Saito (SAITO)

Data Availability Layer

Key Metrics

• Circulating Market Cap: $38M

• Fully Diluted (FDV): $73M

• Daily Tx (30d avg): ~30k

• Total Value Locked: N/A

• Supply Staked: 100%

• Staking Yield: 1%

• Inflation Rate: -2%

Please refer to the bottom of this page for important disclosure information.

estimated reading time: 30–40 minutes

A New Breed of L1s

Saito is a Layer-1 blockchain with a unique solution to Data Availability (DA) scaling. With the recent launch of “Data Availability Layers” like Celestia, a new breed of L1s are drawing more attention to modular blockchains as they further scale L2 throughput. Saito is a unique solution within the modular blockchain ecosystem, where this high throughput L1 can serve as a DA layer for any Virtual Machine while hosting a range of P2P applications from gaming to video chat.

Saito’s Solution to Data Availability Scaling

Saito uses a unique consensus mechanism that rewards all nodes in a blockchain, rather than only rewarding stakers or miners. This contrasts with L1s like Ethereum and Celestia that don’t offer incentives to non-validator nodes. Most L1s foster decentralization by simplifying node technology requirements, aiming to reduce operating costs for nodes. Alternatively, Saito removes blocksize limits and realigns incentives to enable 5-10x more data throughput than Celestia and Solana.

A Unique Solution to L2 Margin Compression

Saito can theoretically support any type of dApp, and it currently hosts several high throughput applications. However, we view Saito as a particularly attractive solution for L2s. Rollups compete by offering lower transaction costs, but this compresses fee margins over time. L2s using low cost DA layers like Celestia can reduce transaction costs and ease margin compression, but this could be temporary as more and more L2s continue to launch. In fact, new L2s like Manta are already experimenting with new revenue models like offering yield on L2 assets. We see Saito as the ideal solution to L2 margin compression. Not only does Saito offer a low cost and highly scalable DA layer, but L2s on Saito would be paid with SAITO tokens for bringing transactions into the network.

Paying dApps for Contributions

Nodes in the Saito blockchain are rewarded for their ability to drive transaction activity into the network. An example would be a popular dApp that’s driving user transactions. This dApp would earn SAITO tokens as rewards, where ~50% of Saito’s blockchain rewards are paid to nodes for propagating transactions. This incentive realignment relative to other blockchains (that only reward validators and miners) encourages more nodes and dApps to contribute to the Saito network.

Growing Interest for SAITO ahead of Ecosystem Growth Initiatives

Since SAITO began trading in 2021, investors have struggled to categorize it among monolithic blockchains like Solana and Algorand. However, with new DA layers like Celestia and EigenDA emerging, we expect greater interest in SAITO as investors better appreciate its advantages over alternative L1s. Saito is finalizing several upgrades in 2024, preparing the network for its initial wave of external development activity. We compare SAITO to a pre-launch token in a late-stage funding round, where Saito hasn’t started using token incentives to attract growth. Yet L1 valuations can expand rapidly once they start decentralizing and targeting external development.

TL;DR

• Protocol Background

  • Saito is a Layer-1 blockchain.

  • SAITO token was launched in 2021.

• Scaling Data Availability

  • Saito has a unique solution to scaling data availability.

  • Most blockchains scale data availability by sacrificing decentralization.

• The Saito Consensus

  • Saito rewards nodes for providing data availability and data storage.

  • Nodes on Saito are incentivized to provide greater transaction throughput.

• Challenging the Trilemma

  • Saito doesn’t put limitations on block sizes.

  • 51% attacks and Sybil attacks are uneconomical on Saito.

• A Solution for L2 Margin Compression

  • Rollups face margin compression as L2s proliferate and gas costs rise.

  • L2s using Saito for DA can offset margin pressure by earning SAITO rewards.

• The Saito Ecosystem

  • Several high throughput dApps are built on the Saito network.

  • Saito has a large community that actively promotes the ecosystem.

• Development Roadmap

  • Saito’s developers have refined the consensus layer over the past 2 years.

  • Decentralization and 3rd party development are major initiatives in 2024.

• Tokenomics

  • Saito is a non-inflationary protocol.

  • The foundation has a healthy treasury with an opportunity to accelerate growth.

• Valuing an Emerging L1

  • We see SAITO capturing more mindshare in 2024 as adoption accelerates.

  • We see potential for SAITO’s FDV to reach $800m in 1-2 years.

• Risks to SAITO

  • Saito competes against better capitalized L1 projects.

  • A bear case scenario could result in SAITO trading as low as $0.005.

Protocol Background

• Saito is a Layer-1 blockchain.

• SAITO token was launched in 2021.

Saito Overview

SAITO is the native token for the Saito protocol, an alternative Layer-1 blockchain. The Saito Consensus whitepaper was published by David Lancashire and Richard Parris in 2018. The project was initially inspired by the “Blocksize Wars” of 2015 when both developers lived in Beijing. Saito’s core contributors are based in the APAC region which includes 6 full-time developers.

A Mainnet Beta version of Saito launched in 2019, followed by SAITO’s token launch in 2021. The L1 currently maintains several high throughput applications, including P2P video calling, social media and gaming. These applications serve as proof of concepts demonstrating the wide range of applications that can leverage Saito’s P2P network for data availability.

Figure 1: Saito Price Chart (SAITO)

Scaling Data Availability

• Saito has a unique solution to scaling data availability.

• Most blockchains scale data availability by sacrificing decentralization.

Differentiation vs L1s

Saito differentiates from other blockchains by incentivizing Data Availability (DA). Below we highlight how protocols like Ethereum, Solana and Celestia address DA when balancing decentralization and scalability.

For readers with a strong understanding of DA, we recommend skipping to page 10.

What is Data Availability?

DA addresses how nodes in a blockchain store and share information. While some nodes in a blockchain will validate transactions via staking or mining, a decentralized network also needs non-validator nodes to store and share data unbiasedly. These non-validator nodes are known as Remote Procedure Call (RPC) nodes. To achieve a “global state”, both RPC and validator nodes need a copy of a blockchain’s distributed ledger.

To help readers better understand DA’s role within blockchain networks, we deconstruct a blockchain into 4 key layers:

• Execution Layer. Web3 users interact with dApps on a blockchain’s execution layer. dApps are programmed into the execution layer via smart contracts, where a blockchain’s set of rules are known as its Virtual Machine (VM). Each VM includes a common programming language like Solidity or Rust.

• Data Availability Layer. When a user interacts with a dApp, their wallet sends a message to a node hosting the dApp’s smart contract. After the smart contract executes the transaction, this node will share the information to other nodes in the network. Each node is constantly storing and sharing data of different transactions. Data Availability refers to the combined effort of blockchain nodes for receiving and sharing information (i.e. propagating data).

Figure 2: Nodes Share Transaction Data on the Data Availability Layer

• Consensus Layer. A blockchain’s “mempool” consists of transactions that have been shared between nodes but not yet confirmed. Some nodes in a blockchain will also confirm (i.e. validate) transactions. These “validator nodes” compete by staking or mining, where the winning validator earns the right to propose which transactions are included in the next block.

  A blockchain’s Consensus Layer determines how block proposers are selected (i.e. Proof of Stake or Proof of Work), as well other security measures that protect blockchains from attackers.

Figure 3: Validator Nodes Confirm Transactions on the Consensus Layer

• Settlement Layer. Once consensus is achieved for a given block, transactions are posted to the blockchain and become immutable. The Settlement Layer includes transactions that were confirmed by a blockchain’s P2P network. Once settlement is achieved, blockchain users can be highly confident that transactions are final and assets are safe.

• Data Storage is a component of the settlement layer. Blockchain nodes store confirmed transition data that was settled on chain. Data storage refers to the history of finalized transaction stored by nodes. While every node stores some data of a blockchain’s history, a “full node” archives the extended transaction history of the network. Alternatively, a “light node” only archives a blockchain’s recent history.

Node Operators: The Unsung Heroes of a Blockchain

Blockchains use staking and mining rewards to incentivize validator nodes to propose blocks and validate transactions. However, standard blockchains do not incentivize non-validator nodes (i.e. RPC nodes) to offer DA. That’s despite RPC nodes often representing 30–50% of the total nodes in a network.

Figure 4: Solana — Number of Nodes on Mainnet

While all nodes in a blockchain offer DA, it’s the RPC nodes that provide checks and balances keeping validator nodes honest. Yet because nodes do not earn blockchain rewards for explicitly providing DA, there’s few internal incentives to self-operate a non-validating nodes. This makes it harder to achieve network decentralization, where blockchains must attract a diverse base of node operators to sufficiently decentralize a network.

Bitcoin has managed to achieve sufficient decentralization by fostering a strong culture of users running their own nodes. While Bitcoin mining pools do create some degree of validator concentration, this is offset by a vast network of ~50,000 nodes. These independent node operators are a leading factor behind Bitcoin’s leading decentralization and security versus all other blockchains.

Figure 5: IPv4/IPv6 Bitcoin Nodes — Top 10 Countries

Bitcoin also fosters decentralization by limiting technology requirements for running a node. Bitcoin blocks can only include up to 4 MB of data, making it relatively cheap to operate a Bitcoin node. However, limiting block capacity also limits Bitcoin’s throughput. Any increases in block capacity that increase the cost of running a node typically reduce a network’s decentralization.

Ethereum blocks can include up to 12 MB of data, resulting in much higher costs to operate a node. Higher costs have deterred a diverse base of users from self-operating Ethereum nodes. In fact, ~50% of Ethereum’s nodes are operated by cloud service providers like AWS. Because of the scale achieved by these companies, it’s significantly cheaper to outsource node management than operate nodes internally.

Figure 6: Blockchain Node Requirements

Even Ethereum’s Vitalik Buterin has acknowledged censorship risks related to node centralization. Ethereum developers are proposing ultra-light nodes that can run on users’ phones; however, this is just an idea and won’t be included in any near-term Ethereum upgrades.

Figure 7: Ethereum Node Concentration

Scaling Data Availability

Meanwhile, Ethereum still faces scalability constraints. Striking a balance between node capacity and operating costs highlights the scalability trilemma. Because blockchains with greater node capacity also require higher operating costs, blockchain scalability typically sacrifices decentralization.

Figure 8: The Blockchain Trilemma

We highlight some popular solutions that Proof of Stake blockchains use to balance scalability and decentralization; however, each requires some level of compromise.

• Solana achieves scalability by leveraging highly performant nodes. A Solana node cost thousands of dollars to operate, but these more performant nodes handle 40x more data per day than Ethereum nodes. However, expensive nodes limit network decentralization. Solana developers believe node operating costs will decline over time as Moore’s Law takes hold and broader technology advancements reduce the costs of running a node on Solana.

• Rollups introduce the concept of a modular blockchain. In a monolithic blockchain like Solana, the L1 facilitates all transaction components (i.e. execution, DA, consensus and settlement). Alternatively, a modular blockchain uses independent protocols for different layers. In a rollup, the execution layer is handled by a separate L2 protocol (like Optimism), while Ethereum is used for DA, consensus and settlement.

  Rollups execute multiple transactions, and then post bundled transaction data onto Ethereum using fault proofs (optimistic) or validity proofs (zero-knowledge). Rollup users can choose to keep their assets on the rollup, or securely bridge assets onto Ethereum for settlement.

• Data Availability Committees (DAC). A true rollup uses Ethereum for DA and settlement; however, a validium uses Ethereum for settlement but not for DA. Data Availability Committees are permissioned groups of validators used by an L2 for DA. DACs significantly reduce L2 transaction costs, where this more centralized DA layer is operated by more complex nodes.

• Data Availability Sampling (DAS) is a relatively new solution to blockchain scaling that creates a more decentralized validium. Rather than use a permissioned set of nodes for DA, blockchains like Celestia use data availability sampling (DAS). This solution significantly reduces node requirements and operating costs, while also fostering high throughput and decentralization.

Figure 9: Blockchain Modularity

When an L2 posts blocks onto the DA layer, nodes using DAS only check some of the block’s data on a randomized basis. By spot checking data rather than verifying all data, nodes using DAS have significantly lower operating requirements. Meanwhile, the repeated randomness of DAS still offers a high degree of confidence (i.e. 99%) that L2 data is valid.

This allows DA layers like Celestia to support extremely high throughput applications despite using ultra-lite nodes. Additionally, DA layers only store data temporarily, typically just ~2–4 weeks. This further reduces node complexity by reducing hardware requirements for data storage.

Figure 10: Manta — L2 DA Savings after Transitioning to Celestia

The Saito Consensus

• Saito rewards nodes for providing data availability and data storage.

• Nodes on Saito are incentivized to provide greater transaction throughput.

The Saito Consensus Whitepaper

A deeper overview of the Saito Consensus can be found in the whitepaper here.

Novel Concepts for Incentive Realignment

Because nodes aren’t typically compensated for providing DA, blockchains with expensive node requirements often become more centralized. Saito takes a unique approach to scaling DA by offering unique incentive realignments. To facilitate cheap, efficient and highly secure blockchain transactions, Saito introduces four unique concepts:

1. Block space is rented rather than stored permanently

2. RPC nodes earn ~50% of blockchain rewards

3. Miners/Stakers earn the other ~50% of rewards

4. Mempools (pre-finality transactions) don’t fully reset after each new block

Concept #1: Automatic Transaction Rebroadcasting (ATR)

Similar to DA, standard blockchains don’t have internal incentives for providing data storage. A full node on Bitcoin needs ~540 GB to store the blockchain’s entire history. This compares to 1.4 TB for Ethereum and over 100 TB for Solana. Most Bitcoin and Ethereum nodes will arbitrarily prune old data to reduce storage costs. Alternatively, Solana stores its transaction history on Airwave, a separate blockchain that specializes in data storage.

Saito addresses blockchain bloat via Automatic Transaction Rebroadcasting (ATR). ATR requires users to pay fees for keeping data (i.e. assets) on chain. This is done by dropping all unspent transaction outputs (UTXOs) every epoch unless the UTXO has enough tokens to pay a rebroadcasting fee. As such, account data is pruned every epoch (~2 weeks), but users with enough SAITO in their wallet will have tokens automatically deducted to renew their data into the next epoch’s first block.

Figure 11: Automatic Transaction Rebroadcasting (ATR)

ATR creates a market mechanism for determining what data is valuable enough to remain on chain (and which data can be dropped). Data renewal fees are based on recent transaction costs (the last epoch’s average fee per byte). Because ATR fees are paid to nodes, this design creates an internal incentive for nodes to store historical blockchain data.

ATR is one of Saito’s most unique features. While full nodes on other blockchains sometimes bear the costs of storing an extended history, it’s more common for nodes to arbitrarily prune data. When full nodes do store a blockchain’s extended history, they’re often operated by centralized cloud providers that can minimize costs by operating at scale. Alternatively, Saito provides internal incentives for all nodes to store relevant data.

Concept #2: Paying Nodes for Data Availability

Most blockchains reward miners/stakers with a mix of transaction fees and token subsidies (i.e. inflation). Some blockchains also burn a share of transaction fees. However, Saito is unique, where the token supply is noninflationary and just ~50% of transaction fees are allocated to stakers/miners. The other 50% of transaction fees on Saito are allocated to nodes for providing DA.

Transaction fees paid to Saito nodes are determined by each node’s contribution to the network. Nodes that drive more transaction activity, such as by hosting a popular dDapp, are rewarded with more fees. Nodes are also rewarded for receiving and sharing transaction data with other nodes (i.e. providing DA).

Saito uses a points-based system called “routing work”, where nodes earn points for every transaction they share. When a node receives transaction data directly from an end user, the node earns the maximum number of points per transaction. As data for a single transaction is shared to other nodes, those nodes will earn points as well. However, the rewards rate is cut by half every time the same transaction is shared to another node.

Figure 12: Saito nodes earn points for sharing transactions

After several different transactions take place, every node will have accrued a number of points based on their position within various transaction paths. The sooner a node receives data for a given transaction, the more points it receives. After a node accumulates a certain number of points (i.e. routing work), it can propose a block to be added to the blockchain.

In a traditional blockchain, a miner or staker will earn the right to propose a block, either through solving a puzzle (PoW) or random selection (PoS). In the Saito Consensus, the node accumulating the most “routing work” points will earn the right to propose a block. However contrary to other blockchains, block proposers in Saito aren’t guaranteed to earn transaction rewards.

Saito transaction fees are randomly distributed to nodes based on accumulated points, where nodes with more points have higher probabilities of earning rewards. The node proposing a block will have the most points and highest probability of receiving a reward, but a block’s reward isn’t necessarily guaranteed to its block proposer.

Figure 13: Saito rewards are based on “routing work”

Concept #3: Miners & Stakers Distribute Rewards

After a Saito node proposes a block, there’s still some steps before blockchain rewards are received. The Saito Consensus also uses miners and stakers to help distribute transaction fees. In return for helping disseminate blockchain rewards, miners and stakers earn the remaining 50% share of transaction fees.

When a node has enough points to propose a block, the block’s data is added to the blockchain but rewards remain locked. Saito miners then race to unlock these rewards by completing a puzzle (similar to Proof of Work). The winning miner finds the “Golden Ticket” that unlocks the block’s transaction fees. 50% of these fees are sent to nodes for sharing transactions while the miner keeps the other 50% of transaction fees.

Figure 14: Saito Consensus – Miners Unlock Block Rewards to Distribute to Nodes

However, Saito’s miners do not always solve each block’s puzzle within the allotted time. The mining algorithm is designed for miners to compute one Golden Ticket every two blocks, on average. As miners commit more energy (i.e. hash power) to unlocking blocks, the difficulty of computing a Golden Ticket increases.

Figure 15: Saito Consensus – Miners Race to Solve a Block’s Golden Ticket

When miners are unable to solve a block’s hashing algorithm, that block’s rewards remain frozen. The next block’s difficulty is then adjusted downward. Once a miner calculates the next Golden Ticket, that miner will receive the latest block’s rewards. However, the previous block’s rewards are distributed to SAITO token holders as staking rewards. In the Saito Consensus, all SAITO token holders are automatically enrolled in staking.

With continuous difficulty adjustments, block rewards are distributed to three parties. For every block, 50% of rewards are disseminated to nodes for providing data availability. The other 50% of rewards are sent to miners that unlock block rewards; however, miners only unlock a block ~50% of the time, on average. This implies that miners earn ~25% of SAITO rewards while token holders earn ~25% via staking rewards.

In Saito’s current consensus design, these ratios are arbitrarily set and the share of rewards earned between nodes, stakers and miners can be changed. Theoretically, the rewards ratio between nodes and stakers/miners could be driven by market forces. As such, it would be possible to adjust Saito’s rewards ratio to allocate 75% of transaction fees to nodes, 12.5% to miners and 12.5% to stakers.

Concept #4: Mempools don’t reset after each block

In traditional blockchains, mempools of unconfirmed transactions shared between nodes reset with every new block. In Saito Consensus, nodes will not forget unconfirmed transaction data if those transactions were not included in the previous block. This is an important nuance that protects Saito from attacks.

For example, a Saito node that accumulates enough points (transaction data) will propose the next block. This block includes the winning node’s data but there could be transactions shared by other nodes that the block proposer never received. In the Saito Consensus, after a block is added to the blockchain, other nodes will still retain unconfirmed transaction data that wasn’t included in the winning block.

As such, points accumulated by nodes will not entirely reset with every new block. Rather, points only reset for the winning node that proposed the last block and included all their transaction data. This feature allows each node to eventually earn enough points to propose a block, even if they accumulate points more slowly than others.

This is an extremely important feature because it protects the network from a single node monopolizing block production. By allowing other nodes to retain mempools of unconfirmed transactions, each node will have an opportunity to eventually propose a block. Even if a single node is bringing a vast majority of transactions to the network (and proposing most of the blocks), it can’t censor transactions from other nodes indefinitely.

Figure 16: Saito’s Defense Against Hostile Nodes

Challenging the Trilemma

• Saito doesn’t put limitations on block sizes.

• Majoritarian (51%) attacks and Sybil attacks are uneconomical on Saito.

Proof of Propagation

The Saito Consensus uses some staking and mining incentives; however, it can’t be classified as PoW or PoS because miners/stakers aren’t proposing blocks. Rather, “Proof of Propagation” is a more accurate description, because block proposers must prove they’re adding value to the network by propagating data to other nodes (i.e. providing Data Availability).

All Saito nodes are rewarded for sharing data, but nodes can compete for more rewards by positioning themselves closer to end users. Examples of how Saito nodes can maximize rewards include:

• Operating an L2. L2s executing multiple transactions can bring meaningful activity into the Saito network. A node operating an L2 can accrue a significant amount of SAITO rewards as it drives a high volume of transactions directly from end users.

• Hosting Popular dApps. Similarly, dApps with meaningful user activity will bring transactions into the network and propagate data directly from end users.

• Efficiently Routing Transactions. A highly performant node could specialize in sharing transaction data with as many nodes as possible. Unlike a dApp or L2, a routing node might not be directly connected to end users. However, a routing node could serve as an attractive second stop for dApps and L2s to share data.

• ISP Providers. Localized ISP providers (such as one covering an apartment building) could route user transactions via a local node. In exchange, the ISP provider could offer discounted internet rates.

• Web3 Infrastructure. Saito nodes can host public facing infrastructure from other blockchains such as NFT content or data storage (like IFPS). For example, an NFT on Ethereum could store the actual image data on Saito.

Removing Blocksize Limits

Blockchains like Bitcoin, Ethereum and Celestia foster decentralization by placing limitations on how much data can be included in each block. While limiting block sizes simplifies node requirements and reduces operating cost, highly decentralized networks also lack scalability.

Alternatively, Saito doesn’t place limitations on transaction or block sizes. Saito gives all nodes an opportunity to earn fees and offset operating costs, thus incentivizing nodes to use more complex machines. This incentive realignment enables extremely high throughput on Saito, even relative to blockchains with expensive nodes like Solana and Algorand.

Saito is currently able to process up to 36 MB of data per second; although, it could theoretically handle greater throughput if Saito nodes were using even more performant hardware. This level of scalability is still significantly greater than other blockchains with implicit constraints on throughput.

Figure 17: Transaction Throughput between Networks

Saito’s ability to process large quantities of data allows the blockchain to support high throughput applications like gaming, social media and even P2P video chat. L2s using Saito for DA are even more scalable, and an L2 could theoretically subsidize user transaction costs with SAITO rewards.

Security Features

Saito’s unique incentive realignments make it prohibitively expensive to attack the network. The key attributes limiting any node (or nodes) from asserting power over the network include:

• Separating block production from PoS or PoW. Nodes on Saito compete for block production and SAITO rewards by sharing transaction data. A Sybil Attack on Saito would require matching the points accumulated by honest nodes on the network. An attacking node can only earn points by spending transaction fees, but this would require an attacker to outspend all other users on the network.

• Nodes cannot earn over 50% of fees. Not only must a Saito attacker match the network’s total spending, but they would need to match the hashing output of honest miners as well. This offers an incremental layer of security against 51% attacks compared to typical blockchains using PoW or PoS.

• Mempools don’t fully reset. In the Saito Consensus, after a block is added to the blockchain, other nodes will still retain unconfirmed transaction data that wasn’t included in the winning block. Even if some nodes are accumulating points more slowly than others, they will eventually earn enough points to propose a block. This deters an attacker from monopolizing block production and censoring transactions. As such, 51% attacks and transaction hoarding are unsustainable on Saito.

  Saito democratizes block production by giving all nodes an opportunity to propose blocks, which contrasts with many PoS and Delegated Proof of Stake (DPoS) blockcahins. Even Ethereum validators must stake 32 ETH (~$90,000) for the ability to propose blocks, while DPoS limits block production to a select minority.

A Solution for L2 Margin Compression

• Rollups face margin compression as L2s proliferate and gas costs rise.

• L2s using Saito for DA can offset margin pressure by earning SAITO rewards.

An Attractive Venue for L2s

Similar to DA layers like Celestia and EigenDA, Saito’s native blockchain does not have a virtual machine. Rather, L2s can operate their own VM independently while leveraging Saito’s decentralization and security. Given Saito’s unique rewards structure that compensates nodes, we believe it provides an ideal DA layer for L2 protocols.

L2 Margin Compression

Data Availability is traditionally an operating cost for L2 protocols, particularly rollups using Ethereum for DA. For every $1 of transaction fees incurred by users, rollups pay ~$0.70 to Ethereum for DA. This results in just ~20-30% gross margins for rollup operators like Optimism and Arbitrum.

Figure 18: L2 Operating Margins (January 2024)

Meanwhile, the competitive landscape for L2s seems to increase every day. After Coinbase launched Base in August, competing CEXs like OKX and Kraken are planning their own L2 chains. Even L1s like Celo and Polygon are transitioning into L2s. With a wide selection of software development kits (SDKs) to choose from (i.e. Optimism’s OP Stack, zkZync’s Hyperchain, Polygon’s CDK, etc.), barriers to launching a new L2 are increasingly low.

Rollups compete for users via many aspects, but offering cheap transactions is a key requirement. As more L2s launch, rollup margins have steadily compressed in their short histories. Rollups with greater scale like Arbitrum can offset margin pressure by bundling more transactions per block; however, L2 proliferation will only fragment transaction activity across different chains.

Figure 19: L2 Fragmentation and Higher Gas Fees Compress L2 Margins

Ethereum’s next upgrade in spring 2024 can ease some L2 margin pressure, but this might only provide temporary relief. “Proto-danksharding” creates a new fee market for rollups posting DA on Ethereum. This upgrade is expected to reduce L2 transaction costs by ~90-99%.

But as barriers to entry for L2s remain low and new rollups continue to proliferate, it’s not unreasonable to expect L2 margins to continue narrowing over time. L2Beat already tracks 37 active L2 protocols and 37 upcoming launches. These numbers exclude recently announced projects by OKX, A16 and Kraken which will further saturate L2 activity.

Figure 20: Major Layer-2 Software Development Kits (SDKs)

Solutions to L2 Margin Compression

L2 developers are already finding new ways to offer competitive transaction fees without sacrificing profitability.

• Alternative DA Layers. L2s using Data Availability Committees (DACs) or Data Availability Sampling (DAS) are significantly cheaper than rollups using Ethereum for DA. Most L2 cost savings will be passed to end users but we’d expect some margin benefit for the L2 as well. That said, using a non-Ethereum DA layer does reduce L2 security guarantees.

• Alternative Income Sources. More recently, new L2s like Manta and BLAST are experimenting with alternative revenue streams. Manta’s New Paradigm is a new L2 protocol that natively offers yield on user assets. Manta stakes users’ assets that are bridged to the L2, or converts bridged assets into liquid staking tokens (LSTs) like stETH or yield baring stablecoins like sDAI. Manta will pass this income to users but will likely retain some spread themselves.

Saito’s Value Prop to L2s

Saito offers L2s a low cost DA layer and an incremental source of income. Because Saito nodes are incentivized to process transactions with high throughput, the network serves as a low cost and highly scalable DA layer. Meanwhile, because L2 nodes would drive transactions into the Saito network, the L2 would be rewarded with SAITO tokens.

Similar to Celestia, Saito does not have a virtual machine (VM). This simplifies rollup integration, where L2s can implement any VM they choose (EVM, SVM, etc.). The L2 could also use validity (i.e. zero-knowledge) proofs to securely bridge assets to Ethereum for settlement.

For L2s using Saito for DA, data publishing becomes a source of revenue rather than a cost. An L2 network could theoretically offer minimal transaction fees to users, and the L2 would still earn income via SAITO rewards. An L2 using Saito could even pass some Saito rewards to honest L2 users.

Saito’s security features would also ensure that L2 transactions remain secure. In typical validiums, the DA layer is incapable of stealing user funds on the L2. However, a DA layer could theoretically “lock” user funds (i.e. ban access) which is an incremental risk relative to standard rollups using Ethereum for DA. Because DA layers like Celestia and EigenLayer use Tendermint BFT consensus, attackers can halt Data Availability if they accumulate a 33% stake weight.

Alternatively, Saito’s security mechanisms deter the DA layer from permanently censoring transactions as long as there are some honest nodes on the network. Saito also verifies that each transaction is legitimate, rather than DA layers using Data Availability Sampling.

The Saito Ecosystem

• Several high throughput dApps are built on the Saito network.

• Saito has a sizable community that actively promotes the ecosystem.

Mainnet Beta

Saito launched a permissioned version of the blockchain in 2019. Since then, a lean developer team has developed several dApps that serve as proof of concepts for Saito’s wide range of use cases. Saito’s development roadmap also includes an open/permissioned version of the blockchain launching in late 2024 or early 2025.

Not dissimilar from rollups using Saito for DA, dApps on Saito have autonomy over their programming language. Saito dApps run as ultra-light nodes directly from users’ web browsers. After running computation on users’ devices internally, Saito dApps share data across the P2P network.

Figure 21: Saito Supports dApps & L2s using Various Programming Languages

Saito’s core team has developed several dApps, including social media, gaming and even P2P video calling. The Saito community also contributes to development efforts, including launching several games. These truly peer-to-peer applications leverage Saito’s security and decentralization, while maintaining high throughput.

Red Square

Red Square is a P2P social media network comparable to X.com [Twitter] where users can share comments, pictures and videos. Red Square is handling ~1,000 user interactions per day, where ~3,000 unique addresses have interacted with the app. Despite minimal marketing efforts by Saito, this level of user activity isn’t far below leading Web3 social media platforms like friend.tech and Farcaster.

P2P Video Calling

Saito’s P2P Video calling dApp demonstrate the types of high throughput applications that can be developed on the blockchain. Saito’s P2P video chat uses public and private keys to securely connect users without censorship risks.

Figure 22: Social Networking dApps – Daily Avg Transactions (January 2024)

Saito Arcade

The Saito Arcade features over 10 different games. Users can play simple games like Chess, or computationally complex games like Mario N64 over the P2P network. After Saito’s core team developed the arcade platform, several community members have launched additional games. Saito Arcade is responsible for 90% of Saito’s ~30,000 transactions per day.

Integrated Wallet

Saito’s core team developed a Web3 wallet that’s integrated into each of these dApps. Most wallet functions are abstracted away which makes using Saito dApps more comparable to traditional Web2 applications. For new dApp users, a new wallet is created upon initially launching one of Saito’s dApps. This significantly reduces barriers for new user onboarding compared to other blockchain ecosystems. Saito’s ATR mechanism also enables disposable wallets that support short-term private browsing.

Integrated Chat

Saito has also integrated P2P messaging into its dApps. Within the web browser of each Saito app, a chat box allows users to send messages while simultaneously using other Saito applications. Users can send P2P messages to anyone in the Saito network, either via direct messages or via group messages. One group chat in particular includes the entire Saito community.

Figure 23: Cumulative Transactions on Saito – 48m to date

Future Applications

Gaming, social media and video chat are just examples of the high throughput applications that can cryptographically use Saito’s P2P network. The Saito community has also began developing several business applications like ERP, DNS registry and digital rights management solutions for transferring digital properties.

Saito Community

Saito’s core team includes six full-time developers but the team relies heavily on community engagement. Community members frequently assist in the development of Saito’s dApps and are very active in promoting the Saito ecosystem via user generated content.

Saito Community

• Registered Saitozens: 7,160 identifiers

• Twitter: 42,800 followers

• Discord: 1,663 members (permissioned)

• Telegram: 7,000+ members

• SAITO holders: over 20,000 wallet addresses

  • saitohodlers.com

Development Roadmap

• Saito’s developers have refined the consensus layer over the past 2 years.

• Decentralization and 3rd party development are major initiatives in 2024.

The Four Eras

Saito’s development team continues working towards building an open and permissionless network. Since Saito launched its Mainnet Beta in 2019, the network has operated on a permissioned basis where nodes and miners are operated by Saito’s core team. However, after completing several milestones, the team expects to begin decentralizing the network in late 2024.

Figure 24: GitHub Code Commits (January 2024)

Saito’s development roadmap consists of “Four Era’s” that officially began in November 2021. The first two Eras are nearly complete and focused on refining Saito’s consensus mechanism. Saito’s 3rd Era is expected to start in late 2024 and will focus on decentralization and supporting mass adoption.

Figure 25: Saito’s Four Eras – Development Roadmap

1st Era: Rust Consensus (Completed in 2022)

When Saito’s Mainnet Beta launched in 2019, the network’s consensus layer was running javascript. In late 2021, Saito added a node running Rust to improve network efficiency. Throughout 2022, the team refined and tested new versions of the javascript and Rust consensus software. This all took place as the Saito network continued to operate.

2nd Era: The Dawn of Persistence (Expected Completion in Late 2024)

The 2nd Era further refines Saito’s consensus layer. The Dawn of Persistence brings automated staking rewards to offset Automatic Transaction Rebroadcasting (ATR). Saito users incur ATR fees every ~2 weeks to keep assets/data on chain; however, this can be offset by staking rewards. With the Dawn of Persistence, SAITO tokens are automatically staked and receive staking rewards in excess of ATR fees.

Saito staking is currently live but requires further testing over the coming months. This will occur alongside further refinements to the consensus layer such as the addition of NFTs. Developers will also continue to focus on improving the user experience for Saito’s dApps, which includes optimizations for mobile device usage.

3rd Era: The Real Economy

Saito is also working on launching an SDK to further promote third-party development. This 3rd Era is targeted to begin in 2H24 and will focus on attracting 3rd party applications and infrastructure. This phase will decentralize the Saito network as 3rd party developers begin operating Saito nodes for their own projects.

Because Saito nodes earn SAITO tokens for contributing to the network, this incentive alignment helps attract 3rd party development. Additionally, because Saito is VM agnostic, where developers can create dApps using any programming language, this makes it easier to onboard a wider audience of dApp developers. With 3.8bn SAITO tokens in the foundation’s treasury, Saito has meaningful flexibility to offer incentives for accelerating adoption.

4th Era: The Open Network

The 4th and final Era will mark Saito’s maturation into an open and permissionless network. In typical PoS blockchains, early investors and developers can influence governance alongside large token allocations. However, Saito is not a PoS blockchain, and will instead decentralize as a diverse base of independent node operators join the network. As such, Saito’s early contributors will have limited influence over the protocol as new dApp developers and infrastructure providers make contributions to the network.

Tokenomics

• Saito is a non-inflationary protocol.

• The foundation has a healthy treasury with an opportunity to accelerate growth.

Token Supply

SAITO launched with a total token supply of 8 billion, where 2.7 billion (33%) are publicly in circulation. This includes 450m tokens (5.6% of total supply) allocated to early investors. Saito’s early investments include a seed round in 2018, and a 2021 investment from Polkadot’s accelerator program. 1.6bn SAITO tokens were also reserved for core team members and developers.

Figure 26: Saito Token Distribution

This leaves 3.8bn SAITO tokens in Saito’s treasury, as well as enough USDC/BTC/ETH to provide core contributors with enough runway for 2+ years. All lockups for early investors expired and have been open to public circulation. Meanwhile, the foundation has not released any new SAITO tokens since 2021 and was careful to prudently manage spending during the 2022-23 bear market.

With 3.8bn SAITO tokens in the treasury, Saito remains well capitalized heading into the next bull market. We see an opportunity to deploy some of these tokens to accelerate growth such as by expanding the core developer team. Additionally, as Saito targets external development during its 3rd Era, we see a meaningful opportunity to use SAITO tokens as incentives to onboard more contributors to the network.

Token Inflation

Saito is a non-inflationary protocol, where blockchain rewards are earned via network fees rather than token subsidies. SAITO tokens are automatically staked to earn ~25% of the network’s transaction fees. These staking rewards more than offset any ATR fees paid by token holders to keep data/assets on chain. Additionally, Saito becomes slightly deflationary as tokens are burned when wallets don’t hold enough tokens to pay ATR fees.

Saito’s token supply is currently declining at a ~2% rate while its staking rewards are offsetting ATR fees by a 1ppt rate. This implies a ~3% real yield for holding SAITO tokens. SAITO’s positive real yield and deflationary structure contrasts with other blockchains that typically use token inflation to pay staking rewards.

Figure 27: Yield Structures of Layer-1 Blockchains

Valuing an Emerging L1

• We see SAITO capturing more mindshare in 2024 as adoption accelerates.

• We see potential for SAITO’s FDV to reach $800m in 1-2 years.

Path to SAITO reaching $800m FDV

We see a path to SAITO reaching a $800m FDV within 1-2yrs, which would imply a $0.10 token price (vs $0.009 today). Saito has achieved meaningful scale despite operating in Mainnet Beta without any token incentives. Our $800m FDV forecast is based on recent private valuations for L1 protocols that took place 6-12 months before token launch.

SAITO’s unique traits leave it well positioned into the next bull cycle, particularly as investors focus on blockchain modularity and DA layers. Accelerating adoption in 2024-25 will be a key driver to SAITO’s valuation as the project focuses on external development and transitions to a permissionless network. Not only should this accelerate user/transaction activity, but greater decentralization typically boosts L1 valuations as well.

Valuing L1s

There’s been extensive debate in the Web3 community over the best valuation frameworks for L1s. However, it’s increasingly evident that the value ascribed to L1s goes beyond the fees accruing to token holders. Even L1s with limited fee activity like Algorand (ALGO) and Celestia (TIA) trade at multi-billion dollar valuations.

Unlike equities or decentralized apps, L1 tokens can derive value as several types of assets:

• Capital Assets: L1 tokens like ETH and SAITO earn staking rewards to generate yields. This drives demand from investors similar to capital assets.

• Consumable Assets: Users must spend L1 tokens to transact on L1 networks. As network activity scales, this drives demand for the L1 token.

• Monetary Assets: L1 tokens like BTC and ETH have earned monetary premiums. L1 tokens with strong ecosystems are used as base currencies, such as within NFT marketplaces. Many investors also own low inflation L1 tokens as hedges against fiat currency debasement.

Meanwhile, competition among L1s is intense and it’s difficult to overcome incumbent’s network effects. Yet token valuations can expand rapidly once L1s do gain adoption and demonstrate staying power. Of the top 20 digital assets by fully diluted market cap, L1s represent over 50%.

Figure 28: Top 20 Protocols by FDV (January 2024)

New Breed of L1s

When SAITO launched in 2021, investors had a hard time categorizing it versus L1 peers. While Saito does support high throughput applications like Solana and Algorand, Saito does not have a native VM to support smart contracts. As such, it was difficult for Saito to draw investor and developer attention as it competed against a sea of monolithic blockchains supporting DeFi and NFT activity.

Yet we view the emergence of DA layers like Celestia and EigenDA as a catalyst for SAITO’s valuation rerating. Similar to Saito, these new Alt-L1s support high throughput transactions but do not natively support smart contracts. Since launching in October 2023, Celestia is already the 11th most valuable protocol by FDV. Meanwhile, several monolithic L1s are now pivoting into L2s (i.e. Polygon; Celo) or into DA layers (i.e. Near).

Figure 29: Saito has more similarities with DA Layers than Monolithic Chains

DA layers also face lower barriers to entry than monolithic chains. Saito can overcome many L1 network effects by supporting any VM. While L1s typically target certain developer communities that build with specific programming languages, Saito can target a much deeper talent pool within the Web3 community and beyond.

Meanwhile, there’s less competition between DA layers than monolithic chains. Last cycle saw the emergence of countless monolithic chains seeking to compete with Ethereum. As VC funding over indexed to L1s, competition has only increased with recent launches of leading edge L1s like Aptos and Sui.

Alternatively, there’s just a handful of DA layers competing to provide DA to a growing ecosystem of L2s. With limited direct competition, we believe SAITO can more easily differentiate itself to both investors and developers this cycle. Additionally, we believe the Web3 community will more easily understand Saito now that DA layers are an emerging and widely discussed theme.

Figure 30: Saito faces less competition than Monolithic L1s

SAITO can reach $0.10

We see potential for SAITO to trade as high as $0.10 over the next 1-2 years which is multiples above its $0.009 price today. Our $0.10 forecast implies a fully diluted market cap (FDV) of $800m vs its current $73m. This 1-2yr forecast would still imply a discount relative to many L1s with limited user activity.

With ~30k daily transactions, 7,000+ registered “Saitozens” and several functioning dApps, Saito has decent activity metrics for an L1 in Mainnet Beta. Yet it’s difficult to compare SAITO to other L1 tokens, particularly since Saito hasn’t started utilizing its treasury to incentive growth. Rather, we view SAITO more comparably to a pre-launch token of an upcoming L1 protocol.

We look at recent investment rounds for L1 ventures that took place 6-12 months before token launch. Similar to Saito, it was too early for these pre-launch protocols to begin using tokens to incentivize growth. On average, recent private L1 ventures were valued at a $800m FDV (pre-money).

Figure 31: Pre-launch L1s – Fully Diluted Valuations ($m)

Saito’s treasury still has 3.8bn tokens, implying ~50% of SAITO’s supply remains untapped. As Saito begins decentralizing the network over the next 1-2 years, the project will be in a better position to start using tokens for incentivizing adoption. Digital assets typically begin trading at prices above pre-launch investment rounds, and we view SAITO as a similar opportunity.

Figure 32: L1 Token Launch Price vs Pre-launch Investment Round

Meanwhile, we don’t believe any of these recent L1 protocols address the scalability trilemma better than Saito. Even when comparing Saito to leading edge DA layers like Celestia, Saito can offer greater throughput alongside higher security guarantees and censorship resistance.

Figure 33: Data Throughput for L1 Blockchains (MB/s)

Risks to SAITO

• Saito competes against better capitalized L1 projects.

• A bear case scenario could result in SAITO trading as low as $0.005.

We highlight key risks that may limit Saito’s broader adoption, as well as risks related to network security. In a bear case scenario, SAITO could retest all-time lows of $0.005.

• L1 Competition: Competition among L1 protocols is intense, while Saito competes against better capitalized projects with significantly larger treasuries. As such, it may be difficult for SAITO to compete for user and developer activity if it’s unable to offer incentives comparable to competing L1s.

• Development: Saito remains in the early stages of its development cycle, where the project remains in Mainnet Beta with a permissioned validator set. As such, there may be unforeseen software bugs as the blockchain begins operating on a fully open and permissionless basis.

• Liquidity. SAITO tokens maintain lower trading liquidity than many digital assets. Lower liquidity could offer higher trading volatility and limit investors’ ability to unwind positions.

• Decentralization: Saito’s unique incentive alignments foster both high throughput and resilient security features. However, it remains to be seen how these incentive alignments impact decentralization. On one hand, an ecosystem with a diverse base of dApps and infrastructure would foster decentralization, given network participants operate their own nodes.

  If one dApp or infrastructure provider facilitates enough transactions, it could theoretically concentrate routing work between a limited group of nodes. This might impact network redundancy. However, given Saito’s incentive alignments, even a node with 90%+ of routing work would be unable to sustainably attack the network.

• Mining Restrictions: Saito’s consensus mechanism uses mining rewards which could expose the protocol to public and government scrutiny. This includes mining restrictions in certain jurisdictions, or taxes targeting mining operations.

• Core Team: Saito’s core developer team includes just 6 full time developers and the project is highly dependent on its co-founders. Any events that lead to team members leaving the project, particularly Saito’s co-founders, could impair the future development of Saito.

  Additionally, Saito’s lean developer team and lower capitalization makes it difficult to accelerate project development. While Saito’s large treasury offers an opportunity to raise capital and expand its developer team, it may take longer than expected to complete Saito’s development roadmap.

• Newer Blockchains: The Web3 ecosystem is constantly evolving and developers are consistently developing newer and more complex blockchains. As blockchain technology continues to advance, Saito’s unique advantages could decline if newer blockchains offer greater security, throughput and decentralization.

• Irrational Attacker: While it’s theoretically unprofitable to sustainably attack the Saito network, an irrational and well capitalized attacker could try to Sybil attack or 51% attack the network for an extended period of time.

Core Contributors

Saito’s core team consists of 9 full-time contributors, including 6 developers. The team is led by David Lancashire and Richard Parris who were the co-authors of The Saito Consensus whitepaper. Both authors’ involvement in blockchain began while living in Beijing in 2012-13.

• David Lancashire: David is the lead developer behind Saito. Prior to founding Saito, David founded a Chinese online education platform in Beijing. He also held Senior Programmer and CTO roles at other Chinese based tech companies. David has degrees from University of Toronto – University of Trinity College and a Masters in Economics at UC Berkeley. His involvement in crypto started in 2012.

• Richard Parris: Richard supports Saito’s software development while also leading business/community development initiatives. Prior to co-founding Saito, Richard was CTO of Edanz, a Japanese technology company that was acquired by M3 in 2019. Prior to Edanz, he had over 15 years of software development experience in multiple sectors across the APAC region. Richard holds degrees in Mathematics and Philosophy from the University of Melbourne. His involvement in crypto started in 2013.

Important Disclosures

The following disclosures relate to relationships between Altcoin Research Limited “Alts RSCH” and the digital assets referred to in this research report.

This report has been prepared by Altcoin Research Ltd which is a consultancy business registered in Hong Kong. All content was produced independently by the author(s) and does not necessarily reflect the opinions of the developers whose protocol is covered in this report. Altcoin Research receives referral fees from third-party Virtual Asset Service Providers (VASPs) that do business with the foundations and core contributors of the digital assets referred to in this report. Referral fees paid to Altcoin Research are variable and based on activities between VASPs and respective token projects. Altcoin Research also received a deposit from the respective token foundation for consultancy services. As a result, investors should be aware that Altcoin Research may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making an investment decision.

General Disclosures

This research report is for informational purposes only.  This research is based on current public information that we consider reliable, but we do not represent it is accurate or complete, and it should not be relied on as such.  Any opinion expressed in this report is subject to change without notice and may differ or be contrary to opinions expressed by other professionals or business areas of Altcoin Research.  We are under no responsibility to update our research.

The views expressed in this research report accurately reflect the publishing analyst’s personal views about the subject digital assets.  No part of the analyst’s compensation was, is, or will be, directly or indirectly, related to the specific recommendations or views expressed by that analyst in the research report. Altcoin Research does not take long or short positions in, and buy or sell, the digital assets or derivatives thereof of tokens referred to in our research reports.

This research is not an offer to sell or the solicitation of an offer to buy any security.  It does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual clients.  Investors should consider whether any advice or recommendation in this research is suitable for their particular circumstances and, if appropriate, seek professional advice, including tax advice.  The price and value of the investments referred to in this research and the income from them may be highly volatile.  Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur.  Certain transactions, including those involving futures, options, and other derivatives, give rise to substantial risk and are not suitable for all investors.

No part of this material may be copied, photocopied, or duplicated in any form by any means or redistributed without the prior written consent of Altcoin Research Limited.