Most Depin projects even use blockchain: true or false?

DePIN – short for Decentralized Physical Infrastructure Network – is one of the new narratives in crypto. However, the extent to which current DPin projects rely on blockchain networks is an open question.

Dubbed the “Internet of Things (IoT) with a blockchain twist,” DePIN projects connect and manage physical devices that collect and exchange data, such as wireless networks, energy grids or storage systems.

These networks generate huge amounts of data, so connecting DePIN to the blockchain requires a huge amount of chains – and even then, the blockchain can only record a small portion of the data.

One of the most famous DPin projects connecting wireless networks, Helium, has dropped its own chain to move to Solana in April 2023, one of the fastest options currently available. However, even Solana's efficiency falls short for DePIN projects that require real-time computing capabilities.

Are DePIN projects really on-chain?

“The current use cases of DP are not really fully onchain,” Shuyao Kong, co-founder of the soon-to-be-launched MegaETH, told the magazine. Ethereum L2 aims to process 100,000 transactions per second, which may be enough for some DePIN projects.

So, is it just to attract venture capitalists that the Deepin projects are branding themselves with “blockchain” as a marketing method?

Some may be. But the bigger issue is that currently fully onchain DePIN applications are inefficient and costly. Therefore, projects use a hybrid structure by performing computations off-chain and recording results on-chain.

In a recent podcast, Kong spoke to Solana's co-founder about a fitness tracker that requires real-time biometric data calculation, which Solana can't do. Kong says the upcoming Ethereum L2 MegaETH will have that capability.

Why Deepin Projects Rely on Off-Chain Computing

Zhe Wang, CEO of the DEPHY network – a project that connects DEPHY data to blockchains – divides IoT into two main categories: sensors and enforcers.

Sensors collect data from physical environments and don't necessarily need to deliver instant messages, think weather sensors or pedometers.

Devices that work on command are more tolerant of sensor delays than processors.

Delays in the calculations of essential functions such as driving a car or controlling lights can cause discomfort or even danger.

“If you ask to turn off the lights in your bedroom, it will be very bad [and] It's within 10 minutes,” Wang told the magazine, “but it's best if your step count is calculated in a day.

Although sensor data can be easily transmitted over a chain, continuous or large-scale data transmission can be expensive and inefficient, so it is still not ideal.

Why use blockchain for DePIN?

So if both sensor and execution data are best calculated off-chain, then what purpose does blockchain even serve for DePIN?

Wes Levitt, head of strategy at Theta Labs, said: “Deep's projects use blockchain because it's a way to coordinate thousands, if not millions, of devices without permission and in a way that doesn't require each device to trust each other.”

Although computation is often required for DePIN use cases, it is typically handled separately from onchain registration.

As the name suggests, DPin aims to decentralize physical infrastructure, requiring participants to host devices, provide bandwidth, or share data. But despite its advantages, blockchain is not really needed to run a decentralized infrastructure. Tor, a decentralized alternative internet with thousands of nodes, is one of the most popular examples worldwide.

Tor also credits volunteers motivated by common interests such as privacy advocacy and technical testing.

Blockchain adds a layer of financial incentives to participation by rewarding contributors with tokens.

For example, DePin marketplaces like CUDOS and Akash allow users to donate computing power, including GPUs, which can be rented out for AI services or other high-demand tasks. This allows contributors to earn tokens in return for sharing their resources, creating a more economically sustainable system.

“That's what's being done with our Theta EdgeCloud platform and many other Deepin projects,” says Levitt. EdgeCloud nodes typically perform AI processing or video editing tasks, which can take nodes from a few minutes to several hours to complete.

Levitt argued that sending partially completed jobs down the chain every few seconds was not worth it. Instead, nodes only need to submit confirmations for authentication and payment completion.

“In most cases, it's sufficient to send a node's proof-of-work or coverage at regular intervals, perhaps with some guarantee to ensure compliance between verifications,” he said.

Best and worst blockchain networks for DePIN projects

Ethereum is the biggest blockchain that can handle DePIN, but L1 is not fully suitable due to slow TPS. Layer 2 solutions, like Base and Arbitrum, can handle more transactions, but these are still well into the hundreds per second.

But in the year 2025 could be the year blockchains take their big leap forward to scale.

MegaETH – short for “Ethereum Great Again” – is an L2 chain backed by Vitalik Buttery and uses an improved algorithm to provide scalability while maintaining Ethereum's security. Despite the promise of 100,000 TPS, it is yet to begin.

Based on Facebook's Move language, Su entered the competition with a theoretical maximum of about 300,000 transactions per second — even surpassing MegaEth's recent efficiency.

However, Sui's highest recorded average TPS in October was 822, with a low of 106 TPS on October 16. (That said, TPS depends on interest and ability, and real-world TPS is always much lower than testnet.)

Solana remains one of the fastest layer-1 blockchains, theoretically maxing out at 65,000 TPS, but mostly averaging around 3,000 TPS, according to Solscan. An upcoming Firedancer client is currently looking for 1 million TPS on the testnet.

But Wang argues that even such a staggering output rate would not be enough to host a DPin project to run entirely on blockchain.

He said that even a blockchain running at 1 million TPS would not be enough to handle the massive amount of computing power DePIN would require.

According to DePIN Scan, WiFi Map has more than 14 million devices. This means that even if a device sends one message every 10 seconds (this is a very generous example for IoT real-time computing examples), that blockchain would require 1.4 million transactions per second for that single project.

At 1.4 million transactions per second and a gas fee of $0.024 per transaction (Solana's typical fees), the project would cost $33,600 per second.

“I don't think full DePIN projects will be formalized in the near future. The real-time computing demands and high data throughput required by many DePin applications outstrips current blockchain architectures fully on-chain, Jeremy Frank, head of engineering at Autonomys Network, a distributed storage project powered by blockchain and AI, told the magazine.

“That said, I expect we'll see more and more DePIN functionality move on-chain as scalability improves.”

The Future of DePIN: Developing Block Chains and Specialized Architectures

Currently, no blockchain can support true real-time transactions for DPin projects, Wang explained.

But real-time computing may not be necessary for all of DePin's applications, and it's not the case that these projects pitch themselves on blockchain as a marketing ploy to attract venture capitalists.

Deepin projects often split their architecture between onchain and offchain components to improve scalability and performance, while the onchain components provide incentives and trustless authentication.

Ultimately, as DePIN pushes the boundaries of blockchain technology, its success depends on finding a balance between onchain and offchain processes.

Frank believes that future DPin projects will move towards modular and specialized architectures that can handle the different computational needs of different applications.

“We see purpose-built execution environments for specific DPin verticals, whether that's real-time biometric processing, large-scale data storage or high-frequency sensor networks.”

John Yun

Yohan Yun is a multimedia journalist who has been reporting on blockchain since 2017. He has contributed to the crypto media outlet Forkast as an editor and covered Asian tech stories as an assistant reporter for Bloomberg BNA and Forbes. He spends his free time cooking and experimenting with new recipes.