It’s easy to get lost in the buzzwords when talking about blockchain technology. We hear about decentralization, security, and of course, speed. But what does “speed” actually mean in the context of transactions per second (TPS), and how do different blockchains stack up? It’s a question that’s crucial for understanding which networks are truly ready for mass adoption.
Recently, the Telos Blockchain Network, powered by AntelopeIO, made some waves with some pretty astounding results. They unveiled their tEVM, an Ethereum Virtual Machine that runs on their native network. While they initially estimated it could handle between 4,000 and 6,000 TPS, recent, rigorous testing painted an even more impressive picture.
To get to the bottom of it, Telos partnered with The Blockchain Collaborative (TBC) at Baylor University. This student-run organization, focused on educating and inspiring future blockchain leaders, conducted unbiased testing. Picture this: a team of students, alongside Telos engineers, configuring multiple EVM cannons to bombard the network with transactions. The outcome? A record-setting 15,200 TPS for simple token transactions (send/receive) on the tEVM. For swap transactions, they observed 1,826 TPS. These numbers are significant, especially when you consider they were achieved without relying on complex scaling solutions like sharding or rollups. It’s a testament to the underlying architecture.
To put this into perspective, let’s consider other networks. While the reference material doesn't provide a direct comparison table, it mentions Ethereum’s average block time is around 12 seconds. Telos, on the other hand, boasts block times of just 0.5 seconds. This fundamental difference in block production speed directly impacts how many transactions can be processed within a given timeframe. Imagine a highway: if one has 0.5-second intervals between cars passing a point, and another has 12-second intervals, the first one is clearly going to move a lot more traffic.
What’s particularly interesting is how these tests were conducted. They used a “benchnet” environment that closely simulated the Telos Mainnet, employing standard hardware and code. Crucially, they adjusted configuration variables like max_block_net_usage and max_block_cpu_usage. These settings essentially define the capacity of each block. The benchnet was configured with much higher limits than the mainnet, which is intentionally set lower to prioritize stability over peak performance. This is a common practice; networks often dial back peak performance on their live mainnet to ensure reliability for everyday users. Enabling “overclocking” for the WASM VM and adjusting CPU effort were also key to pushing the limits.
So, when we talk about blockchain transactions per second, it’s not just a single, static number. It’s a dynamic figure influenced by network architecture, block times, and specific transaction types. The Telos results highlight the potential for high throughput, especially when networks are optimized for performance. It’s a reminder that the blockchain space is constantly evolving, with innovation pushing the boundaries of what we thought was possible.
