Algorand vs. Aptos

Overview 

Algorand is a blockchain introduced in 2019 by Silvio Micali, a famous person in the cryptography space, a Turing Award-winning computer scientist, and a professor at the Massachusetts Institute of Technology (MIT).

‍
Whereas Aptos is much younger - it was introduced in 2020 by Avery Ching and Mo Shaik, two prominent META employees. They have been working on technology around Diem, a META-designed blockchain, to introduce the stablecoin Libra to the market.

‍

Consensus Mechanism

Algorand

Algorand's consensus, named Pure Proof of Stake, is based on Silvio Micali’s invention, which is VRF (Verifiable Random Function). 

‍

Advantages:

• Simplicity & Fairness.

Algorand consensus is relatively simple compared to others and highly fair. It can randomly select the next block builder and block verifiers - and nobody knows who it is until the block is propagated to the network.

• No need to have Liquid Staking.

Even if you have a low stake compared to others, you can still cheaply participate in the consensus and receive rewards from it. 

• No slashing.

There is also no slashing because VRF randomly selects 50% of the network to attest to a given block.

‍

Disadvantages:

• Bandwidth-intensive mechanism.

Algorand never fully addressed the fact that the mechanism is quite bandwidth-intensive. It requires what are called Relay Nodes, beefy nodes that move messages around for nodes that participate in consensus. Without these nodes, the throughput would be really low and could even halt.

‍

A property of Algorand's consensus is that the finality is very short, and there are no forks. When a block is proposed and attested by verifiers, it is impossible to revert.

‍

Aptos

Aptos' consensus mechanism, called AptosBFT, is based on Meta Libra’s Diem, which in turn is based on VMWare’s HotStuff consensus mechanism. It is a variation of the Proof of Stake algorithm, where you stake your tokens and participate in block production and validation for rewards. 

It is hard to find any relevant information about the details of the consensus mechanism. From what I have read in forums and chats, there is no slashing mechanism implemented yet, and the finality is subsecond.

I think it’s still quite novel and was never properly validated by other researchers; it needs time to learn its disadvantages.

‍

Staking

‍

Algorand

Algorand changed its staking rewards program once. 

In the beginning, all accounts received staking rewards. This has changed - now, only accounts participating in Governance and locking their coins receive rewards.

In the future, there will be an incentive to run a participation node and lock the tokens there, but it is not yet available. Running a participation node is cheap and easy. This would allow anyone to run consensus on their own machines with their own stake. 

There are liquid staking options, but they are not necessary for smaller players.

‍

Aptos

Aptos, on the other hand, requires a whopping 1M tokens to participate in the consensus. There are only 100 nodes participating in the consensus. Liquid staking is the only valuable option for others. As of March 2023, there is not a single liquid staking app that has amassed 1M tokens to run its own node, sadly.

‍

Performance

Algorand 

Algorand, as of today, handles an average of 11 TPS with a peak performance of 6,200 TPS, with the goal of sustaining a constant throughput of 10,000 TPS.

‍

Aptos

APTOS, as of March 22, 2032, the average rate of transactions is 8 TPS with 102 active validators, with a peak of 400 TPS. 

This is a far cry from the promised 160,000 TPS but is comparable with other blockchains. 

In the 30 days of November 2022, the peak rate of transactions in the Aptos blockchain was 2,107 TPS.

‍

Programming language

Algorand

Algorand's smart contract language is an assembler-like language called TEAL. 

It uses low and high-level opcodes to perform most operations, such as assets, branching, and asset transfers. It comes packed with many high-level opcodes like signature checks or high-precision math operations, so you don't need to develop libraries for that (like safe math, etc.).

On top of that, there are several languages that either generate TEAL or compile to it.

‍

Most notable are:

• PyTEAL, a Python library that generates TEAL. Easy to understand, it helps maintain a broader view of how a smart contract looks.

• Beaker, a framework for designing smart contracts built on top of PyTEAL. If you need an even higher level of abstraction, this should be your choice.

• Reach, a completely new language that compiles (among others, like Solidity) to TEAL, provides some solid security guarantees, but development for Algorand slowed down in 2023.

• Tealish, a language developed by one of Algorand's DEXes - Tinyman. Super simple but elegant and powerful. Worth checking out.

‍

Aptos

Aptos prepared its own language - Move:

‍

• Based on Rust and has similar syntax
• Is much simpler than Rust
• Focused on security
• Incorporates the best parts from Rust (Example: memory and execution safety)
• Has other features for a modern smart contract language (Example: protection against reentry attacks, proper interfaces for moving resources between accounts)

‍

Both languages have still many things to build. The most important are the ecosystem and the community. 

Currently, both struggle with standardizing approaches to new things. 

There are no strong forces in ecosystems similar to EVM/Solidity that, when there is a new thing, it is standardized immediately, like ERC-20 or ERC-1155. 

There is no debate within the community about how to do something in the best way. Mostly, it is centralized around respected foundations or development companies.

‍

Tokenomics

Aptos

Aptos tokenomics are straightforward and simple, with 1 billion tokens to unlock in 10 years and just two vesting types for the community, investors, and core contributors.

13% (130 million) of all tokens are unlocked from the beginning to support the community and will be distributed by the Foundation and Aptos Labs.

Then, they gradually unlock 0.3% (0.3 million) of tokens every month up until September 2032.

Starting from November 2023, investors' and core contributors' tokens begin to be unlocked, starting with 2.2% (or 22 million) in November 2023 and then 0.74% (or 7.4 million) each following month up until November 2026 (45 months).

‍

By November 2026, all 35.5% (355 million) of all investors' and core contributors' tokens will be unlocked, and 144 million of community and support tokens.

‍

This is what it looks like:

‍

Apart from that, there is also a staking reward program that mints new tokens for validators that participate in consensus. It gives a constant 7% yield and decreases yearly by 1.5% until it reaches 3.5%. With current figures, this provides around 60 million tokens a year extra.

‍

This is what it looks like in a graph:

‍

‍

Algorand

‍

Algorand is more complicated, but most of the tokens have already been unlocked. So we don't need to go into details about how. 

Algorand has 10 billion tokens that were to be unlocked over a period of 10 years, as per the graph below:

‍

‍

‍

7.5 billion have already been unlocked, including all tokens for investors. The remaining 2.5 billion are tokens in possession of the Algorand Foundation for supporting community growth.

‍

Ecosystem

‍

Algorand

‍

• Has most of the core protocols that are already built and widely adopted. Examples: AMMs like Pact, CLOB like Algodex, lending protocols like Folx Finance, Algorithmic stablecoins like xBacked, and NFT Marketplace like Rand Gallery
• Contains native stablecoins (USDC and USDT) that can easily help build liquidity in most products.
• Best practices are already established, and audit companies have strong know-how when it comes to verifying protocols. 

‍

Aptos

‍

• Young ecosystem. Started just a couple of months ago, but seeing fast growth.
• No native stablecoins (USDC and USDT)
• Some apps are already on the mainnet, but most of them are still in the testnet stage. 
• The ecosystem is growing rapidly - they already have wormhole integration to pull other tokens to the ecosystem.

‍

I think it is a matter of time when most of the missing pieces will be in place, probably even before the end of 2023.

‍

Investors 

‍

‍

‍

Summary

‍

In conclusion, both Algorand and Aptos are innovative blockchain platforms with unique features and potential. 

Algorand has a more established ecosystem, a simple and fair consensus mechanism, and a diverse range of smart contract languages.

Whereas Aptos is still in its early stages, rapidly growing, and features a secure programming language called Move. 

Investors and developers should consider each platform's strengths and weaknesses, keeping in mind that Aptos may require more time to mature. At the same time, Algorand has a more extensive ecosystem already in place. 

‍

The future development of these platforms will depend on their ability to attract talent, build strong communities, and foster collaboration.

‍

Choosing the technology that would be a perfect fit for your current and future business strategy can be quite a challenge. When you entrust product development to your Blockchain partner, rest assured they will not only choose the perfect blockchain for you but also leverage it to get maximum results and avoid costly mistakes. 

Here's our guide on choosing the right Blockchain partner.

‍