A comprehensive analysis of the evolution trends of the five core tracks in blockchain

Written by: Lao Bai, “Ten Thousand Words Long Article - The Technological Trends Evolving in Various Tracks of Blockchain in My Eyes”

Let’s talk about what is happening in various sectors recently, as well as the trends that may emerge in the next 6-12 months.

Overall, it will be divided into several sectors: public chains, DeFi (Dex, lending, Stablecoin, synthetic assets, etc.), NFT, GameFi, and storage.

I. First, let’s talk about public chains

The new trends in public chains mainly fall into three categories: ETH + L2, Move + Diem gang, and modularization.

1. ETH + L2

ETH, the sequence for the next few months to years is as follows:

· Merge (expected between September and December)

· EIP4488 (expected by the end of this year or early next year) - Reducing the cost of CallData, lowering the cost of L2 by several times

· EIP4844 - Proto-Danksharding (next year or the year after) - Introducing a new transaction type Blob to replace CallData, further reducing Rollup costs

· Danksharding - (2-4 years later) - Equivalent to Proto-Danksharding + sharding

ETH is great in many ways, but it’s just slow…

2. L2

OP System

Arb - Fully aware of the current pain points of L2, which are - still not cheap enough, not fast enough (compared to Alt L1 like Solana or sidechains like Polygon), so while preparing for the Nitro mainnet (using WASM architecture to significantly improve performance and reduce costs), they also launched Nova (a sidechain based on AnyTrust, but with Rollup as backup, making it much safer than typical sidechains).

Op - The biggest highlight is the token issuance, and the recently released BedRock architecture, which makes multi-client support possible in the future - not just supporting OP clients, but also ZK.

Fuel - Created a Swayswap using a new language called Sway, and in the testnet demo, the speed was astonishing; it remains to be seen if this speed can be maintained after the mainnet launch. Notably, Fuel uses the UTXO model, which is known for its performance advantages in concurrent processing compared to the Account model, but due to the lack of shared state, it is not suitable for complex contract designs. Fuel addresses this issue using UTXO ID + Strict Access List technology.

ZK System

ZK-Sync - Announced plans to launch the mainnet within 100 days. It’s worth noting that many DApps on ZK-Sync are native, not just some DApps ported over.

Polygon - Announced ZK-EVM.

Starkware - Also announced expectations for token issuance, with the most anticipated feature being recursive Rollup, which can further branch into L3 or even L4, achieving higher TPS and lower costs.

Scroll - Just announced Pre-Alpha, claiming to be the most EVM-compatible ZK-Rollup.

Aztec - Created Aztec Connect (privacy as a service), finally providing the privacy I wanted to see, allowing existing ETH L1 DApps to connect to Aztec through bridging contracts and SDKs for privacy protection. There’s no liquidity fragmentation, and no need for redeployment.

Although the current implementation is quite rudimentary, not all types of DApps can connect, and it heavily depends on the team’s PR and collaboration capabilities, this is at least the kind of privacy I envision. I never thought we needed privacy coins or a separate privacy public chain; what we need is a privacy plugin!

Finally, two points about ZK and concurrent processing.

I just saw a tweet from Min Dao teacher, stating that a large number of ZK systems will go live in the coming months, and then? It will surely be the same old DeFi trifecta: dex, lending, Stablecoin… The public chain’s native tokens + various DEX and lending projects will likely see dozens of new tokens for everyone to pick up, and worse, the further fragmentation of liquidity, which currently has no solution.

LRC’s collaboration with Starkware on DAMM might be a solution, but first, this thing can only integrate liquidity between ZK Rollup and ETH L1; second, no one knows when it will go live; third, the complexity of this solution is very high. The MEV issue on a single chain is already a problem, and now integrating multiple chains increases the asynchronicity, MEV, impermanent loss, and these issues will become even more complex. I’m really overwhelmed…

Regarding parallel processing - this is somewhat related to the Move and Diem gang mentioned later.

A bottleneck limitation of EVM performance is that even with high TPS, it is still serial. For example, if A sends different tokens to B and C at the same time, it can be processed in parallel on Solana (of course, different objects of the same token cannot be processed in parallel), but on EVM, it can only be processed serially, regardless of L1 or L2, as long as it is EVM architecture, parallel processing is not possible.

This is a natural flaw of EVM. To overcome it, a new VM must be designed. Arb, Op, ZK-sync, and Starkware still inherit this EVM issue. So theoretically, the strongest Rollup might be a ZK-Rollup that resembles Solana’s architecture but is built on ETH? (Overcoming Solana’s security and stability issues while achieving Solana’s TPS and parallel processing)

The only L2 exploring parallel processing is Fuel mentioned above. Although Aztec also uses the UTXO model, it seems that its UTXO design is mainly for privacy considerations rather than parallel processing and TPS.

3. Move & Diem Gang + Modularization

I initially thought that the launch of ICP had closed the door on public chains, as it takes several years for a public chain to develop and flourish, and the current time window is insufficient. However, under the magic of capital, this is not an issue…

Recently, a new article on the top ten public chains has been circulating, predicting that more than ten new L1s will emerge in the next year, along with eight to ten new L2s, resulting in dozens of L1 + L2s, leading to liquidity fragmentation… Anyway, the most popular new public chains are the two from the Move language + Diem team, Aptos + Sui.

The two are basically identical, both emerging from the original Diem team, both using the new Move language (Sui modified it a bit, but the core is still Move), both claiming 100,000 TPS+, backed by A16Z and Coinbase Ventures, both valued at $2 billion, and both focusing on parallel processing… The difference is that Aptos has a fast state synchronization and account recovery technology, while Sui treats Object rather than Account as the basic unit and introduces a pay-for-state-storage model.

How to put it, saying there’s no innovation would be false, but the degree of innovation seems far less than Solana (Solana is prone to downtime, but its consensus and propagation mechanisms are indeed unique innovations, achieving the extreme of single-chain TPS). These two launched with a $2 billion valuation, feeling even more “capital-chain” than Solana. I wonder how much profit can still be made in the secondary market.

In fact, another project from Diem is Linera, which uses Rust instead of Move. This feels like a state-owned enterprise has gone bankrupt, and everyone is pulling out their own equipment to start their own businesses… But no matter what, you cannot ignore them, especially Aptos, which, under the hype of various media, gives a sense of the next bull market akin to Solana. While focusing on technology, you also need to make money trading tokens.

Then there’s the modularization thing, which was basically popularized by Celestia. Originally, Polkadot and Ethereum 2.0 could also be considered modular, but Celestia has split the two-layer structure into three layers, making this “modular” feel much more pronounced. Moreover, with a separate DA layer, many combinations can be played out, such as various arrangements of different execution layers + different settlement layers, with the most important being Sovereign Rollup.

According to Celestia, the ultimate Layer 0 of a chain is actually Social Consensus. Although Ethereum’s Rollup is secure, it is essentially an extension chain of ETH, losing sovereignty, more like a state in the United States. If you want to build your own country, you need to be able to flexibly adjust and fork at any time; you still need your own sovereignty. The first sovereign Rollup should be Celo.

As for which approach, focusing on sovereignty or Ethereum’s focus on security, will prevail or coexist in the long term, it’s too difficult to judge now; we’ll have to wait for the market to take two years to tell us the answer.

Besides Celestia, there’s also a Dymension team in Cosmos developing RDK, Rollups Development Kit - a Rollup development kit, which is based on the Cosmos SDK and ecosystem, somewhat reminiscent of Polygon Edge. Coupled with the recently promoted CosmWasm virtual machine, Cosmos is currently definitely at the forefront of modularization.

On the Polkadot side, the recently launched XCM and governance V2 are points worth noting, but the current heat and prices of parachain auctions are quite low. It feels like unless a blockbuster chain or application emerges, it will be difficult to capture the current market’s attention.

As for one-click chain issuance, the number of players is increasing. Besides Cosmos and Polkadot, Avalanche recently joined (command line can issue a chain in 42 seconds, terrifying…), and now there are BAS (BSC sidechain) and Polygon Supernet. The infrastructure for issuing chains in the market has far exceeded the demand for issuing chains, creating a déjà vu of 2019 when trading platforms outnumbered users…

II. DeFi

1. Dex

I think the market has finally recognized that impermanent loss is a feature, not a problem.

It’s just that, you earn transaction fees, and if there’s no impermanent loss, wouldn’t that mean you’re just eating meat without getting hit?

Where’s the good deal in that?

Many LPs who wanted to eat meat without getting hit have fallen victim to Bancor this time; it was said that many LPs still couldn’t withdraw, and I don’t know how it is now.

Uniswap has become a true infrastructure.

Besides long-tail assets, mainstream tokens have mostly moved to V3, and there are no longer V2 trading pairs. More importantly, we see more projects using V3 as infrastructure starting to implement. These are not just simple composability like using Uniswap’s LP for liquidity mining, but rather utilizing V3 to achieve more complex functional projects.

For example, Perp V2 and Rage Trade are both leveraging V3 to implement leverage and more complex derivative trading, providing LPs with more diverse choices (they can choose higher yields, of course, but also bear greater risks).

This is somewhat similar to the path of Web2, where early applications were based on operating systems like Windows, and after the emergence of super applications like WeChat and Chrome, many apps began to be built on these super application platforms, such as various WeChat mini-programs and Chrome extensions.

You will see similar scenarios in the upcoming lending space.

2. Lending

Asset isolation pools have become standard. The recently launched Euler and AAVE V3 have introduced this concept to avoid a single bad asset or pool damaging the entire platform.

Multi-chain deployment - represented by AAVE V3’s portal and the multi-chain protocol code released by Compound V3. Full-chain DEX currently faces numerous challenges, while full-chain lending seems much easier.

Morpho Labs - This is a peer-to-peer lending matching protocol based on Compound and AAVE, which has launched a peer-to-peer lending feature on the two major pool lending platforms, allowing eligible users to enjoy better interest rates and higher capital efficiency. This is similar to the aforementioned scenario based on Uniswap for derivative projects.

As for permissionless and credit lending, to be honest, I’m not optimistic about them at the moment. Permissionless lending is meaningless at this stage because 99% of tokens do not qualify as collateral for lending. As for credit lending… we’ll talk about it when the DID system matures. It could even be a pessimistic view that blockchain-based credit lending might just be a false proposition.

3. Derivatives

There still hasn’t been a “Uniswap moment” for on-chain derivatives. Besides DYDX relying on mining to support liquidity and GMX relying on innovative trading volume, derivatives in the DeFi circle are still being crushed by CEX.

The trends worth watching are as follows:

1. DYDX V4 - What will a sovereign perpetual platform based on Cosmos be like compared to the Starkware era?

2. Will derivatives like Rage Trade, based on Uniswap V3, have better depth and experience?

3. There will be more derivatives mimicking the GMX platform, allowing users to choose to be LPs, acting as counterparties to all traders (traders tend to lose money on average, so LPs always profit).

4. Synthetix V3 and TracerDAO, theoretically, as long as the oracle supports it, can create synthetic financial derivatives or synthetic leveraged tokens for everything in the world (slightly exaggerated).

Originally, there was also an Overlay Protocol, similar to the above two but more aggressive, utilizing the AMPL rebase system to eliminate the need for counterparties. However, after following it for a year and a half, it’s still in the most primitive testnet stage, with progress moving at a snail's pace; I almost wanted to unfollow it.

4. Stablecoin

After Luna, it can be said that stablecoins are basically dead, with only Frax, a “semi-stable” coin, still struggling to hold on. And it’s important to note that a large proportion of Frax’s components are USDC, and DAI is also…

So basically, stablecoins have returned to the era of pure dollar stablecoins. Although USDC’s issuance is still slightly lower than USDT, its status in the market is actually the top. The only stablecoin worth looking forward to is the over-collateralized stablecoin that AAVE and Curve claim to be developing. I wonder if these two will thicken the leverage in DeFi once they come out.

5. RWA (Real World Asset)

This could have been the most important aspect, but it is also the most overlooked. Everyone involved in DeFi has a dream of going mainstream, but when it comes to action, they find that ideals are abundant, but reality is thin. The only notable event in the past two years is last year’s collaboration between MakerDAO and Centrifuge for real asset-backed DAI loans, and the recent case of using 500 million DAI to buy government bonds and corporate bonds. Is there anything else?

To summarize, the trends in DeFi can be described as follows:

1. Uniswap, AAVE, Curve, and other first-generation DeFi have nominally become true infrastructure. Last year’s so-called DeFi 2.0 has basically been debunked. The real next wave of DeFi 2.0 may be projects like Rage Trade and Morpho Labs, which are built on first-generation DeFi as “stacked” projects, or Synthetix V3 and TracerDAO, which can reach off-chain and external worlds through synthetic assets.

2. From AAVE and Curve’s plans to create stablecoins, it’s clear that leading DeFi projects are building stacks. We may see super applications that surpass public chains in the next round.

3. RWA remains lukewarm, and it can even be said that it is ignored. Will it ultimately be like NFTs, where burning famous paintings to put them on-chain is a false proposition, and that native IPs like monkeys going off-chain is the right path? So the idea of bringing off-chain assets on-chain is wrong (oh, I’m taking a hit here…), while bringing on-chain native assets out of the chain through DeFi or some form is the right way?

III. NFT and Storage

The trends in NFTs can be summarized into three main points:

1. Application Scenarios

Currently, NFTs seem to have many application scenarios, including digital art, tickets, domain names (ENS), game items, Poap, etc. Even the LP of Uni V3 itself is an NFT… But the only thing that can be talked about right now is profile pictures, and to create a profile picture, the only way out is to turn it into a brand or IP like the monkeys, followed by secondary creations and narratives like the metaverse. So you will see that all NFT projects making profile pictures will paint a picture of the metaverse, but few truly have the ability to realize it or create value; in short, it’s very competitive.

Aside from profile pictures, I personally see the most potential in SBT and game items after GameFi enters its 2.0 era.

2. Financialization

The financialization of NFTs is basically an irreversible trend. I previously tweeted about this, with various oracles, peer-to-peer lending, and NFT + DeFi projects emerging one after another, essentially all aimed at solving the liquidity problem of NFTs.

The biggest current issue in financialization is: there are too few blue-chip NFTs, yet too many NFTFI projects. Dozens of projects are basically competing for less than ten blue-chip NFT projects, with a target user ceiling of only a few tens of thousands, a classic case of too many wolves and too little meat.

There’s an even bigger problem: based on point 1, the future explosion of NFTs is expected to be in SBT and GameFi items. However, SBTs are non-tradable, and you cannot build FI-like products around them. GameFi items do represent a large market, but based on Web2 experience, when a GameFi project becomes popular, it is highly likely to establish its own DEX or FI-related market, rather than handing over this lucrative segment. For example, Dota, CSGO, and Dream of Red Mansions… Web3’s Stepn initially used Orca for DEX, which led to a surge in Orca’s trading volume, but later they simply created their own DEX, causing Orca’s trading volume to plummet.

So, the currently seemingly popular NFTFI projects may have limited growth potential, as they mainly serve the top few NFT profile picture users. Of course, if I’m wrong about point 1, then this point also doesn’t hold.

The most noteworthy project in terms of financialization recently is Sudoswap, which is somewhat like a more streamlined version of NFTX + Sushiswap, turning NFT buying and selling and market-making into an experience similar to Uniswap’s AMM. I recommend everyone to give it a try, especially to study its LP mechanism, which is quite creative. Although the mechanism is a bit more complex and cumbersome than Uniswap, I really can’t think of a better design solution at the moment.

Additionally, the recently confirmed ERC4907 - Rentable NFTs protocol theoretically allows for more creative financialization of NFTs, but it seems that the biggest application adaptation scenario is still in the next generation of GameFi.

3. Market Competition

Uniswap acquired Genie and claims to integrate Sudoswap, which means it has extended its reach into the NFT field. Opensea acquired Gem and created a set of open-source protocols called Seaport. X2Y2 and Looksrare have been trying to poach users from Opensea through listing and trading rewards. So the landscape seems to be Uniswap + Genie + Sudoswap vs. Opensea + Gem vs. Looksrare + X2Y2 vs. new projects developed based on Seaport?

Finally, a quick note on storage

There’s not much to say about storage; I currently see one major trend, which is that storage is beginning to cross over and extend its reach into computing.

The reason is simple: storage isn’t profitable, at least at this stage of blockchain development, storage isn’t making much money. The core value of the entire blockchain is still oriented towards a trustless mechanism based on “decentralized computation,” rather than “decentralized storage.”

This has led many projects aiming to create decentralized cloud storage to have little business. People still use Amazon if they need it, and Dropbox if they need that. Besides hearing about a batch of small images stored on AR, what significant business is there? ETH has used a lot of IPFS, but Filecoin is still primarily about packaging garbage data for “mining.”

Currently, several major storage projects are basically expanding towards computing.

On AR, a SCP (Storage-based Consensus Paradigm) based on Smartweave has been developed, with Everpay being a representative. The SCP paradigm focuses on off-chain computation and client verification for scalability, theoretically breaking performance bottlenecks completely. Bitcoin’s RGB protocol, BSV’s large blocks, and ETH’s Truebit are somewhat similar to this paradigm. Whether this “non-full-node computation verification” can ultimately gain market recognition remains to be seen, as it differs from traditional blockchain routes.

Filecoin also plans to create a virtual machine, claiming to develop FVM (Filecoin-based virtual machine), but it’s unclear when it will be completed.

Other well-known storage projects like Ceramic (more like a blockchain database system) and Subspace are also moving towards a model that includes storage + computing, rather than being 100% storage projects.

IV. GameFi

To be honest, I’m quite conflicted about writing GameFi because I haven’t fully seen or figured out where the next trends in GameFi lie.

First, let’s discuss the current state

Firstly, X2Earn has basically reached a dead end. This Ponzi model starts with significant advantages, then grows exponentially until it hits a bottleneck and begins to struggle, resembling the early liquidity mining that attracted users with high APY. However, because it is cloaked in the guise of GameFi, the timeline can be extended significantly, but its core essence remains a DeFi.

Some say that introducing external revenue is the solution, just like DeFi projects; as long as the project itself has revenue, it can break this Ponzi model. Crv is a prime example. GameFi cannot generate revenue like DeFi projects, but it can introduce external world advertising through a SocialFi-like approach. For instance, some suggest that Stepn could collaborate with Starbucks to create a running route that passes by Starbucks, requiring users to stop there to earn rewards. The advertising fees could serve as external revenue to break the Ponzi.

This idea is sound, but it still doesn’t solve the fundamental problem. The reason is that not everyone in DeFi is mining for profit; it’s those providing liquidity, bearing impermanent loss and risks that are making money. In GameFi’s X2Earn, everyone is making money, and such a world does not exist.

In an extreme hypothetical scenario - let’s assume Stepn grows to the scale of Facebook, with 3 billion users. You would discover a terrifying fact - even if each person earns only one dollar worth of GST daily, that’s a daily selling pressure of 3 billion dollars, amounting to 1 trillion dollars a year. Even if we assume that half of that is consumed in upgrading shoes and gems, that’s still 500 billion dollars…

What was Facebook’s advertising revenue last year? 80 billion dollars.

This means that once Stepn reaches that scale, its advertising profitability would need to be 6-7 times that of Facebook to maintain a daily income of 1 dollar per user - which is clearly unrealistic.

So, what about a daily income of 0.1 dollars?

Ask yourself, would you download another app and go running for an extra 0.1 dollars a day? Even if you just open the app while running?

Clearly, that doesn’t make sense.

So where’s the way out?

The current viewpoints and trends I see are as follows:

1. Leaning towards traditional - Free 2 play, play for fun, and Skill 2 Earn

In simple terms, this means getting closer to traditional models, allowing everyone to participate for free, and then making money becomes “only some players earn.”

The filtering method would be players’ skills, character levels, time and effort invested, understanding of the game, etc. In short, skilled players can earn money.

What about the rest? Naturally, they would just play for fun.

Then the question arises: Steam releases thousands of games every year. If it’s for fun, wouldn’t I enjoy playing on Steam? Why come to Web3 to play GameFi? At least currently, GameFi is significantly lacking in terms of fun compared to Web2 games.

Without high playability, there won’t be a large number of players playing for fun. Without enough players, there won’t be any big spenders, and without big spenders, those skilled players won’t have Skill to Earn… It’s a flywheel. Well-made games like World of Warcraft and Dream of Red Mansions have achieved Skill 2 Earn without needing Web3.

So this path is a direction, but I don’t know when GameFi’s production and playability will align with traditional games. Currently, it seems quite far off…

2. Blockchain Native - Crypto Native

If you think about it, you’ll find that in the long run, native things always have more vitality.

For example, when the internet first emerged, people stopped reading newspapers and turned to the internet for news. That’s not native; it’s “internet reform.” The truly impressive ones are Google, YouTube, Douyin, and WeChat.

When NFTs first appeared, we thought about turning famous paintings into NFTs and then burning the paintings. Later, we realized it should be the other way around; native Punks and monkeys became popular, and offline entities began to connect.

Following this line of thought, the first point above resembles traditional game chain reforms and coin reforms, without leveraging the true native characteristics of blockchain.

So what should GameFi built on blockchain-native characteristics look like? Fomo3D is a great example, though it’s quite rudimentary and leans towards gambling.

Later games like Wolf and Sheep and Dark Forest also contain very “native” elements, as the core logic of the entire game is on-chain and relies on smart contracts, rather than off-chain servers. Therefore, it has significant advantages in terms of open architecture, self-sustainability, and composability.

However, this model also has significant issues. The most direct problem is that the barrier to entry is too high. If Axie is Dota, and Stepn is Honor of Kings, then Wolf and Dark Forest are like the Civilization series, often praised but not popular, appealing to a narrow audience, exclusive to hardcore players.

So if you ask me where the way out is, or what the trend is, I really don’t know. Currently, I see projects like “Big Time” exploring the first direction (including Axie recently trying the Free 2 Play path), and there are also projects like Isaac exploring the second direction. Whether the issues I mentioned can be overcome, I believe the market will tell us in due time.

I have always believed that GameFi is a significant trend because there’s no better way to open up the market and connect with everyone than through games. But right now, I’m truly at a loss. I only know that GameFi 2.0 will not be what it shouldn’t be, but I don’t know what it should be or ultimately will be…