AMD EPYC 72F3 vs Intel Xeon Gold 6154 Comparison

Intel Xeon Gold 6154 and AMD EPYC 72F3 are both server-class processors aimed at similar workloads, yet they approach the task from opposite design philosophies. The data shows a remarkably consistent picture across six Cinebench benchmarks, with the Intel part winning every single test by a narrow 1.5% margin. However, the specifications behind these results could not be more different, making the choice between them a matter of workload characteristics rather than raw performance dominance.

Head-to-Head Benchmarks

The benchmark results are striking in their uniformity. Across all six tests, the Intel Xeon Gold 6154 posts a 1.5% lead over the AMD EPYC 72F3. In Cinebench R15, the Intel chip scores 2370 in multi-core and 334 in single-core, while the AMD chip scores 2334 and 329 respectively. This pattern holds exactly through Cinebench R20 and R23, with the Intel part scoring 9878 and 1394 in R20, and 23521 and 3320 in R23, against AMD's 9728 and 1373 in R20, and 23164 and 3270 in R23.

This consistent 1.5% delta suggests that the two processors are effectively equivalent in raw computational throughput for these rendering workloads. The Intel Xeon Gold 6154 wins all six head-to-head matchups, but the margins are small enough to be considered within run-to-run variance on most systems. In practical terms, a user running Cinebench R23 multi-core would see the Intel part finish at 23521 versus 23164 for the AMD — a difference of roughly 350 points out of over 23,000, which amounts to less than two percent.

The percentile rankings reinforce this near-parity. The Intel Xeon Gold 6154 sits at the 63rd percentile of all CPUs, while the AMD EPYC 72F3 sits at the 62nd. Their average benchmark scores tell a similar story: 6803 for Intel versus 6700 for AMD, a difference of about 1.5%. Each processor's nearest rivals also occupy similar territory. The Intel chip's closest competitor is the Intel Core i7-12700E at 6776 (a 0.4% difference), while the AMD chip's closest is the Intel Xeon Gold 5317 at 6710 (a -0.1% difference). Neither processor dramatically outperforms its immediate peers.

Where Each One Wins

Based strictly on the benchmark data, the Intel Xeon Gold 6154 wins every tested category. It leads in Cinebench R15, R20, and R23, in both single-core and multi-core variants. The single-core wins are notable because they indicate that Intel's older Skylake architecture still delivers competitive per-thread performance against AMD's newer Zen 3 design. In single-core tests, the Intel part scores 334, 1394, and 3320 across R15, R20, and R23, versus AMD's 329, 1373, and 3270.

However, the multi-core results are where the architectural differences become interesting. The Intel Xeon Gold 6154 has 18 cores and 36 threads, while the AMD EPYC 72F3 has only 8 cores and 16 threads. Despite having fewer than half the cores, the AMD part nearly matches the Intel chip in multi-core performance — trailing by just 1.5% in every multi-core test. This means AMD's per-core performance is substantially higher, even though the aggregate scores are similar. The Intel chip wins the multi-core tests, but it needs 10 additional cores to achieve that victory.

For workloads that scale with core count, the Intel part's advantage in raw multi-core scores makes it the nominal winner. For workloads that are lightly threaded, the Intel part also wins the single-core tests, though again by a slim margin. The AMD EPYC 72F3 does not win any benchmark in this dataset, but its performance per core is clearly superior — it achieves roughly the same multi-core score with 10 fewer cores.

Architecture Differences

The two processors come from different eras and different design philosophies. The Intel Xeon Gold 6154 is built on Intel's 14 nm process and uses the Skylake-SP architecture, released in July 2017. It has 18 cores and 36 threads, with a base clock of 3.00 GHz and a boost clock of 3.70 GHz. Its cache layout includes 64 KB of L1 per core, 1 MB of L2 per core, and a shared 24.75 MB L3 cache. The chip is fabricated by Intel with 8,000 million transistors and fits the Intel Socket 3647.

The AMD EPYC 72F3 is a much newer design, built on TSMC's 7 nm process with the Zen 3 architecture (codenamed Milan), released in March 2021. It has just 8 cores and 16 threads, but runs at a higher base clock of 3.70 GHz and boost clock of 4.10 GHz. Its cache configuration is dramatically different: 64 KB of L1 per core, 512 KB of L2 per core, and a massive 256 MB shared L3 cache. This large L3 cache is a hallmark of AMD's server chips, designed to keep frequently accessed data close to the cores. The AMD chip has 33,200 million transistors and uses the AMD Socket SP3.

The memory and I/O subsystems also differ significantly. Both support DDR4 memory and ECC, but the AMD EPYC 72F3 has an eight-channel memory bus with a peak bandwidth of 204.8 GB/s, while the Intel chip's memory bus is not specified in the data. The AMD part also supports PCIe Gen 4 with 128 lanes (CPU only), whereas the Intel chip's PCIe capabilities are not detailed. The AMD chip has a TDP of 180 watts, while the Intel chip draws 200 watts — a modest difference given the Intel part's higher core count.

FAQ

Q: Which processor has more cores?

A: The Intel Xeon Gold 6154 has 18 cores and 36 threads, while the AMD EPYC 72F3 has 8 cores and 16 threads.

Q: Which processor wins in Cinebench R23 multi-core?

A: The Intel Xeon Gold 6154 wins with a score of 23521, compared to the AMD EPYC 72F3's 23164, a 1.5% difference.

Q: What is the clock speed difference between the two?

A: The Intel Xeon Gold 6154 has a base clock of 3.00 GHz and boost clock of 3.70 GHz, while the AMD EPYC 72F3 has a base clock of 3.70 GHz and boost clock of 4.10 GHz.

Q: How much L3 cache does each processor have?

A: The Intel Xeon Gold 6154 has 24.75 MB of shared L3 cache, while the AMD EPYC 72F3 has 256 MB of shared L3 cache.

Q: Which processor has a higher TDP?

A: The Intel Xeon Gold 6154 has a TDP of 200 watts, compared to the AMD EPYC 72F3's 180 watts.

Q: What is the process node for each chip?

A: The Intel Xeon Gold 6154 is built on Intel's 14 nm process, while the AMD EPYC 72F3 is built on TSMC's 7 nm process.

Specification Differences

The two processors differ in almost every major specification category. The Intel Xeon Gold 6154 has 18 cores and 36 threads, versus 8 cores and 16 threads for the AMD EPYC 72F3. Base clocks are 3.00 GHz for Intel and 3.70 GHz for AMD, with boost clocks of 3.70 GHz and 4.10 GHz respectively. The TDP is 200 watts for Intel and 180 watts for AMD.

The process nodes are different: 14 nm for Intel (fabricated by Intel) versus 7 nm for AMD (fabricated by TSMC). Transistor counts are 8,000 million for Intel and 33,200 million for AMD. The AMD chip has a die size of 8x 81 mm², while the Intel chip's die size is not listed. L2 cache is 1 MB per core for Intel and 512 KB per core for AMD, while L3 cache is 24.75 MB shared for Intel and 256 MB shared for AMD.

Memory support differs as well: the AMD EPYC 72F3 has an eight-channel memory bus with 204.8 GB/s bandwidth, while the Intel chip's memory bus and bandwidth are not specified. The AMD chip supports PCIe Gen 4 with 128 lanes (CPU only), while Intel's PCIe support is not listed. The sockets are also different: Intel Socket 3647 for the Intel chip and AMD Socket SP3 for the AMD chip. The AMD EPYC 72F3 has a launch MSRP of $2468, while the Intel Xeon Gold 6154 has no launch MSRP listed.

The Verdict

The data presents a clear performance picture: the Intel Xeon Gold 6154 wins every benchmark test, but by a uniform 1.5% margin. For users prioritizing maximum absolute performance in Cinebench workloads, the Intel chip is the choice based on the numbers. Its 18 cores and 36 threads provide a solid multi-core foundation, and its single-core scores are also slightly ahead.

However, the AMD EPYC 72F3's performance is remarkable given its hardware. With only 8 cores and 16 threads, it nearly matches the Intel chip's multi-core output. This suggests that for workloads that are not heavily threaded, the AMD chip may offer better efficiency per core. The AMD part also has a lower TDP (180 watts versus 200 watts), a smaller process node (7 nm versus 14 nm), and substantially more L3 cache (256 MB versus 24.75 MB). The eight-channel memory bus with 204.8 GB/s bandwidth and PCIe Gen 4 support with 128 lanes give it a modern I/O feature set.

The choice between these two processors depends on what the benchmark results do not show: workload characteristics beyond Cinebench. For heavily threaded workloads that scale across many cores, the Intel chip's additional cores provide a safety margin, even if the measured difference is small. For workloads that are memory-bandwidth sensitive or benefit from large caches, the AMD chip's specifications suggest advantages that do not appear in these particular tests. The Intel chip's 63rd percentile ranking versus AMD's 62nd percentile is essentially a tie. Ultimately, the benchmark data shows near-parity, so the decision should rest on the specific requirements of the target applications, with the Intel chip taking the performance crown by the slimmest of margins.