In- depth interpretation of Layer 2 scalability: How does a project choose between ZK-Rollups and subnets?


Recently, popular one-tier networks like Solana and BNB Chain have experienced network downtime due to concurrently running DeFi protocols, NFT marketplaces, and Web3 games. Likewise, Ethereum has been criticized for network congestion and expensive gas fees that plagued its network. As a result, the industry is beginning to realize the importance of Layer 2 (L2) scalability solutions.

According to the Ethereum Foundation, rollups are currently the go-to L2 solution to the scalability problem. Besides rollups, other L2 solutions (such as sidechains and subnets) exist to solve the scalability trilemma.

In particular, Subnets, Avalanche’s L2 scalability solution, has recently gained considerable popularity following the deployment of DeFi Kingdoms and Crabada subnets earlier this year. In some cases, subnets are essentially sidechains, except that subnets can choose to use validators from the mainnet. Currently, subnets and sidechains seem to be the preferred L2 scaling solution for web3 games.

With these different L2 solutions, how do they actually stack up against each other? In this article, we’ll take a deep dive into the rollups space to discuss recent developments and pain points within it. We will also analyze rollups related to custom chains (subnets or sidechains) and examine how these different L2 solutions actually stack up against each other, especially in the GameFi space.

ZK-Rollups Prospects, Recent Developments and Pain Points

By packing compressed off-chain transactions into a single order on the L1 network, Rollups achieves lower gas fees and expanded transaction capacity for any given L1. This way, L2 can perform calculations, while L1 only needs to verify them. From this, the gas fee will be split among a bunch of transactions, and thousands of transactions can be done collectively.

Rollups come in two different schemes: Optimistic Rollups and zero-knowledge rollups (“ZK-Rollups”). The two differ in their verification methods.

Optimistic Rollups assumes that all transactions are valid and publishes them to the base layer without verification. Validators then modify the disputed transactions within a 7-day challenge period.

ZK-Rollups, on the other hand, create a cryptographic proof and publish it to the mainnet as a proof of validity. Currently, transactions using ZK-Rollups can be completed in about 10 minutes.

In-depth interpretation<span class=


As can be seen from the graph above, it seems that ZK-Rollups will enable a faster and more scalable Ethereum ecosystem compared to Optimistic Rollups.

But it is worth noting that ZK-Rollups is still an emerging technology and lacks EVM compatibility. While zkSync did launch an EMV-compatible zkEVM on its public testnet earlier this year, there is still a long way to go in terms of EVM compatibility for ZK-Rollups. Otherwise, ZK-Rollups will be limited to token transfers and exchanges.

In-depth interpretation of Layer 2 scalability: How does a project choose between ZK-Rollups and subnets?

Source: (data as of June 7, 2022)

As can be seen from the graph above, Optimistic Rollups currently holds 72% of the overall market share in terms of TVL. The aforementioned lack of EVM compatibility seems to have put ZK-Rollups on the back burner, with large DeFi players not opting for ZK-Rollups when scaling to rollups.

As Optimistic Rollups have matured, most large DeFi projects have adopted the scheme on an as-needed basis. However, the one-week waiting period often overwhelms new users in the field. As an alternative, when considering GameFi scalability, NFT marketplaces and web3 applications, ZK-Rollups may be a possible solution for networks that require higher throughput and cheaper transaction fees.

In-depth interpretation<span class=

Source: and the protocol’s website (data as of June 7, 2022)

As of this writing, there are more than 12 ZK-Rollup solutions in the field, with the underlying technology developed by StarkWare, Loopring, Matter Labs, and Polygon Hermez.

In-depth interpretation<span class=

There are two main cryptographic proof methods in ZK-Rollups. The technology employed by Loopring and zkSync creates a zero-knowledge, succinct, non-interactive zero-knowledge proof (ZK-SNARK) to verify transaction packets in the mainnet. ZK-SNARKs came into focus in 2012 and were widely adopted by ZK-Rollups pioneers.

On the other hand, Zero-Knowledge, Scalable, Transparent Zero-Knowledge Proof (ZK-STARK) technology is relatively new, released by StarkWare in 2018. ZK-STARKs are considered more transparent and extensible than ZK-SNARKs. The promising technology recently pushed StarkWare’s valuation to $8 billion in a round led by Tiger Global.


StarkWare offers two main ZK-STARK solutions: StarkEx and StarkNet.

StarkEx is a Validium solution designed for application-specific ZK-Rollups, adopted by derivative trading platform dYdX, web3 games and NFT Marketplace Immutable X.

Validium is a scaling solution that employs the same proofs of validity as general-purpose ZK-Rollups, but does not store data on the mainnet. However, the risk of adopting Validium is that the data availability manager may take control of the user’s assets, causing security issues. However, off-chain data storage bypasses transaction processing on the mainnet, which in turn significantly enhances scalability compared to general-purpose ZK-Rollups.

In-depth interpretation of Layer 2 scalability: How does a project choose between ZK-Rollups and subnets?

Source: (as of June 8, 2022)

StarkNet is a general rollup launched by StarkWare, which was launched in February this year. It is worth mentioning that Immutable X includes a Volition model that allows users to choose between a Validium model or a generic ZK-Rollup. As TVL gradually picks up, StarkWare focuses on accelerating StarkNet’s TPS to an optimized level.

Cairo is the programming language of StarkNet. StarkNet itself does not support EVM, and at the time of writing, the Warp team at Nethermind is developing a transpiler from Ethereum Solidity to Cairo.

StarkEx’s Web3 Gaming Footprint

With ZK-Rollups, web3 games can significantly reduce transaction costs for users, avoid network congestion, and benefit from Ethereum’s measured security. As a result, ecosystems such as Sorare, Immutation X have adopted StarkWare’s scalability solutions in an effort to support web3 games on their platforms.

Sorare is a famous football card game that became popular on the Ethereum mainnet and then moved to L2 by adopting StarkEx. Sorare chose ZK-Rollups to scale without sacrificing security and decentralization.

Immutable X is an L2 platform for Ethereum-based NFT marketplaces. It offers zero gas fees and instant transactions by moving computation off-chain. According to DappRadar data, there are currently four active web3 games performing NFT transactions with the help of Immutable X. Gods Unchaine is the most prominent one, with around 5,000 users and a daily trading volume of $1.5 million at its peak.

Illuvium is the latest web3 game to join the Immutable X platform, and ahead of its official game release, Illuvium made headlines with the sale of $72 million of virtual land.

In conclusion, building a game in ZK-Rollups can be done in two ways. The first is through solutions like Immutable X, and the second is through building a separate ZK-Rollup, similar to what Sorare chose. Comparing the two approaches, Immutable X’s solution has proven to be more cost-effective. However, it’s important to note that Immutable X still lacks adoption, and ZK-Rollups have so far failed to gain traction in the web3 gaming space.


Loopring is an in-house developed ZK-Rollup solution. As of this writing, the protocol is TVL’s largest general-purpose ZK-Rollup chain. While Loopring is strategically focused on transactions and payments, its efforts to support web3 gaming are only reflected in its partnership with GameStop to build the Gamestop NFT marketplace. There are currently no web3 games running on the platform.

Matter Labs (the team behind zkSync)

Matter Labs is the team behind the zkSync scaling solution. It uses ZK-SNARK and provides off-chain Validium solution zkPorter and on-chain universal rollup zkSync.

ZKSpace is forked from zkSync and uses the same ZK-SNARK technology to solve the scalability problem. zkSync pointed out that ZKSpace has been closed since the fork, with the potential for security risks. As of this writing, ZKSpace (formerly known as ZKSwap) owns the protocols ZKSpace, ZKSwap, and ZKSea (not yet live).

Matter Labs launched zkEVM, the first EVM-compatible ZK-Rollup, on the public testnet in February 2022. zkEVM is still in development and is designed to run 99% of Solidity contracts. This marks an important milestone for ZK-Rollups to run in parallel with the Ethereum mainnet and compete with Optimistic Rollups.

Matter Labs is also currently designing a Validium solution, zkPorter. If Matter Labs is going to attract web3 games to its platform, zkPorter may be just what it needs. However, Matter Labs still seems to be focused on building general-purpose ZK-Rollups.


Polygon is very active in forming its own L2 scalability solution. Its ZK-Rollups products include Polygon Hermez (ZK-SNARK), Polygon Zero (ZK-SNARK) and Polygon Miden (ZK-STARK). Polygon acquired ZK-Rollups expansion project Hermez and startup Mir in 2021 and gained expertise to build its ZK-Rollups.

As of this writing, only Polygon Hermez is active. It can be used as a general ZK-Rollup for DeFi projects with a TVL of only $366,000. Polygon Hermez has yet to attract many projects, possibly due to the lack of EVM compatibility. However, Polygon may have strategically targeted web3 game developers through its PoS sidechains or supernets, and has yet to shift its focus to ZK-Rollups.

Pain points and summary

A recurring problem with generic ZK-Rollups is their EVM compatibility. We noticed that most DeFi players in Ethereum choose Optimistic Rollups or Polygon PoS as their L2 scaling solution. For example, Curve has chosen Optimism; Aave V3 is used for Polygon, Optimism and Arbitrum; Uniswap has been extended to Optimism and Polygon. zkEVM is being tested by Matter Labs, while Nethermind is developing a transpiler for StarkNet. It is possible that any breakthrough in EVM compatibility will ultimately help attract DeFi projects to ZK-Rollups in the future.

Another problem is the lack of communication in ZK-Rollups. This can lead to liquidity disruptions and the centralized sorting process can raise security concerns. However, Immutable X’s new cross-rollup liquidity solution could herald a more liquid rollup ecosystem.

ZK-Rollups have yet to gain traction in the NFT and web3 gaming space, and for good reason. Gods Unchained keeps losing users and has seen a sharp drop in transaction volume since adopting Immutable X’s solution. Likewise, Sorare was unable to repeat its success in the Ethereum mainnet after entering the rollup. Despite the lack of success of ZK-Rollups in the above examples, Avalanche’s subnets have found a more solid foundation in both NFT and web3 games by comparison.

Avalanche subnets and Polygon

Introduction to Avalanche Subnets

In-depth interpretation of Layer2 scalability: How does a project choose between ZK-Rollups and subnets?


Avalanche’s main network consists of three blockchains – the X, P, and C chains. Each chain serves a different purpose in maintaining the integrity of Avalanche and in creating new assets or blockchains.

  • The exchange chain (X-chain) is the default asset blockchain on Avalanche, supporting the creation of new assets, transactions, and transfers across subnets.
  • The platform chain (P-chain) is the metadata blockchain that coordinates validators, tracks active subnets, and enables new subnet creation.
  • The contract chain (C-chain) is a smart contract blockchain. It supports the creation of EVM-compatible smart contracts.

Avalanche uses the scaling method to create subnets for horizontal scaling. Subnets are different from a single blockchain, and the relationship between them is as follows:

  • A subnet is a set of validators on the blockchain.
  • Each blockchain has only one subnet (a set of validators), and each subnet can maintain multiple blockchains simultaneously.

You can start with five validators per subnet, but 10 is recommended for network health and security. Only one AVAX is required to create a subnet on Avalanche. However, each validator of the subnet is also required to validate the main network — 2,000 AVAX tokens, worth around $50,000 as of June 9, 2022.

Subnet creation is designed to be permissionless and flexible in requirements to allow for different use cases. For example, a private blockchain might require transactions to be viewable only by subnet validators; or a validator must undergo KYC/AML verification before entering a subnet.

Web3 Games on Subnets

Subnets have always been the holy grail of L2 scalability, especially for web3 games. Avalanche has been supporting and growing its GameFi ecosystem and finding a home for them in the subnet. Crabada and DeFi Kingdoms are two of their pioneers, and Castle Crush is currently developing its testnet.

In-depth interpretation of Layer2 scalability: How does a project choose between ZK-Rollups and subnets?


As can be seen from the table above, Crabada’s share of gas fees on Avalanche’s C chain was the highest at 38.93%. However, similar to many web3 games, Crabada faced network congestion on the mainnet, prompting them to move to their own chains, such as subnets. Recently, Crabada moved from the Avalanche C-chain to their own subnet, Swimmer Network. With the move to the Swimmer Network, Crabada players’ gas fees have been reduced by 85% and transaction times have been greatly reduced.

In addition to the obvious benefits of lower gas fees and higher TPS, there are other benefits that can attract web3 games to adopt subnets. By having its own subnet, web3 games inherently have better control over the predetermined parameters of the blockchain. Their gas fee can be paid in a token of their choice. Crabada now uses $TUS as a fuel token instead of $AVAX, burning up to 80% of its fuel for network scaling. Token burning dampens token inflation and increases token utility.

Notably, game studios have been building one subnet for the entire gaming ecosystem, rather than the typical one-game-per-subnet model like Crabada or DeFi Kingdoms employ.

Avalanche also demonstrated its support for web3 games built on its subnets. After its initial planned $190 million Avalanche Rush to advance the DeFi ecosystem, there is the launch of the $290 million Multiverse program aimed at accelerating the development of the ecosystem.

Polygon Supernet and Sidechains

In-depth interpretation of Layer2 scalability: How does a project choose between ZK-Rollups and subnets?

Source: Polygon Blog/Supernet

A competitor to the Avalanche subnet is the Polygon Supernet, a scaling solution after two well-known sidechain solutions, Polygon Edge and Polygon PoS. As mentioned, Polygon’s scaling solution is quite radical, with ZK-Rollups coexisting with supernets and sidechains.

Polygon supernets are still new in web3 gaming and haven’t gone through many practical applications. However, the supernet solution is positioned as an alternative to Avalanche subnets with more functionality. It is worth highlighting that the Polygon Supernet offers the adoption of a Proof of Authority (PoA) or Proof of Stake (PoS) chain; selection from a group of professional validators to enhance decentralization; and the possibility to share security with the Polygon mainnet .

Polygon PoS has always been at the center of web3 gaming, with successful titles like Aavegotchi and Crypto Raiders seeing huge success in 2021. Popular games come with high volumes and transaction issues. Similar to Crabada on Avalanche, Sunflower Land caused severe network congestion issues, consuming up to 42% of the gas on Polygon during peak periods.

In short, Polygon’s various scaling solutions (rollups and supernets) are premature. The sidechain Polygon PoS that hosts countless web3 games is the most powerful option. However, Polygon PoS has more similarities to layer 1 general purpose sidechains in that it has its own set of validators and still suffers from network congestion. We consider its supernet to be more similar to Avalanche’s subnet, which we compare with ZK-Rollups below.

Comparing Custom Chains to ZK-Rollups

Essentially, both L2 ZK-Rollups and custom chains (Avalanche Subnet and Polygon Supernet) are solutions to the scalability issues their respective L1 blockchains suffer. Both have vastly different technical specs and the benefits of using them, and we’ll try to break them down to better understand the differences.

In-depth interpretation of Layer 2 scalability: How does a project choose between ZK-Rollups and subnets?

If we only focus on the technical aspects of web3 games to decide whether to build on the L2 or Avalanche subnet, one might argue that L2 offers a better solution. However, the L2 solution is still in development and has not been battle tested. Subnets seem to be gaining popularity in the web3 game due to the need for their own chain.

Generally speaking, web3 games do not require much interaction with the larger ecosystem compared to DeFi projects. They often have closed-loop ecosystems where their tokens and NFTs support each other. Outside of their ecosystem, the only interaction is through their network tokens every now and then, like AVAX or MATIC.

In-depth interpretation of Layer2 scalability: How does a project choose between ZK-Rollups and subnets?


On the other hand, top web3 game studio Gala Games or the soon-to-be-released Myria prefer to build on their own blockchain. The two studios have multiple games that interact to some extent; therefore, it makes more sense for them to build games that live together on the same blockchain.

Still, given that most games are currently made by a single project team, not part of a larger collective, there is very little collaboration between games. Therefore, it may be more reasonable for smaller game studios to adopt subnets rather than building their own chains.

Aside from technical specifications, the core difference between Avalanche Subnets and ZK-Rollups is the isolation of Web3 games on individual subnets or sidechains. At this stage, Immutable X is still an ecosystem where web3 games and NFTs can interact and build upon each other, while Avalanche subnets are individually segregated into modular blockchains.

How does a project choose between ZK-Rollups and subnets?

Subnets and supernets are not designed for DeFi projects due to the lack of composability with the mainnet. EVM-compatible ZK-Rollups are suitable for projects deploying smart contracts and requiring a high level of security. However, as EVM compatibility for ZK-Rollups is still in its infancy, Optimistic Rollups appear to be the only option available to DeFi projects.

Clearly, for web3 games, subnets have shown to be a better choice than ZK-Rollups. Separate custom chains for single-game/single-chain models have gained traction, as evidenced by the popularity of Avalanche subnets.

ZK-Rollups and Validium may have better technical specifications than custom chains, but they are still in development. The ZK-Rollups scalability solution does not provide a separate custom chain, it can be “crushed” due to sudden hype of a single project. Also, creating a separate rollup like Sorare comes with high costs and transition pains.

In-depth interpretation of Layer2 scalability: How does a project choose between ZK-Rollups and subnets?

Source: ImmutableX, StarkNet, StarkEx

On the other hand, Immutable X recently released an innovative Layer 3 solution for web3 games. With this approach, game studios will be able to enjoy hyperscale solutions in Tier 3 without sacrificing security and composability. Layer 2 and Layer 3 integration integrates the liquidity of the rollup solution and strengthens the interaction between DeFi and web3 gaming. While it still appears to be in early development, we’re excited to see what else Immutable X has to offer.

All in all, the skyrocketing valuations StarkWare and Immutable X have received from venture capital have led institutional investors to pour large sums of money into them. This in turn increases confidence in the ability of ZK-Rollups to scale scalability within the Ethereum ecosystem. Additionally, Polygon is still taking an active approach to leveraging ZK-Rollups. So it’s clear that ZK-Rollups’ potential has yet to be fully unlocked – we’ll be keeping a close eye on how ZK-Rollups will flourish in the future.

Posted by:CoinYuppie,Reprinted with attribution to:
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Like (0)
Donate Buy me a coffee Buy me a coffee
Previous 2022-07-04 11:37
Next 2022-07-04 11:38

Related articles