 6736P supports eight-channel DDR5 with 409.6 GB/s bandwidth, compared to the Core Ultra 9's dual-channel DDR5 at 115.2 GB/s. PCIe lanes also favor the server chip: the Xeon provides 88 Gen 5 lanes (CPU only), while the Core Ultra 9 provides 20 Gen 5 lanes. The Core Ultra 9 includes integrated Arc Xe-LPG Graphics 64EU, whereas the Xeon has no integrated graphics.

Clock behavior separates the two. The Core Ultra 9 290K Plus has a base clock of 3.70 GHz and boost clock of 5.80 GHz, and its multiplier is unlocked. The Xeon 6736P runs at 2.00 GHz base and 4.10 GHz boost with a locked multiplier. The Xeon carries a 205 W TDP, while the Core Ultra 9 draws 125 W. The Xeon 6736P launched at a $3351 MSRP on February 23, 2025, while the Core Ultra 9 has no launch MSRP listed.

The Verdict

The data supports a clear split: the Intel Core Ultra 9 290K Plus is the better choice for single-threaded and lightly threaded workloads, while the Intel Xeon 6736P is the better choice for memory-intensive and multi-threaded server tasks. The Core Ultra 9 wins 12 of 17 head-to-head benchmarks, including every Cinebench test and Passmark single-thread, but its victories are concentrated in areas where raw clock speed matters most.

The Xeon 6736P wins 5 of 17 tests, but those wins include some of the largest margins in the comparison. Its 97% lead in Passmark physics and 30.1% lead in random string sorting suggest workloads that stress memory bandwidth or large caches will strongly favor the server chip. The Xeon also leads in integer math by 24.5% and data compression by 14.1%.

For users choosing between these two, the decision hinges on workload type. The Core Ultra 9's 58% single-thread advantage makes it the obvious pick for interactive desktop use, gaming, and applications that scale poorly beyond a few threads. The Xeon 6736P's 409.6 GB/s memory bandwidth and 144 MB L3 cache make it the data-driven choice for virtualization, database workloads, or scientific computing where memory throughput dominates. Both processors sit at the 96th percentile, so neither is a weak performer—they simply optimize for different bottlenecks.

Specification Differences

The Intel Xeon 6736P and Intel Core Ultra 9 290K Plus differ across nearly every major specification category. The Xeon has 36 cores and 72 threads, while the Core Ultra 9 has 24 cores and 24 threads. Base clocks are 2.00 GHz versus 3.70 GHz, and boost clocks are 4.10 GHz versus 5.80 GHz. TDP is 205 W for the Xeon and 125 W for the Core Ultra 9.

Sockets and architectures differ completely: the Xeon uses Intel Socket 4710 with Granite Rapids, while the Core Ultra 9 uses Intel Socket 1851 with Arrow Lake Refresh. Process nodes are 5 nm (Intel) versus 3 nm (TSMC). The Xeon's die is 598 mm², compared to 243 mm² for the Core Ultra 9.

Cache configurations are 144 MB shared L3 for the Xeon versus 36 MB shared L3 for the Core Ultra 9. Per-core L1 cache is 112 KB for the Xeon and 192 KB for the Core Ultra 9; per-core L2 is 2 MB versus 3 MB. Memory channels are eight for the Xeon and two for the Core Ultra 9, with bandwidth of 409.6 GB/s versus 115.2 GB/s. PCIe Gen 5 lanes are 88 for the Xeon and 20 for the Core Ultra 9. The Core Ultra 9 has integrated graphics and an unlocked multiplier; the Xeon has neither. The Xeon 6736P has a launch MSRP of $3351.

Head-to-Head Benchmarks

The Intel Core Ultra 9 290K Plus dominates the Cinebench suite with consistent 17.7% leads across all tests. In Cinebench R15 multicore, the Core Ultra 9 scores 5214 versus 4290 for the Xeon 6736P. Single-core R15 shows 736 versus 605. R20 multicore is 21727 versus 17875, and R20 single-core is 3067 versus 2523. R23 multicore is 51731 versus 42561, with single-core at 7303 versus 6008.

Passmark results paint a more nuanced picture. The Core Ultra 9 wins Passmark multithread with 60860 versus 50072, a 17.7% margin. It also leads in floating point math by 30.3% (214760 versus 149770) and find prime numbers by 22.1% (503 versus 392). Data encryption favors the Core Ultra 9 by 12% (52563 versus 46236). Single-thread Passmark is a dominant 4823 versus 2024, a 58% gap.

The Xeon 6736P's wins are equally decisive in its favored workloads. Passmark physics is the largest margin at 6531 versus 3315, a 97% advantage. Random string sorting goes to the Xeon at 103723 versus 79744, a 30.1% lead. Integer math favors the Xeon by 24.5% (206833 versus 166194). Data compression is a 14.1% win (796658 versus 698346). Extended instructions go to the Xeon by 6.2% (55563 versus 52338).

Where Each One Wins

The Intel Core Ultra 9 290K Plus wins in every single-threaded benchmark and all Cinebench multi-core tests. This makes it the clear choice for desktop applications, real-time workloads, and any software that relies on high clock speeds. Its 5.80 GHz boost clock and 58% single-thread advantage over the Xeon 6736P indicate a processor tuned for responsiveness and lightly threaded performance. The Core Ultra 9 also wins in floating point math, data encryption, and prime number finding, suggesting strength in scientific and cryptographic workloads that leverage modern instruction sets.

The Intel Xeon 6736P wins in workloads that stress memory bandwidth, large caches, or integer-heavy processing. Its 97% lead in Passmark physics points to simulation and physics engines that benefit from massive memory throughput. The 30.1% win in random string sorting and 24.5% win in integer math indicate database operations and general server tasks will favor the Xeon. Data compression at 14.1% ahead further supports its server positioning. Extended instructions, where the Xeon leads by 6.2%, suggests some specialized server workloads will prefer the Granite Rapids architecture.

The average benchmark scores reflect this split: the Xeon 6736P averages 87864 versus 84003 for the Core Ultra 9, but the Core Ultra 9's wins are more numerous. The Xeon's 5 wins carry larger average margins in its specialties, while the Core Ultra 9's 12 wins include many narrow Cinebench margins around 17.7%. For a user running varied workloads, the Core Ultra 9 offers broader wins, but for a user running memory-bound server applications, the Xeon 6736P's specific advantages are substantial and non-negotiable.