After many years of development and iteration, a multi-faceted crypto ecosystem has finally started to emerge. For sectors like DeFi and NFTs in particular, the innovation shows no signs of slowing down. Rather it’s the opposite. Each new application and use case builds upon what came before. None of this would have happened without Ethereum, which settled $1.5 trillion of value in Q1 alone. Ethereum is where much of the liquidity, tooling, and users are. It’s where the action happens. For organizations interested in pursuing a NFT strategy of their own, Ethereum is arguably the best ecosystem to launch it on. There’s just one big problem at the moment holding, it and your plans, back from prime time – scalability.
The decentralized nature of Ethereum provides the network with its primary benefits. Namely its security, immutability, and transparency. But all of that comes at a cost, as the network can only handle so many transactions at a given time. When demand exceeds capacity, transaction fees rise, pricing out less valuable applications and smaller users. As a business, the costs you pay and the burdens you place on your users matter deeply. It’s easy to understand why this could be a cause for concern. However, don’t let this deter you. There are scaling solutions for Ethereum and alternative networks that can help you accomplish your goals. The purpose of this post is to compare those options across key criteria, so that you’re better equipped to answer this question – where do you want to build the future of your digital brand?
While there is no shortage of possibilities to go with, in some form or another, our intent with this comparison is to be focused rather than exhaustive. As it stands now, there are relatively few production-ready NFT solutions on the market. In particular, we will be looking at Ethereum’s base layer, Immutable X (Starkware/Volition), Polygon (PoS Chain), Flow, and WAX (built on EOS’ L1). Generally speaking, other alternatives are still in development, have yet to gain sufficient traction, or were bespoke solutions. At the end of this post, we’ll include an honorable mentions section to shine light on some of these. Before we progress further, we’d also like to touch on environmental concerns due to some recent NFT criticism on that point. You can take comfort in knowing that all of the solutions covered in this post are either already green or moving towards it (i.e. Ethereum PoS).
At a high level, we evaluated these solutions across key criteria that all factor into the end user experience. Now, for our grading to make sense, let’s first zoom out and see how all of this looks at a network level.
As we alluded to earlier in the post, we place a premium on the Ethereum network. The diagram above illustrates how these various solutions exist relative to it and certain attributes of each. Namely, security and compatibility. On the left hand side you’ll notice the labels L1, which refers to a “layer 1”, and L2, which refers to a “layer 2”. Simply put, a layer 1 is a standalone network with its own set of validators and security guarantees, whereas a layer 2 is a scaling solution built on top of an existing layer 1, with security guarantees very close to the underlying network it sits on top of. For those new to the crypto scene, having strong security in this context means that the validators running the decentralized network can’t easily manipulate or steal from users.
It’s important to emphasize that security varies significantly across the different L1s, with Ethereum far superior relative to the others. Immutable X, built using StarkWare‘s new bleeding edge scaling technology, is the only true layer 2 (“L2”) technology shown. This allows it to offer throughput orders of magnitude higher than layer 1 Ethereum, while at the same time possessing similar security to it. Polygon PoS, Flow and WAX (which is built on EOS’ layer 1), offer weaker security assurances. We should note that Polygon does have a live Plasma implementation that can be considered a layer 2 solution with better security guarantees. However, it doesn’t currently support any NFTs (although it could in the future), which is why we’ve disregarded it for this post. Instead, we’ll focus exclusively on their PoS chain. Polygon also has plans to support other L2 solutions in the future, such as Optimistic and ZK Rollups. Again, since those are not live yet, we’ve chosen to disregard them for this analysis. While the timeline for that is uncertain, if/when Polygon adds support for rollups, it would place it in a similar position to Immutable X from a security and throughput perspective, depending on their implementation. If you’re interested in learning more about Polygon, we recommend checking out our recent deep dive on it here.
A crucial consideration for blockchain scaling, NFT-focused or otherwise, is that there is a fundamental interplay between scalability, security, and decentralization. This is known as the scalability trilemma, which suggests that in order to increase one, you necessarily make trade-offs in the others. For the purpose of simplifying this report, and prioritizing attributes we believe the target audience of this post cares most about, we essentially treat decentralization and security as synonymous, given that the latter is often a consequence of the former.
While the trade off between security and scalability holds true for each of the platforms shown, the magnitude of the trade off can vary materially between them, based on the specific implementation. This dichotomy is best illustrated by comparing Ethereum and Immutable X. Ethereum’s base layer (“L1”) is largely optimized for decentralization and security, which limits its throughput to ~15 transactions per second (“TPS”). Contrasting this, Immutable X has marginally weaker security (which can also vary based on the Volition choice) but, importantly, can facilitate ~9,000 TPS. This is the promise of L2 solutions and one reason to be optimistic about Ethereum’s future.
While estimates vary significantly for throughput potential, Polygon’s PoS chain was able to reach 7,200 TPS on testnet with 100 validators. That’s an impressive amount of scalability at that level of decentralization. On mainnet, we’ve seen instances of 300+ TPS during periods of high demand, but there needs to be more stress testing before we know the true upper limits. We’d prefer to score this purely based on Mainnet figures but, to avoid penalizing Polygon unfairly, we assumed throughput is closer to to the testnet TPS, placing it in 2nd place behind Immutable X. For security, Polygon comes in 3rd place since its PoS chain has 99 validators and joining is permissionless. However, we should note that there is a high degree of validator centralization with Binance accounting for 37% of total staked supply.
EOS, the layer 1 that WAX is built on, reached an all time high of ~4,000 TPS which places it in 3rd place for throughput. Here, however, the scalability tradeoffs are far more pronounced, evidenced by the fact that the network only has 21 block producing validators. The opinion of our team, and others as well, is that EOS is prone to collusion, explaining why we ranked it last in terms of security. Finally, Flow clocks in at ~100 TPS. That level of throughput is probably less than what will be required for mainstream adoption, but it would be a respectable number if achieved by a truly decentralized network. Oddly enough though, that isn’t the case. While Flow has 140 consensus validators, the vast majority are run directly by the Flow Team themselves. Furthermore, you need permission to run a node in the first place. As a result, Flow is highly centralized in its current form. With this in mind, you might be wondering why we ranked it slightly ahead of WAX with regards to security. We believe that, in the future, it will be able to decentralize more easily than EOS and, in our opinion, Dapper Labs (the company behind Flow) is less likely to cheat its users than the EOS cartel.
Next up, we assessed these platforms on asset mobility and ecosystem support. In this context, asset mobility refers to the ease in which the NFTs can be moved from one chain or scaling solution to another (e.g. withdrawals to Ethereum L1). Ecosystem support looks at the number of quality applications that can easily plug into the solution and utilize the assets created via it (e.g. OpenSea support). For platforms that offer both, token holders will benefit from having the freedom to move their assets wherever they please and experience greater utility from how they may be used.
Starting off, nothing surpasses Ethereum in terms of having a vast ecosystem of applications and numerous asset bridges. With this in mind, you can see why a solution that interoperates closely with Ethereum can also benefit from these attributes as well. In an earlier graphic, we showed a high level overview of where the different solutions sat at a network level. Let’s revisit that now with some added context.
The dark purple color annotates EVM compatibility. Since Polygon uses Solidity, which is the same smart contract language as Ethereum, anything deployed on Ethereum can also be easily deployed on Polygon. How easy you may ask? Think “copy + paste”. This is a strong advantage for Polygon because it has minimal switching costs for applications looking to move over and allows the developer experience to effectively mirror that of Ethereum. To date, this has led to a flourishing DeFi ecosystem on Polygon which continues to grow. Polygon’s compatibility with Ethereum is what earned it 2nd place for ecosystem support.
Flow and WAX are designed differently and thus don’t directly benefit from the resources and tooling available to Ethereum developers. Those networks also have different token standards which, when coupled with lower security guarantees, may lessen the pricing premium that the market is willing to assign to those assets, as we’ll discuss later on. Neither have bridges to Ethereum at the moment, but we’ve ranked WAX higher than Flow in asset mobility because a bridge is under development.
Immutable X is in a unique position here. The Starkware-built L2 it lives on implements the newly developed CAIRO programming language, meaning there aren’t that many developers familiar with it yet. In addition, existing code can not be easily ported over from Ethereum to it, like it can with Polygon. With that being said, we’ve labeled it “Semi-Compatible” in the diagram above because it interoperates with Ethereum and utilizes the same token standards. Regardless of where the NFT is initially created, whether it be on Ethereum, Polygon, or Immutable X, the assets can flow between them, albeit at varying speeds. From a technical standpoint, Immutable can facilitate faster NFT withdraws back to Ethereum, without compromising on security, giving it an edge over Polygon in asset mobility.
With regards to ecosystem support we should note that, in actuality, the NFT sector has relatively low composability at the moment. There are multiple reasons for this such as the assets being more unique (they are non-fungible after all) and because there is often less frequent / reliable pricing available. When we talk about ecosystem support, we’re really focusing on applications like prominent NFT marketplaces. If you’ve ever looked at a diagram of a Yearn vault strategy, then you already know composability plays a much greater role in a sector like DeFi. Polygon, which has high composability just like Ethereum, showcases this well. It’s been able to quickly form an impressive DeFi ecosystem and accumulate over $3.5B in liquidity. This is certainly worth noting on the topic of ecosystem support.
However, the growth of Polygon’s DeFi ecosystem could also prove to be a double edged sword for NFT projects. Applications moving to Polygon still end up competing for scarce block space on its PoS chain and DeFi activity, if it continues to grow rapidly, could come to dominate the chain. It ultimately depends on what the true, max throughput is for Polygon’s PoS chain and how DeFi demand grows. This may not be an issue at all, but it is possible lower-value NFT activities get priced out, just as they have been on Ethereum L1. Of course, NFT-specific activity could be moved to another Polygon scaling solution in the future too. Compared to Immutable X’s scaling architecture, this is less of a concern given its NFT-focus and batching efficiencies at scale. Its L1 on-chain footprint (i.e. the proof size) grows logarithmically. In layman’s terms, the more transactions there are being batching, the more cost efficient it becomes.
Now that we’ve had a chance to talk through each of the solutions and their unique attributes, let’s move on to our scorecard. As seen below, we’ve ranked the solutions across a variety of key factors relative to each other. A score of 5 represents the best option in its respective category, while a score of 1 represents the worst option. In an unsurprising twist, Ethereum dominates every category except for throughput, which is where the alternatives have their edge. It should go without saying that there is certainly a level of subjectivity here, although we’ve tried our best to provide justification. Below, we’ll run through each of these categories, what they mean, and what factored into our scoring.
To begin, throughput was relatively straightforward to grade since it can be easily quantified. Immutable X is able to achieve 9,000 transactions TPS, Polygon 7,200 TPS, WAX 4,000 TPS, Flow 100 TPS, and Ethereum 15 TPS.
Next, we assessed security. A significant portion of this post has already been dedicated to this topic so we’ll keep this brief. The Starkware-built layer 2 that Immutable X lives on can offer security guarantees close to that of Ethereum, although this can vary, which is why its ranked 2nd. Polygon came in 3rd place given its permissionless set of ~100 validators. Flow came next, despite being highly centralized at the moment. As we mentioned earlier, Flow will likely become more decentralized as time passes and we personally believe there is less of a chance for manipulation from the Dapper Labs team, than there is from the EOS cartel of 21 validators that WAX relies on.
Wallets, despite being relatively easy to increase support for, as we’ll soon discuss, nonetheless have an important, direct impact on user experience. Polygon comes in 2nd here given the wide array of options it currently supports, even though users need to do some additional configuration. Immutable X can support any Ethereum wallet with minimal lift falling on the user but, since only MetaMask is currently available, we had to rank it 3rd for now. We decided to rank WAX ahead of Flow because they’ve built a very intuitive cloud wallet whilst Flow still appears to only offer custodial solutions.
Fiat gateways are likely a top priority for mainstream brands given that most of their audience doesn’t transact with cryptocurrencies, yet. The ability to directly use credit cards was a major contributor to the success of NBA Top Shot and justifies why Flow was ranked 2nd. We should note, however, that card support can negatively impact UX due to chargebacks. Polygon comes in 3rd for fiat gateways, given the options it offers and the ease in which liquidity can flow to it from Ethereum. WAX was next given the functionality available in its cloud wallet. We had to rank Immutable X last in this category for now since it currently lacks fiat gateways, but this will change soon.
We approached developer experience based on how easy it is to build. Many things factored into this such as detailed documentation, available Github repos, adoption of the programming language used, existing tools, etc. Ethereum (Solidity) reigned supreme, followed closely by Polygon (Solidity) given its strong EVM compatibility. After that, the developers we interviewed believed the order ranking should be Flow (Cadence), WAX (C++), and Immutable X (CAIRO). It’s important to note that CAIRO is a very new language and its adoption should increase with time.
Ecosystem support refers to how easy it is to leverage existing building blocks for each solution, as well as how widely integrated that ecosystem is. Once more, Polygon’s EVM compatibility shines through in 2nd place. Next we went with Immutable X. As noted earlier, while it’s not EVM compatible in the same way that Polygon is, Immutable X is still able to benefit from much of the existing Ethereum ecosystem. EOS, while nowhere close to Ethereum’s ecosystem, still has its perks which WAX stands to benefit from. Flow doesn’t have the same type of open ecosystem or external marketplace support yet which is why we’ve ranked it last.
We believe that existing brand power is a strong consideration for attracting new IP to a given platform. As such, Flow takes second place behind Ethereum here given the massive success of NBA Top Shot and other IP in their pipeline. Immutable X follows having already attracted several strong game brands. Polygon comes in next since it’s been able to attract several DeFi blue chips and NFT projects like Aavegotchi. Lastly, we have WAX which has failed to attract any significant brand power outside of more recent Topps MLB NFTs. WAX’s lower historical transactions by purchase amount is reflective of hosting primarily cheap and lesser quality products, which in turn attracts less capital flows and interest.
Lastly, we define asset mobility as the ease in which the NFTs can be moved from one chain or scaling solution to another (e.g. withdrawals to Ethereum L1). Technically speaking, Immutable X can facilitate withdraws to Ethereum that happen in the next block without sacrificing on security. This is followed by Polygon which can take up to 30 minutes for NFT withdrawals from its PoS chain. We ranked WAX in 4th place as they are actively building a bridge to Ethereum which has been on their roadmap for sometime. Plans for a two-way bridge with Flow remain unclear, so they are placed last.
While the scores above are helpful on their own, they lack nuance and context. Not all of the categories are equally important and some are more difficult to improve on moving forward than others. For example, its far easier for a solution to increase their wallet support than it is to suddenly offer better security and throughput. The latter would likely require an overhaul and rebuild of the design architecture. With this in mind, we set out to rank the categories themselves so that we could use weightings to adjust the scores from earlier. The end goal was to make the scoring more useful for our readers. The framework we arrived at can be seen in the graphic below.
In accordance with the above weightings, we adjusted the scores and final rankings of the solutions in question. Once again, please note that there is a degree of subjectivity to the weightings but we have made an informed estimation to the best of our abilities.
As shown above, Immutable X received the best composite score after Ethereum, largely driven by its high throughput and strong security assurances. It excels in all of the categories which are most important and hardest to improve upon. With that said, Polygon came in closely behind it. While Polygon didn’t receive the best score in any single category, it also didn’t receive the worst score in any either. In all the ways that matter, Polygon is a balanced solution without any major flaws relative to the field. Flow came in next, driven by the two factors that made NBA Top Shot such a success – its brand power and fiat gateways. Lastly, we have WAX where, aside from having solid throughput, it was a laggard everywhere else.
It’s important to note just how quickly things can change in this market, where the ground regularly shifts beneath our feet. As shown on the chart below, Flow hardly registered any metrics until January 2021 when it exploded onto the scene in spectacular fashion. We are encouraged by the mainstream attention NFTs have been able to draw, and are eager to track this fascinating landscape as it evolves.
In the context of NFTs we are often confronted with the question – “to what extent does decentralization and security truly matter?”. It’s an interesting point, particularly when said NFTs rely on IP owned by centralized companies.
Regarding decentralization, the secular trend of this decade has been a shift towards user-owned communities in place of rent-seeking middlemen extracting value. The desire for true digital ownership on open platforms is increasing, whist walled gardens with lock-in effects begin to pale in comparison. The emergence of credibly neutral, shared digital infrastructure has given rise to NFTs. It’s this foundation, and the possibilities it can enable, which makes NFTs appealing and distinct from other forms of digital ownership that have come before. NFTs are effectively atomic units of “ownable internet”, which can be used to express a stunning variety of digital items and imbue them with their own histories, provenance and more complex properties. We believe that the credible neutrality of the networks they’re issued on is a core driving force behind what makes them valuable. These networks are attractive because they are owned by everyone, and yet no one, simultaneously. Does the market actually value this property as we do? The graphic below seems to reaffirm that they do.
The best way to articulate why security is important to both your brand and users is to highlight the consequences from it being absent. Manipulation, theft, and a lack of faith in the future existence of the NFTs in question all come into play. For large IP holders considering putting AAA brand assets on-chain, the risks associated with poor security should not be casually overlooked.
Ultimately, if NFTs truly succeed at gaining mainstream adoption, the end user probably won’t even be aware of the underlying technology that makes them possible. Nor should they be expected to. A good user experience abstracts away much of the intricacies which we’ve just discussed. However, these attributes do matter. We hope this post has proved helpful in explaining why. Now that you know the tradeoffs you need to consider, where do you want to build the future of your digital brand?
Honorable Mentions
Although this report primarily focused on production ready NFT solutions with traction, we wanted to briefly touch on other options we didn’t have a chance to cover in-depth. A notable one is Binance’s NFT marketplace which is coming in June and should benefit from Binance’s powerful distribution channel. It’ll run primarily on Binance Smart Chain (a L1) but Ethereum will be supported as well.
Solana is another layer 1 offering high throughput that we’re starting to see NFT projects commit to building on. Rollup solutions like zkSync and Arbitrum are also planning NFT support but the primary focus still appears to be building out robust DeFi ecosystems.
We are encouraged to see projects such as Moonbeam and Unique working to deliver Ethereum-compatible solutions built on Polkadot. Lastly, we’d like to also note the existence of other ecosystems such as that of Forte who are focused on building tooling for game developers to integrate blockchain mechanics in their games. As it stands, there is relatively little public information on their approach and technology stack. What is known is that they opted to use a fork of Ripple’s Interledger. We are unsure of this approach given the focus of the current environment.
Crucially, the window of opportunity for alternative NFT platforms may be closing as Ethereum’s network effects continue to deepen while performant solutions now exist that don’t compromise on compatibility.
