DeFi is not the only driver of Solana adoption. Digital artists have also embraced the platform’s low fees and created some smash hits in the process.

The two top NFT projects on Solana are “Degenerate Ape Academy,” which hit the ball out of the park in August 2021 and sold out in minutes, and rival “Solana Monkey Business,” whose secondary sales have continued growing in the last months.

Total NFT sales reached $247 million in September 2021 and $246 million in October. At the same time, Ethereum NFT sales amounted to $2.2 billion and $1.7 billion, respectively.

Most of that volume comes from big-ticket sales such as “CryptoPunks” or “Bored Ape Yacht Club,” where a single deal can be worth millions of dollars.

Secondary sales for Solana’s top NFT projects from August to October 2021

Solana NFT Sales

Looking at just the dollar-denominated volume doesn’t paint a complete picture, however. Ethereum saw 132,879 and Solana 68,235 unique buyers in October, according to Messari. The average amount paid for an NFT on Solana was $1,097 in October 2021, signifying 224,000 purchases. 

From a liquidity standpoint, selling NFTs on Solana might be even more attractive to artists than Ethereum because competition is less intense, and the number of transactions is still half.

Solana is backed by the high and mighty in crypto, and the project showcases promising technological innovation. Centralization risk is a caveat, but large token holders are unlikely to dump significant amounts of SOL any time soon — not while there is money to be made in DeFi, futures and NFTs.

But Solana is not the only crypto project competing with Buterin’s “world computer” which itself is a force of nature and still the main driver of innovation and value in crypto. We will therefore also take a look at projects like Polkadot and Algorand.

This article is an extract from the 80+ page Scaling Report: Does the Future of Decentralized Finance Still Belong to Ethereum? co-published by the Crypto Research Report and Cointelegraph Consulting, written by ten authors and supported by Arcana, Brave, ANote Music, Radix, Fuse, Cryptix, Casper Labs, Coinfinity, Ambire, BitPanda and CakeDEFI.

Transactions per second are not the best metric to gauge users’ felt experience. Crypto influencer Packy McCormick stated that using Solana felt like “using the internet” in his analysis for Not Boring. But what does that mean?

The first slide in Solana’s seed round deck reads: “Solana is blockchain at NASDAQ speed.” 

Solana block times are measured below 400 milliseconds, which demands considerable network speed and node processing capacity. As a result, Solana node requirements are steep. A prospective node needs a 12 core CPU, 128GB RAM (256GB for an RPC node) and a blazingly fast 500GB SSD.

Theoretical TPS

As a proof of concept, Solana’s testnet has demonstrated 400,000 TPS on a single machine without any networking, which is almost at Nasdaq speed, where the trading servers handle up to 500,000 TPS.

Out in the wild, Solana’s testnet has reached bursts of 59,400 TPS, making it “faster than Visa.” In lab environments, 50 nodes were able to conduct 111,609 TPS on their mainnet. Real-world speed in a distributed system with nodes spread across the globe is, of course, affected by available network speeds. 

Table: Solana TPS in a lab environment

Solana’s white paper claims that the theoretical limit to its capacity is even higher than 400,000 TPS and will continue to increase as network speeds and node processing capacity rise and network latency shrinks.

Solana’s performance is achieved without sharding, which is the approach that Ethereum will implement in its next iteration. With sharding, a blockchain is split up into multiple pieces that work in parallel. Still, it introduces complex problems for DeFi when assets processed on different shards are composed.

Time to finality

Transactions are only deemed final after three to 12 validators have confirmed them, depending on the desired security level. The time it takes for these three to 12 confirmations is called the time to finality. Solana takes five seconds on average, with outliers at 12 seconds — a long time for internet standards. 

Research by email client Superhuman revealed that users experience delays of more than 100 milliseconds as noticeable friction. Rival layer-one blockchain Avalanche boasts only 1.3–1.6 seconds to finality.

So it seems that Solana, despite its high speed, could never reach the point where it could build a decisive lead over its competition.

This article is an extract from the 80+ page Scaling Report: Does the Future of Decentralized Finance Still Belong to Ethereum? co-published by the Crypto Research Report and Cointelegraph Consulting, written by ten authors and supported by Arcana, Brave, ANote Music, Radix, Fuse, Cryptix, Casper Labs, Coinfinity, Ambire, BitPanda and CakeDEFI.

Low fees are one of the existential selling points for Solana. As with Ethereum, the actual fee is a function of supply and demand. When demand for block space rises, the price to include a given transaction in a block appreciates accordingly. 

Solana features a much higher transaction capacity than Ethereum. We’ll cover just how much in the chapter on theoretical transactions per second.

A look at network explorer Solana Beach reveals transactions cost between 5,000 and 10,000 lamports. One lamport equals one-billionth SOL. In dollar terms, the average Solana transaction has cost $0.00025.

Actual transactions per second (TPS)

Between 2,000 and 3,000 TPS are conducted on the Solana network at the time of this report. This number dwarves Ethereum’s 35 TPS by almost two orders of magnitude.

Figure: Solana transaction breakdown

On Solana, 80%–90% of all transactions are used for voting and synchronization, so this number is misleading on its own. Other blockchain projects have ridiculed Solana for its inflated numbers in the past.

Before comparing apples to apples, voting has to be factored out of transaction counts since Ethereum nodes don’t vote. Assuming the upper bound of 90% votes, Solana would still process 200–300 TPS or 10x Ethereum at a fraction of the cost.

This article is an extract from the 80+ page Scaling Report: Does the Future of Decentralized Finance Still Belong to Ethereum? co-published by the Crypto Research Report and Cointelegraph Consulting, written by ten authors and supported by Arcana, Brave, ANote Music, Radix, Fuse, Cryptix, Casper Labs, Coinfinity, Ambire, BitPanda and CakeDEFI.

Solana has a unique consensus mechanism called TowerBFT and proof-of-history (PoH). Co-founder Anatoly Yakovenko, with a background in distributed systems design, thought hard about blockchain scalability problems in 2017 after Bitcoin transactions took days when demand surged. 

According to an interview with Acquired, he discovered that most consensus issues vanish when the systems involved agree on a common timeline. Take the dreaded double-spend issue, for instance. In a synchronized system, you can assume that the first transaction is valid and the second is thus fraudulent.

Solana implements a surprisingly straightforward method of synchronizing nodes. It uses a sequential hash that runs over itself continuously, creating a rhythm that all nodes follow.

Proof-of-history uses recursive calculations where the previous output is used as the next input. Only with the output of the current function “X” will a validator be able to calculate the output of the next function “Y.” All validators need to solve the same function “X” and then be able to derive the output for the next function “Y” around the same time. Like this, Solana creates synchronization across its network.

Figure: The proof-of-history flow of control

Besides PoH, Solana uses its version of the practical Byzantine fault tolerance (PBFT) consensus mechanism called Tower. PBFT is an industry standard.

Programming language

Solana uses Rust, a recent, functional programming language for programs that run on top of its blockchain and base layer.

Rust has seen a remarkable rise in popularity for blockchain applications thanks to its performance advantages. From a purely technical point of view, it seems like a clear winner compared to Ethereum’s Solidity.

However, the lack of tooling, libraries and knowledgeable developers means that many wheels need reinventing to get DApps off the ground. The advent of the Anchor framework has ameliorated that somewhat by reducing the amount of work necessary just to get started by 80%.

This next article will look at the question if Solana is able to scale. What is the network’s transaction speed in theory and practice? And what are the advantages and disadvantages of the design choices involved?

This article is an extract from the 80+ page Scaling Report: Does the Future of Decentralized Finance Still Belong to Ethereum? co-published by the Crypto Research Report and Cointelegraph Consulting, written by ten authors and supported by Arcana, Brave, ANote Music, Radix, Fuse, Cryptix, Casper Labs, Coinfinity, Ambire, BitPanda and CakeDEFI.

SOL holders enjoy a variety of options for putting their tokens to work. Non-custodial staking is available in the Exodus wallet or with the native Solana-CLI command-line tool. Staking rewards are around 6%–6.5% APY at the time of writing.

Custodial staking is possible on Binance Earn, Kraken and FTX and, typically, offers fewer earnings. Binance Earn offered 6.5% APY this November 2021.

Then there’s lending on platforms such as Solend or Tulip Finance. Even staked SOL can be lent on Tulip, albeit for a meager 1.79% yearly yield, while Solend offers 3.87% for supplied SOL.

Lending becomes more exciting when providing stablecoins. Solend offers 24% on USD Coin (USDC), and Tulip grants 15% APY on USDC-USDT pairs via Raydium.

Solana initial coin distribution breakdown

The degree to which Solana is decentralized was the subject of heated controversy on Crypto Twitter this autumn. The pièce de résistance is the number of tokens held by the team and by VC backers. Solana has an initial token supply of 500 million SOL with a yearly inflation rate of 1.5%.

Binance Research found out that the team holds 12.79%, and VCs bought 29.15% of all tokens during the seed and funding sale, a total of 41.94%.

Figure: Solana initial token supply distribution

The pie chart doesn’t include the $314.15-million token sale that Polychain Capital and a16z completed in June 2021, and the exact amount of tokens involved was not published. The exchange price for SOL was $30–$40 in the months ahead, though it’s probably fair to assume that a steep discount was applied, given the scale of the purchase. Presupposing a $20 token price, 15.7 million SOL or 3.14% of the initial supply would have changed hands. 

Staking validators have to pay transaction fees on voting and syncing transactions but earn staking rewards as well as block rewards. Running a viable validator requires a stake that produces rewards in excess of transaction costs. In September 2021, the minimum stake required had surpassed $1 million — a significant barrier to entry for new validators. 

Despite that, almost 1,200 validators are operational at the time of this writing. The top 19 validators control 33% of all SOL staked and could theoretically halt the network if they colluded.

This article is an extract from the 80+ page Scaling Report: Does the Future of Decentralized Finance Still Belong to Ethereum? co-published by the Crypto Research Report and Cointelegraph Consulting, written by ten authors and supported by Arcana, Brave, ANote Music, Radix, Fuse, Cryptix, Casper Labs, Coinfinity, Ambire, BitPanda and CakeDEFI.