AMD EPYC 4584PX vs Intel Core Ultra 9 290HX Plus Comparison
AMD EPYC 4584PX
Core Ultra 9 290HX Plus
PERFORMANCE BENCHMARKS
Analysis: AMD EPYC 4584PX vs Intel Core Ultra 9 290HX Plus
FAQ
Q: Which processor has the higher overall average benchmark score?
A: The AMD EPYC 4584PX records an average benchmark score of 83090, while the Intel Core Ultra 9 290HX Plus scores 79574. The AMD part also sits at the 96th percentile versus the Intel part's 95th percentile among all CPUs in the database.
Q: How do these two chips compare in Cinebench R23 multi-core performance?
A: The AMD EPYC 4584PX scores 51524 in Cinebench R23 multi-core, which is 29.8% ahead of the Intel Core Ultra 9 290HX Plus's 39684. This is one of the largest multi-core gaps in the head-to-head results.
Q: Which processor wins the single-threaded PassMark test?
A: The Intel Core Ultra 9 290HX Plus scores 4951 in PassMark single-thread, beating the AMD EPYC 4584PX's 3795 by 23.3%. The Intel chip wins both the passmark_single_thread and passmark_singlethread entries by the same margin.
Q: What is the difference in core and thread counts?
A: The AMD EPYC 4584PX has 16 cores and 32 threads, while the Intel Core Ultra 9 290HX Plus has 24 cores and 24 threads. The AMD chip uses simultaneous multithreading, while the Intel chip does not.
Q: Which processor has more L3 cache?
A: The AMD EPYC 4584PX has 128 MB of shared L3 cache plus a 1x 64MB 3D V-Cache slice. The Intel Core Ultra 9 290HX Plus has 36 MB of shared L3 cache, a substantial difference favoring AMD.
Q: What are the release dates for these two processors?
A: The AMD EPYC 4584PX was released on 2024-05-20, while the Intel Core Ultra 9 290HX Plus was released on 2026-03-16. The Intel chip is a later product by nearly two years.
Architecture Differences
The AMD EPYC 4584PX belongs to the EPYC 4004 series, built on the Zen 4 architecture with the codename Raphael. It uses a 5 nm process node from TSMC, with a transistor count of 17,840 million spread across a die size of 2x 71 mm². The Intel Core Ultra 9 290HX Plus is part of the Core Ultra Series 2, using the Arrow Lake-HX Refresh codename, built on a 3 nm process node also from TSMC, with 17,800 million transistors on a single 243 mm² die.
The cache hierarchy reveals a significant architectural divergence. The AMD chip provides 64 KB of L1 per core, 1 MB of L2 per core, and a massive 128 MB of shared L3 cache, supplemented by a 1x 64MB 3D V-Cache slice. The Intel chip offers 192 KB of L1 per core and 3 MB of L2 per core, but only 36 MB of shared L3 cache. The AMD part's 3D V-Cache technology is a unique feature not present on the Intel chip.
Memory support is DDR5 for both, with dual-channel memory buses, but the bandwidth figures differ: the Intel chip achieves 102.4 GB/s versus the AMD chip's 83.2 GB/s. Both support ECC memory. PCIe lane counts also differ, with the AMD chip providing 28 Gen 5 lanes (CPU only) versus the Intel chip's 20 Gen 5 lanes (CPU only).
Integrated graphics distinguish the two as well. The AMD EPYC 4584PX includes Radeon Graphics, while the Intel Core Ultra 9 290HX Plus ships with Arc Xe-LPG Graphics 64EU. The AMD chip is a server/workstation part on Socket AM5, while the Intel chip is a mobile part on Intel BGA 2114. The AMD processor has a locked multiplier, whereas the Intel processor has an unlocked multiplier.
The Verdict
The data supports a clear split based on workload priorities. The AMD EPYC 4584PX wins 10 of the 17 head-to-head benchmarks, and its victories tend to be decisive in multi-core render workloads. The Cinebench R23 multi-core result is the headline: 51524 versus 39684, a 29.8% advantage. The single-core Cinebench R15 and R23 results are even more lopsided, with the AMD chip leading by 115.6% and 208.7% respectively.
For users who prioritize raw multi-core rendering, integer math, data compression, or physics simulation, the AMD EPYC 4584PX is the stronger choice. Its 128 MB L3 cache plus 3D V-Cache likely explains the 22.5% lead in PassMark integer math and the 12.6% lead in data compression.
The Intel Core Ultra 9 290HX Plus, however, wins in floating-point math, single-threaded PassMark, encryption, and prime number finding. Its 39.8% lead in PassMark floating-point math and 23.3% lead in single-threaded PassMark indicate a different strength profile. The Intel chip also has a 55 W TDP versus the AMD chip's 120 W TDP, making it a more power-efficient mobile option.
The database suggests the AMD EPYC 4584PX is the productivity and server workhorse, while the Intel Core Ultra 9 290HX Plus excels in floating-point and single-threaded tasks within a mobile power envelope. Neither chip is universally superior; the choice depends on whether the workload favors AMD's cache-heavy multi-core design or Intel's higher-bandwidth, higher-frequency single-thread performance.
Specification Differences
The two processors differ across nearly every major specification category. The AMD EPYC 4584PX has 16 cores and 32 threads, while the Intel Core Ultra 9 290HX Plus has 24 cores and 24 threads. Base clocks are 4.20 GHz for AMD versus 2.70 GHz for Intel, and boost clocks are 5.70 GHz versus 5.50 GHz respectively. The TDP ratings are 120 W for AMD and 55 W for Intel.
The AMD chip uses the AM5 socket, while the Intel chip uses BGA 2114. The process nodes are 5 nm and 3 nm respectively. Transistor counts are nearly identical at 17,840 million and 17,800 million, but die sizes differ significantly: 2x 71 mm² for AMD versus 243 mm² for Intel.
Cache configurations differ as described: AMD offers 64 KB L1 per core, 1 MB L2 per core, and 128 MB shared L3 plus a 64 MB 3D V-Cache slice; Intel offers 192 KB L1 per core, 3 MB L2 per core, and 36 MB shared L3. Memory bandwidth is 83.2 GB/s for AMD and 102.4 GB/s for Intel. PCIe lanes are 28 for AMD and 20 for Intel. The integrated graphics are Radeon Graphics versus Arc Xe-LPG Graphics 64EU. The AMD chip has a launch MSRP of $699, while the Intel chip has no listed launch MSRP. The AMD multiplier is locked; the Intel multiplier is unlocked.
Head-to-Head Benchmarks
The Cinebench R15 multi-core result goes to the Intel Core Ultra 9 290HX Plus with a score of 5981 versus 5193 for AMD, a 13.2% difference. But the Cinebench R15 single-core result flips dramatically: AMD scores 733 versus Intel's 340, a 115.6% advantage for AMD. This pattern of Intel winning multi-core in older Cinebench versions but losing single-core by wide margins is striking.
Cinebench R20 multi-core is a narrow AMD win: 21640 versus 21198, a 2.1% margin. Cinebench R20 single-core is also AMD by 2.1%, with scores of 3055 and 2992. The R23 results are where AMD separates itself: multi-core at 51524 versus 39684 (29.8% ahead), and single-core at 7274 versus 2356 (208.7% ahead).
PassMark results show a mixed bag. Data compression goes to AMD: 741648 versus 658724, a 12.6% win. Data encryption goes to Intel: 50008 versus 45902, an 8.2% win. Extended instructions favor AMD: 53774 versus 51290, a 4.8% win. Find prime numbers favors Intel: 519 versus 441, a 15% win.
Floating-point math is a major Intel victory: 201773 versus 121460, a 39.8% margin. Integer math is a major AMD victory: 201919 versus 164839, a 22.5% margin. The multithread PassMark test is close, with Intel at 59439 and AMD at 58117, a 2.2% Intel lead. Physics favors AMD substantially: 4574 versus 3387, a 35% margin. Random string sorting goes to AMD: 87690 versus 80327, a 9.2% win. The single-thread PassMark tests both go to Intel by 23.3%, with scores of 4951 versus 3795.
Where Each One Wins
The AMD EPYC 4584PX is the clear winner in rendering and compute workloads that favor large caches and high multi-threaded throughput. The Cinebench R23 multi-core result at 29.8% ahead, the 22.5% lead in integer math, and the 35% lead in physics all point to a processor that excels in CPU-bound productivity tasks. The data compression win at 12.6% and random string sorting win at 9.2% further reinforce this pattern. The 128 MB L3 cache plus 3D V-Cache appears to provide a substantial advantage in data-heavy workloads.
The Intel Core Ultra 9 290HX Plus dominates in floating-point math, with a 39.8% lead that is its largest single victory. It also wins single-threaded PassMark by 23.3%, data encryption by 8.2%, and prime number finding by 15%. The multithread PassMark win, while narrow at 2.2%, plus the Cinebench R15 multi-core win at 13.2%, suggest the Intel chip has competitive multi-threaded performance in certain older benchmarks despite its lower L3 cache.
The split is nearly clean: AMD wins in integer-heavy, cache-sensitive, and multi-core render workloads; Intel wins in floating-point, encryption, and single-threaded tasks. The Intel chip's higher memory bandwidth of 102.4 GB/s versus 83.2 GB/s likely contributes to its floating-point and encryption advantages. The AMD chip's massive L3 cache, meanwhile, explains its compression and integer math wins.
For server and workstation deployments where multi-core rendering, data compression, and physics simulation dominate, the AMD EPYC 4584PX is the data-backed choice. For mobile workloads that prioritize floating-point math, encryption, and single-threaded responsiveness within a 55 W TDP, the Intel Core Ultra 9 290HX Plus holds the advantage. The 10-7 win split in favor of AMD, combined with the larger margins on AMD's victories, gives the EPYC part the overall edge in the database's aggregate scoring.