DePIN: Reheating Old Rice—Immature Traditional Technology + Immature Blockchain ≠ Low-Cost Myth
TechFlow Selected TechFlow Selected
DePIN: Reheating Old Rice—Immature Traditional Technology + Immature Blockchain ≠ Low-Cost Myth
From a technical standpoint, DePIN currently lacks feasibility and usability; it might become viable in the future, but that prospect remains distant.
Navigating Web3 tides with focused insightsInterview | Beichen
Guest | Steven
The Web3 industry is currently in a very awkward phase — the market is turning sluggish as liquidity tightens, while no new technological paradigm has emerged. As a result, old ideas are being reheated, such as DePIN (Decentralized Physical Infrastructure Networks).
In fact, various blockchain-based hardware infrastructure projects have been emerging since 2017. While none has truly succeeded yet (we can only say none has completely failed — projects like Filecoin and Arweave still hold promise), enthusiasm today is surging. Some even claim that DePIN could cut costs by over 75%.
Last month, “Loud Whistle” published an article titled "Why Helium Was Fundamentally Flawed from the Start, and on Web3 Research Methodology", using Helium — a flagship DePIN project — as a case study to analyze its commercial shortcomings.
The key arguments can be summarized in two points:
1. Helium’s blockchain merely serves token issuance and transaction settlement — it doesn’t require a public blockchain at all;
2. The IoT services provided by Helium run counter to real-world trends in the IoT industry (as manufacturers increasingly prefer building their own closed ecosystems).
For this episode, we invited communications technology expert Steven to explain, from an engineering perspective, why blockchain is unsuitable for hardware infrastructure.
1. Beichen: DePIN is extremely hyped right now, with claims it could drive a new wave of physical infrastructure. But beyond business logic, is it even technically feasible? Could you start with a bold take on DePIN from a telecommunications engineering standpoint?
Steven: From a technical standpoint, DePIN currently lacks feasibility and practicality. It might become viable someday, but that day is far off.
2. Beichen: Why do you think the blockchain industry is revisiting DePIN now? To me, this seems like the same "blockchain transformation" narrative that collapsed in 2018. My personal bias is that people turn to blockchain integration only when they’ve failed in traditional industries.
Steven: DePIN isn’t quite the same as "blockchain transformation." The idea behind DePIN was actually proposed before blockchain even existed — for instance, in telecom, there were attempts at decentralized Wi-Fi or CDN networks. Now, with blockchain, fundraising is easier and financial incentives are stronger, so these ideas naturally get wrapped into blockchain narratives — what we now call DePIN. But current technology still can't support DePIN’s ambitions.
3. Beichen: Since DePIN spans many sectors, let’s begin with the area you know best: DeWi (Decentralized Wireless). Take Helium — do its LoRaWAN and 5G networks actually have any competitive edge?
Steven: From a communications engineering perspective, LoRa supports only a very limited range of IoT applications. And by adding blockchain, Helium makes its system prohibitively expensive — it's simply not viable for commercial use. Its 5G offering is just a supplement to existing telecom operators, essentially acting as a 5G reseller, which holds little significance.
Despite Helium promoting numerous use cases, the reality is that LoRa can serve only a narrow set of IoT scenarios.
Wireless IoT physical layers fall into two broad categories based on coverage: long-range (hundreds of meters to tens of kilometers) and short-range (within hundreds of meters). Familiar examples like Wi-Fi and Bluetooth are short-range. Helium’s LoRa belongs to LPWAN (Low-Power Wide-Area Network), a type of ultra-narrowband IoT protocol requiring minimal bandwidth. The trade-off is extremely low data throughput, restricting it to lightweight nodes and transmissions — such as sensors like thermometers or water meters that report data every few minutes with payloads of just dozens of bytes.
Moreover, LoRa operates on unlicensed frequency bands that are free to use, but vary by country. In Europe and the U.S., it uses prime bands like 433, 868, and 915 MHz (ideal wavelengths with decent penetration and acceptable speeds — perfect for voice and low-to-medium-speed data). In China, it uses 470–510 MHz.
(Beichen: So must Helium’s mining devices be adjusted accordingly? Steven: Yes, both frequency band and transmission power must be properly configured, otherwise signals will interfere.)
Although LoRa uses unlicensed spectrum, allowing any company to build its own network (like setting up a home Wi-Fi), each frequency band has capacity limits. For example, 5G theoretically supports up to one million connected devices per square kilometer, whereas LoRa, based on our calculations, handles only a few hundred to about 2,000 devices in practice.
Helium uses gateways (miners) to perform coverage-based proof-of-work, meaning it rewards network coverage regardless of actual demand — even if it’s in uninhabited areas. Real-world IoT networks optimize node placement based on actual usage needs.
IoT systems require ongoing operations and maintenance, with failures needing rapid resolution. Traditional IoT networks invest heavily in dedicated teams for maintenance. In contrast, Helium miners lack professional capabilities — and their devices are widely dispersed, making reliable maintenance nearly impossible.
4. Beichen: After integrating blockchain, Helium claims to reduce costs. Yet you argue it actually increases them. How so?
Steven: I strongly disagree with the claim that Helium reduces costs.
First, Helium’s network deployment cost is high. I once disassembled a Helium miner — it was quite expensive, nearly 10,000 RMB. When I opened it up, I found that the components, including chips, could be replicated on the open market for around 100 RMB. The only difference was a so-called “custom chip” with an embedded ID for anti-counterfeiting.
Second, the distributed architecture inherently raises data transmission costs. Each Helium miner requires internet connectivity, monitoring tools, and operational oversight. Traditional IoT systems also need internet interfaces and monitoring, but IoT and internet layers are strictly isolated, enhancing security. They don’t require extensive individual device monitoring — for example, Alibaba Cloud manages hundreds of thousands of sensors with just a few servers and platforms, spreading out costs efficiently.
LoRa networks don’t need blockchain to function. Helium added blockchain purely for fundraising, artificially introducing verification mechanisms that add unnecessary overhead.
Strip away the blockchain layer, and you’ll see that Helium’s core business is fundamentally centralized. It merely moved automated transactions onto a blockchain — making blockchain a completely artificial requirement for Helium.
5. Beichen: If Helium’s model doesn’t work, what kind of solution does IoT actually need?
Steven: Industrial enterprises — especially those engaged in continuous or large-scale discrete manufacturing — demand extremely high reliability in data transmission. They prefer building private networks, often implementing electronic fencing and other isolation measures to ensure all data stays within internal networks. Only low-security applications would use cloud IoT services like AWS or Alibaba Cloud, or technologies like NB-IoT. They would never adopt third-party LoRa networks like Helium — not even for something as simple as streetlight control.
5G is a key direction for IoT, capable of supporting massive numbers of sensors or controllers within a small area. LoRa networks are unsuitable for industrial or critical infrastructure use — especially Helium’s version, which combines higher cost with lower efficiency due to its distributed design.
6. Beichen: There are also concepts like decentralized storage and decentralized computing. Since storage and compute are genuine market needs, do they really offer advantages over centralized solutions?
Steven: Whether decentralized storage and computing offer benefits depends entirely on the problem they’re solving — decentralized systems inevitably sacrifice efficiency and increase costs. However, they make sense when the application demands multi-node connectivity — such as higher redundancy, reliability requirements, or compliance with legal restrictions.
So yes, there is demand for decentralized storage and computing — no doubt about it. But how big is that demand? And can it replace centralized systems? That remains questionable.
Take Filecoin as an example. It’s a solid project that solved many technical challenges, particularly around Proof-of-Spacetime and Proof-of-Replication. Honestly, developing it was no easy feat. But Filecoin’s fundamental economic model is flawed, triggering a cascade of further issues.
First, FIL is a utility token whose price volatility directly impacts real storage demand. Ideally, participant incentives shouldn’t be tied to token market fluctuations.
This flawed economic model creates significant sunk costs. Miners are afraid to use hard drives because any disruption (power outage, internet failure) won’t just prevent earnings — it risks losing staked FIL. This contradicts Filecoin’s original vision of utilizing idle storage space. Worse, Filecoin miners need GPUs to verify storage validity (via replication proofs), leading to actual storage costs that exceed even cloud storage — let alone basic hard drive expenses.
High costs further undermine usability — no enterprise will store mission-critical data on Filecoin’s network. This isn’t something blockchain incentives can fix.
7. Beichen: In your view, what technical hurdles must a mature “blockchain-based decentralized network architecture” overcome? Or are they fundamentally unsolvable?
Steven: DePIN refers to blockchain-based decentralized network architectures. Blockchain relies on triple-entry accounting and consensus across multiple nodes, which inherently makes it less efficient and more costly than traditional digital systems.
For DePIN to progress toward its ideal form, three major problems must be solved.
First, infrastructure reliability. Distributed networks impose higher maintenance demands on nodes than centralized systems. Miners would need strong technical expertise — but current hardware and software maturity are far from sufficient.
Second, decoupling of hardware and software. Dominant centralized firms often bundle software and hardware together (like Apple’s Mac ecosystem). Only when hardware and software are sufficiently decoupled — enabling software-defined everything — can DePIN’s vision of decentralized, distributed deployment become realistic.
Third, trustworthy mapping of real-world data. A reliable bridge between the digital world and the physical world must be established. For blockchains, this means oracles must mature significantly.
(Beichen: This is also the foundation for RWA. Steven: More accurately, it’s the foundation for whether blockchains have a future at all. Centralized systems rely on social contracts among people; blockchains rely on code and probabilistic game theory. If input data is unreliable, how can any vision stand?)
Beyond technical bottlenecks, DePIN also faces major challenges in community efficiency — a huge issue.
Take 5G: the latest standard, Release 18, runs over 35 million words in Chinese alone. Then, different vendors derive even more detailed product specifications from it. From R&D to deployment, it involves massive coordinated effort.
Given the current efficiency of DePIN communities, achieving anything like this is impossible. They can only wait for centralized organizations to do the heavy lifting and then consume the results. So how meaningful is decentralization then? At best, blockchain lowers the barrier to entry — enabling experiments that couldn’t happen in traditional systems. That’s valuable, but it’s not true technological innovation.
8. Beichen: That point really hit home. DePIN deals with off-chain domains, so communities must import external, centralized-world technologies. Only native crypto areas like DeFi have the potential to generate new innovations organically within the community.
Steven: First, DeFi aligns naturally with blockchain’s strengths because finance is inherently rules-based. Second, DeFi implements engineering through code, while DePIN must interact with complex physical devices — the difficulty levels aren’t comparable.
9. Beichen: Finally, how do you view the lifecycle of the DePIN narrative? Or put differently, how long before the market realizes it’s unworkable?
Steven: Technologically speaking, neither blockchain nor the legacy technologies involved in DePIN are mature enough. Attempting such ambitious projects with today’s technology has almost no chance of success. Only teams with real technical depth — like Filecoin — might see progress after five to ten more years of sustained effort.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
