AMD EPYC 9224 vs Intel Xeon Gold 6336Y Comparison

AMD
AMD

AMD EPYC 9224

CORE STATE Genoa
CORE SPECS 24 Cores / 48 Threads
CLOCK SPEED 2.5 Base / 3.7 GHz Turbo
CACHE 64 MB (shared)
MAX TDP 200W
ARCHITECTURE Zen 4
nm
PROCESS 5 nm
LAUNCH DATE 2022
VS
Intel
INTEL

Xeon Gold 6336Y

CORE STATE Ice Lake-SP
CORE SPECS 24 Cores / 48 Threads
CLOCK SPEED 2.4 Base / 3.6 GHz Turbo
CACHE 36 MB (shared)
MAX TDP 185W
ARCHITECTURE Ice Lake
nm
PROCESS 10 nm
LAUNCH DATE 2021

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
3,874
3,899
cinebench_cinebench_r15_singlecore
546
550
cinebench_cinebench_r20_multicore
16,144
16,249
cinebench_cinebench_r20_singlecore
2,278
2,294
cinebench_cinebench_r23_multicore
38,439
38,689
cinebench_cinebench_r23_singlecore
5,426
5,462

Analysis: AMD EPYC 9224 vs Intel Xeon Gold 6336Y

The Intel Xeon Gold 6336Y and AMD EPYC 9224 are two 24-core server processors that land remarkably close in raw compute performance, despite being separated by a full architectural generation. Across the six Cinebench benchmarks, the Intel part wins every single test, but the margins are exceptionally narrow. The largest gap is a 0.7% advantage for the Intel Xeon Gold 6336Y, seen in Cinebench R15 multi-core (3899 vs 3874), Cinebench R20 multi-core (16249 vs 16144), Cinebench R20 single-core (2294 vs 2278), and Cinebench R23 multi-core (38689 vs 38439). The smallest edge is 0.6% in Cinebench R15 multi-core, where Intel scores 3899 against AMD’s 3874. In single-core tests, the Intel Xeon Gold 6336Y leads by 0.7% in both Cinebench R15 (550 vs 546) and Cinebench R23 (5462 vs 5426).

The average benchmark scores confirm the near-parity: Intel’s average is 11191, while AMD’s is 11118, a difference of just 0.7%. Both processors sit at the 66th percentile among all CPUs, meaning they occupy the same performance tier in the broader database. The head-to-head delta percentages are so consistent that they suggest the two chips are effectively trading blows within measurement noise. Intel wins all six tests, but the data does not support calling it a decisive victory. In practical terms, a workload that saturates all 24 cores will see almost identical completion times on either platform.

Head-to-Head Benchmarks

The Intel Xeon Gold 6336Y sweeps all six Cinebench comparisons, yet the margin never exceeds 0.7%. The most telling result is Cinebench R23 multi-core, where Intel scores 38689 against AMD’s 38439. That 0.7% lead translates to roughly 250 points—a negligible difference in a render that runs for minutes. The same pattern holds in Cinebench R20 multi-core, where Intel’s 16249 tops AMD’s 16144 by 0.7%. Single-core results are equally close: Intel takes Cinebench R20 single-core with 2294 versus 2278, a 0.7% edge, and Cinebench R23 single-core with 5462 versus 5426, also 0.7%.

The smallest delta appears in Cinebench R15 multi-core, where Intel’s 3899 beats AMD’s 3874 by just 0.6%. Interestingly, Cinebench R15 single-core shows Intel ahead by 0.7% (550 vs 546), which is the same relative margin as most other tests. The consistency of these deltas—all falling between 0.6% and 0.7%—indicates that the performance difference is uniform across both single-threaded and multi-threaded workloads. The AMD EPYC 9224 does not win a single head-to-head benchmark. Its best relative showing is Cinebench R15 multi-core, where it trails by the smallest margin. The data shows that any workload selection between these two will not be driven by raw Cinebench performance, as the scores are statistically indistinguishable.

Looking at the nearest rivals, both processors cluster around the same average score. The Intel Xeon Gold 6336Y’s closest competitor is the Intel Core i3-1305U, which scores 11200, a delta of -0.1%. The AMD EPYC 9224’s nearest rival is the AMD EPYC 7542 with a score of 11152, a delta of -0.3%. This reinforces that both chips sit in a crowded performance band where a few points separate many different products.

Architecture Differences

The architectural gulf between these two processors is vast, even though their benchmark scores are not. The Intel Xeon Gold 6336Y is built on the Ice Lake-SP architecture, fabricated on Intel’s 10 nm process node. The AMD EPYC 9224 uses the Zen 4 architecture, codenamed Genoa, on TSMC’s 5 nm node. This node advantage allows AMD to pack 26,280 million transistors across four 72 mm² dies, while Intel’s transistor count is not listed. The process difference explains the power characteristics, not the performance.

Cache configurations diverge significantly. Both have 64 KB of L1 cache per core and 1 MB of L2 cache per core, but the L3 cache differs dramatically: Intel provides 36 MB shared, while AMD offers 64 MB shared. That 28 MB difference is substantial for workloads with large working sets, even if Cinebench does not reveal it. Memory support also differs: Intel uses DDR4 with an eight-channel memory bus delivering 204.8 GB/s of bandwidth, while AMD uses DDR5 with a twelve-channel memory bus delivering 460.8 GB/s. The AMD platform has over double the memory bandwidth, a critical factor for memory-bound server workloads.

The PCIe capabilities are another major split. Intel provides Gen 4 with 64 CPU lanes, while AMD offers Gen 5 with 128 CPU lanes. This gives the EPYC 9224 twice the lane count and double the per-lane bandwidth, making it the clear choice for systems with many high-speed NVMe drives or GPUs. Both processors support ECC memory, but the memory type and channel count are fundamentally different. The AMD part also has a larger transistor budget due to the 5 nm process, though the Intel part compensates with a higher boost clock in some scenarios.

Clock speeds are close but favor AMD in the specifications. The EPYC 9224 has a base clock of 2.50 GHz and a boost clock of 3.70 GHz, while the Xeon Gold 6336Y runs at 2.40 GHz base and 3.60 GHz boost. Despite the lower clocks, Intel manages to edge out AMD in every Cinebench test, suggesting that the Ice Lake architecture extracts more instructions per clock in these specific workloads, or that thermal/power behavior favors Intel under the test conditions.

The Verdict

The data points to a clear but narrow conclusion: the Intel Xeon Gold 6336Y is the faster processor in Cinebench workloads, winning all six tests with deltas between 0.6% and 0.7%. For any application that mirrors Cinebench’s rendering pattern, the Intel part is technically the better choice, but the margin is so small that it will not be perceptible in real-world use. The AMD EPYC 9224, despite losing every benchmark, is not a weaker product—it simply trades raw Cinebench performance for other capabilities that the benchmark does not measure.

Server buyers should choose the Intel Xeon Gold 6336Y if their primary metric is single-threaded or multi-threaded Cinebench scores and they want the absolute highest number in that specific test suite. The Intel part also has a lower TDP at 185 watts versus AMD’s 200 watts, which could reduce cooling requirements in dense deployments. However, the AMD EPYC 9224 is the better fit for systems that need maximum memory bandwidth, as it offers 460.8 GB/s versus Intel’s 204.8 GB/s, or for users who require more PCIe lanes—128 Gen 5 lanes versus 64 Gen 4 lanes.

The AMD part’s 64 MB of L3 cache, versus Intel’s 36 MB, may also benefit workloads with large data sets that exceed Intel’s cache capacity. The fact that AMD uses DDR5 memory while Intel uses DDR4 means the EPYC 9224 is positioned for next-generation memory infrastructure, which could matter for future upgrade paths. The Intel processor, released in April 2021, is the older design; the AMD EPYC 9224 launched in November 2022. For a purely benchmark-driven purchase focused on Cinebench, Intel wins. For a system design that values memory bandwidth, PCIe expansion, and cache capacity, the AMD EPYC 9224 is the justified choice despite the slight Cinebench deficit.

Specification Differences

The two processors differ on nearly every major specification except core and thread counts. Both have 24 cores and 48 threads, but the AMD EPYC 9224 has a base clock of 2.50 GHz and boost of 3.70 GHz, while the Intel Xeon Gold 6336Y runs at 2.40 GHz base and 3.60 GHz boost. The AMD part consumes 200 watts TDP, compared to 185 watts for Intel. The socket is different: AMD Socket SP5 for the EPYC 9224, Intel Socket 4189 for the Xeon Gold 6336Y.

The process node is 5 nm for AMD (TSMC foundry) and 10 nm for Intel (Intel foundry). AMD lists 26,280 million transistors across four 72 mm² dies, while Intel provides no transistor or die size figures. L1 and L2 caches are identical at 64 KB and 1 MB per core, but L3 differs: 64 MB shared on AMD versus 36 MB shared on Intel. Memory support is DDR5 with a twelve-channel bus and 460.8 GB/s bandwidth on AMD, versus DDR4 with an eight-channel bus and 204.8 GB/s on Intel. PCIe is Gen 5 with 128 lanes on AMD, Gen 4 with 64 lanes on Intel. The AMD part has a launch MSRP of $1825, while Intel’s launch MSRP is not listed. Intel’s architecture is Ice Lake (Ice Lake-SP), AMD’s is Zen 4 (Genoa). The AMD part has a part number of 100-100000939; Intel’s is not listed. Both are active production server/workstation parts with ECC support and no integrated graphics. Neither has an unlocked multiplier.

FAQ

Q: Which processor has a higher boost clock?

A: The AMD EPYC 9224 has a boost clock of 3.70 GHz, while the Intel Xeon Gold 6336Y has a boost clock of 3.60 GHz.

Q: How much L3 cache does each processor have?

A: The AMD EPYC 9224 has 64 MB of shared L3 cache, while the Intel Xeon Gold 6336Y has 36 MB of shared L3 cache.

Q: What is the memory bandwidth difference?

A: The AMD EPYC 9224 provides 460.8 GB/s over a twelve-channel DDR5 bus, while the Intel Xeon Gold 6336Y provides 204.8 GB/s over an eight-channel DDR4 bus.

Q: Did the Intel Xeon Gold 6336Y win any benchmark against the AMD EPYC 9224?

A: Yes, the Intel Xeon Gold 6336Y won all six Cinebench tests, with the largest margin being 0.7% in Cinebench R20 multi-core, Cinebench R20 single-core, Cinebench R23 multi-core, and Cinebench R23 single-core.

Q: Are the average benchmark scores similar?

A: The Intel Xeon Gold 6336Y has an average benchmark score of 11191, and the AMD EPYC 9224 has an average of 11118, a difference of 0.7%.

Q: Which processor has more PCIe lanes?

A: The AMD EPYC 9224 has 128 Gen 5 lanes, while the Intel Xeon Gold 6336Y has 64 Gen 4 lanes.

DETAILED SPECIFICATIONS

SPECIFICATION
EPYC 9224
Gold 6336Y
Core Specs
Cores
24
24 0.0%
Threads
48
48 0.0%
Base Clock (GHz)
2.5
2.4 -4.0%
Boost Clock (GHz)
3.7
3.6 -2.7%
Frequency (GHz)
2.5
2.4 -4.0%
Turbo Clock (GHz)
3.7
3.6 -2.7%
Multiplier
25
24 -4.0%
SMP CPUs
2
2 0.0%
Cache
L1 Cache
64 KB (per core)
64 KB (per core)
L2 Cache
1 MB (per core)
1 MB (per core)
L3 Cache
64 MB (shared)
36 MB (shared)
Power
TDP (W)
200
185 -7.5%
Configurable TDP
200-240 W
—
Architecture
Architecture
Zen 4
Ice Lake
Codename
Genoa
Ice Lake-SP
Generation
EPYC (Zen 4 (Genoa))
Xeon Gold (Ice Lake-SP)
Process Size
5 nm
10 nm
Transistors
26,280 million
—
Die Size
4x 72 mm²
—
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
DDR4
Memory Bus
Twelve-channel
Eight-channel
Memory Bandwidth
460.8 GB/s
204.8 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket SP5
Intel Socket 4189
PCIe
Gen 5, 128 Lanes(CPU only)
Gen 4, 64 Lanes(CPU only)
AMD Multi-Die
IO Process Size
6 nm
—
Other
Market
Server/Workstation
Server/Workstation
Production Status
Active
Active
Launch Price
$1825
—
Part Number
100-100000939
—
Package
FC-LGA6096
FC-LGA4189
View EPYC 9224 Details View Xeon Gold 6336Y Details