RISC-V is currently slow compared to modern CPUs

After watching the rise of ARM displace Intel from most of my personal and server computing, I’ve been following the RISC-V architecture with intense interest. It’s an exciting open-source ISA that has the potential to reshape the CPU landscape, but there’s an important caveat: RISC-V is currently quite slow compared to modern CPUs in 2024. It is even slow compared to Raspberry PI boards.

How Slow is RISC-V?
You can start with these Geekbench performance scores for various RISC-V processors: https://browser.geekbench.com/search?q=RISC-V.

I haven’t seen any single threaded scores above 150 and no multi-threaded scores higher than 1500. The best result I’ve seen is the Sophgo Mango’s score of 147 single threaded and 1300 multi-threaded. And the average score for RISC-V CPUs in Geekbench is around 80 single threaded and 400 multi-threaded.

In comparison, Apple’s M4 series is seeing scores around 3800 single-threaded and 15000 multi-threaded. A more fair comparison would be the Raspberry Pi series of CPUs, but even then it isn’t close. The Raspberry Pi 5 scores around 1000 single-threaded and 2200 multi-threaded, while the Raspberry Pi 4 scores 300 single-threaded and 739 multi-threaded.

So in that context, the fastest RISC-V CPU I could find is 25x slower than a top-of-the-line Apple M-series chip and ~7x slower than a Raspberry Pi 5 and 2x slower than a Raspberry Pi 4 for single-threaded performance. Those that say that RISC-V is a viable replacement for x86 or ARM in the near term are kidding themselves.

Why Is RISC-V So Slow?
RISC-V implementations in the wild lack advanced features that modern CPUs rely on for speed, including sophisticated pipelining mechanisms, out-of-order execution capabilities, advanced branch prediction, and multi-tiered cache hierarchies. Most commercial RISC-V chips remain in-order processors, meaning they execute instructions sequentially rather than optimizing their order for performance. This architectural simplicity creates a fundamental performance ceiling that’s difficult to overcome without significant architectural changes.

RISC-V is Great for Embedded Systems
But even with slow performance, RISC-V is gaining significant traction in embedded applications where performance isn’t the primary concern. In these contexts, cost and power efficiency are mort important than raw speed. A great example is Espressif’s SoC lineup, which has largely moved to RISC-V. They demonstrate that RISC-V is ideal for IoT devices and other embedded systems where low power consumption is critical.

Current Path to High Performance
Several initiatives are working to address RISC-V’s performance limitations. The Berkeley SonicBOOM project has developed an experimental core featuring out-of-order execution and more advanced architectural features, though it hasn’t yet reached mainstream usage. Similarly, Tenstorrent, a company founded by legendary chip designer Jim Keller (known for his work on AMD’s Zen architecture), is working on high-performance RISC-V processors. While we haven’t yet seen these chips in mainstream benchmarks, Tenstorrent’s involvement suggests serious effort toward making RISC-V competitive in high-performance computing.

ARM’s Path as a Historical Reference
ARM’s trajectory provides an interesting context for understanding RISC-V’s current position. ARM began in low-power embedded systems, initially facing similar performance limitations. Through years of architectural improvements and ecosystem development, ARM gradually expanded into mobile devices, then servers, and now even high-performance desktop systems. Today, ARM’s success has contributed to Intel’s market plateau, demonstrating how an initially limited architecture can evolve to challenge established players. However, it’s important to note that this evolution took decades of sustained development and investment.

Looking at today’s landscape, if you’re building something that needs raw performance, RISC-V simply isn’t ready. While there are promising developments in research labs and startups, the current generation of commercially available RISC-V processors remains significantly behind modern x86 and ARM implementations in terms of raw computational power.

Discussed on Hacker News here.