July 10th, 2024

Benchmarking ARM Processors: Graviton 4, Graviton 3 and Apple M2

The blog post compares ARM processors, highlighting Graviton 4's enhanced performance over Graviton 3. Graviton 4 matches Apple M2 in URL parsing but lags in Unicode validation and JSON parsing. Despite falling behind Apple M2 in some tasks, Graviton 4 shows significant improvements over Graviton 3, especially in base64 encoding/decoding.

Read original articleLink Icon
Benchmarking ARM Processors: Graviton 4, Graviton 3 and Apple M2

The blog post discusses benchmarking ARM processors, specifically comparing the performance of the Graviton 4, Graviton 3, and Apple M2 chips. The Graviton 4, based on the Neoverse V2 design, is Amazon's latest ARM processor, showing improved performance over the Graviton 3. Tests reveal that the Graviton 4 can match the Apple M2's performance in URL parsing despite running at a lower frequency. However, in Unicode validation and JSON parsing benchmarks, the Apple M2 outperforms the Graviton 4, although the latter shows significant improvements over the Graviton 3. The Graviton 4 also demonstrates better performance in base64 encoding/decoding tasks compared to the Graviton 3 but falls behind the Apple M2. Number parsing benchmarks show similar trends. In memory-level parallelism tests, the Graviton 4 slightly outperforms the Graviton 3. Overall, while the Graviton 4 does not match the Apple M2 on a per-core basis, it represents a notable advancement over the Graviton 3, showcasing improved instruction retirement rates and overall performance gains.

Related

Surface Laptop review: Microsoft's best MacBook Air competitor yet

Surface Laptop review: Microsoft's best MacBook Air competitor yet

The Microsoft Surface Laptop 7th Edition competes with the MacBook Air, offering Qualcomm chips for enhanced performance and battery life. It features AI capabilities, a larger display, and smooth operation with some app compatibility challenges.

AMD Zen 4 vs. Intel Core Ultra 7 "Meteor Lake" in 400 Benchmarks on Linux 6.10

AMD Zen 4 vs. Intel Core Ultra 7 "Meteor Lake" in 400 Benchmarks on Linux 6.10

The article compares AMD Zen 4 laptops with Intel Core Ultra 7 "Meteor Lake" SoC in 400+ Linux 6.10 benchmarks. Testing involved Ryzen 7 7840HS/U vs. Core Ultra 7 155H. Insights on performance, power consumption, and upcoming releases were discussed.

Scan HTML faster with SIMD instructions: .NET/C# Edition

Scan HTML faster with SIMD instructions: .NET/C# Edition

WebKit and Chromium enhance HTML content scanning with fast SIMD routines, boosting performance significantly. .NET8 now supports speedy SIMD instructions for C#, achieving impressive speeds comparable to C/C++ implementations.

Apple M5 Chip's Dual-Use Design Will Power Future Macs and AI Servers

Apple M5 Chip's Dual-Use Design Will Power Future Macs and AI Servers

Apple plans to use advanced SoIC packaging for M5 chips, enhancing performance for Macs, data centers, and AI tools. Collaboration with TSMC includes carbon fiber composite molding for improved thermal management. Mass production expected in 2025-2026, aligning with Apple's vertical integration strategy.

Mac shipments grew 6% year-on-year, outperforming the PC market

Mac shipments grew 6% year-on-year, outperforming the PC market

Mac shipments outpaced global PC market growth at 6%, securing Apple the fourth position with 5.5 million units shipped and a 9% market share. Apple's growth is expected to continue with the upcoming Apple Intelligence launch. Microsoft's Windows 11 release and Copilot+ PCs with ARM processors are set to drive further market expansion. Apple's transition to Apple Intelligence across all Apple Silicon Macs, including older models, may attract Intel Mac users.

Link Icon 2 comments
By @ksec - 4 months
Graviton3 = Neoverse V1 ~= ARM Cortex-X1

Graviton4 = Neoverse V2 ~= ARM Cortex-X3

M2 ~= Apple A15

This makes it easier for those who may be more familiar with Smartphone SoC. And it should be no surprise it is slower.

By @cozzyd - 4 months
I know Apple clang differs from upstream but I wonder if it makes sense to use clang instead of gcc for this comparison.