Starknet: Analysis of the Advantages of Full-Chain Game Ecosystem Development and Quality Projects

Author: Jian Shu

After experiencing the death spiral of Gamefi, the market in 2023 has shifted its focus more towards full-chain games, with Starknet being the most robust ecosystem for full-chain games. So, why has Starknet garnered the support of numerous full-chain game projects? What advantages does it have in developing full-chain games? Additionally, which projects within this ecosystem deserve our attention? This article will delve into these questions.

For those who are not yet familiar with full-chain games, you can first read Jian Shu's previous article “Nine Questions About Full-Chain Games”.

Advantages of Starknet in Developing Full-Chain Games

On social media, Starknet has gradually been labeled as "the most popular development platform for full-chain games." In addition to announcing the migration of Loot Realms from Arbitrum to Starknet in 2021, there are several other full-chain game projects such as Cartridge, Influence, Matchbox DAO, Briq, and Cafe Cosmos.

What advantages does Starknet possess that make it worthy of numerous full-chain games going all in? I believe there are several key points:

Continuously Expanding Performance Advantages (High TPS, Low Gas)

In full-chain games, the blockchain serves as the game's server, and all player actions must be completed through interactions with smart contracts. Therefore, for a chain to develop a full-chain game ecosystem, it must meet the performance requirements of high TPS and low Gas. Starknet, which is continuously upgrading, has the potential to meet these requirements.

From a foundational algorithm perspective, Starknet employs the core algorithm STARK, which is more suitable for large-scale computational applications. Its average verification time and proof size will decrease as the scale of verification expands. As shown in the figure below, when the hash invocations are 3072, the verification time is 40 milliseconds, averaging 76.75 times/millisecond; when the hash invocations are 49152, the verification time is about 60 milliseconds, averaging 819.2 times/millisecond, with proof size decreasing similarly. Therefore, in creating large full-chain games, the advantages of the STARK algorithm are evident, enabling high TPS.

Additionally, the recursive proofs used by STARK help reduce on-chain verification costs. In simple terms, recursive proofs can compress multiple upstream STARK proofs into a single proof, meaning that only the aggregated single proof needs to be verified on-chain to validate the effectiveness of multiple upstream proofs, which further reduces the Gas cost per transaction.

From the perspective of actual team actions, Starknet's Quantum Leap upgrade is also progressing rapidly. The mainnet has already upgraded to version V0.12.0, and the testnet has launched version V0.12.1, significantly improving TPS. Future transaction costs will also be continuously optimized.
Starknet Upgrade Roadmap (Source: Twitter user @100Sheng)

Native Account Abstraction and Its Infinite Possibilities for Enhancing User Experience

If the performance requirements for blockchain are the factors that previously constrained the development of full-chain games from a hardware perspective, then from the user's perspective, the constraint on the development of full-chain games is the even worse user experience compared to Gamefi.

Taking the recently popular Loot Survivor as an example, every time a user performs an "attack," "escape," or "upgrade" action, they need to pop up their wallet to sign and authorize the transaction, making the process quite cumbersome. Moreover, this is just a lightweight mini-game; if a large multiplayer online full-chain game has a similar experience, then there's no point in discussing playability further.

So, how does Starknet address this issue for full-chain games? The answer is native account abstraction. Unlike Ethereum's dual account model (EOA and CA), Starknet has a single account design called "account contract," which implements native account abstraction based on ERC-4337.

For those unfamiliar with the concept of account abstraction, you can read Jian Shu's previous article “Account Abstraction: Simplifying Your Crypto World”.

Specifically, a creative feature initiated by the Briq, Loot Realms, and Topology teams at the Starknet hackathon a year and a half ago—Session Key—utilizes Starknet's native account abstraction design, allowing users to operate in full-chain games without needing to sign for every action.

This idea was applied this year, with the Arcade Account developed by the Realms team on August 7 being the first implementation of the Session Keys feature. In the demonstration video of Loot Survivor using the Arcade Account released by the Loot Realms team, the improvement in user experience is evident: no need to sign for every "attack" and no long waiting times.

Of course, the enhancement of user experience in full-chain games through Starknet's native account abstraction goes far beyond just the Session Keys feature; there will be more possibilities in the future.

Crypto Game Engine: Dojo

It is well known that game engines are crucial for game development. With their support, developers do not need to create basic systems (physics simulation, graphics rendering, and basic game mechanics) from scratch for each game, allowing them to invest the saved costs and time into innovating game content.

To some extent, the massive growth of traditional commercial games can be attributed to two major game engines: Unity and Unreal Engine, with many familiar games like "Fortnite" and "Hearthstone" being developed on them.

From a developer's perspective, promoting the formation of a scale in the full-chain game ecosystem requires an engine that helps reduce the difficulty of game development. Therefore, in February of this year, the first full-chain game engine on Starknet, Dojo, was officially born, developed jointly by Loot Realms, Cartridge, and Briq.

However, unlike traditional game engines, Dojo currently does not involve advanced features such as physics simulation or 3D rendering, as full-chain game development is still in its early stages and faces more issues related to interaction with the blockchain's underlying technology. Therefore, according to official sources, Dojo is focusing on development in four areas:

• First, building a dedicated L3 for games to further enhance on-chain performance;
• Second, running fraud proofs within games so that players do not need to wait for on-chain verification for every action, enhancing game fluidity;
• Third, executing some proofs locally on the player's device, with only key ZK proofs uploaded to the chain, greatly improving game performance and player privacy protection;
• Fourth, utilizing storage proofs for shared assets. Storage proofs are similar to oracles; they provide evidence of information authenticity without requiring third-party intervention. Ownership of assets from different chains can be proven through storage proofs, allowing users to use their assets across different games on various chains.

As a foundational infrastructure for game development, game engines play a key role in the virtuous cycle of "helping developers build games—producing hits—attracting more developers." With the development of Dojo, the advantages of Starknet's full-chain game ecosystem will become increasingly apparent.

Programming Language Friendly for Complex Full-Chain Game Development: Cairo

In fact, before Dojo, there was another full-chain game engine elder brother: MUD, and in terms of both ecosystem quantity and technical maturity, MUD currently outperforms Dojo significantly.

So, why do we still need Dojo? One major reason is that, unlike MUD, which uses the Solidity language, Dojo employs a programming language more friendly for complex full-chain game development: Cairo.

Cairo is a smart contract language designed specifically for Starknet, based on STARK. Every coin has two sides; while Cairo has a certain learning curve for developers accustomed to Solidity, it is considered more modern. Many developers find it similar to Rust, offering more advanced and abstract functionalities.

For example, the previously mentioned generation of STARK proofs, recursive proofs that can reduce on-chain verification costs, and executing some proofs locally on the player's device are all implemented through Cairo. Additionally, Cairo is more versatile than Solidity; it is not only suitable for Starknet contract development but can also run provable programs on other large servers or personal devices, offering broader future applications.

Therefore, for complex full-chain games like large multiplayer online games, Cairo will have a stronger advantage. This will also attract more developers to Starknet to realize their dreams of creating complex and playable full-chain games.

Notable Projects to Watch

Having understood the advantages of Starknet in developing full-chain games, let's take a look at the notable projects currently in the Starknet full-chain game ecosystem.

Loot Survivor

Loot Survivor is a lightweight mini-game launched by Loot Realms, currently running on the Starknet testnet.

• Play to Die Mode

Unlike the various "X to Earn" models adopted by Gamefi, Loot Survivor uses the "Play to Die" mode designed by the Loot Realms team. In simple terms, players need to spend a small amount of funds to mint an on-chain adventurer character. If the adventurer dies in the game, the character's life will be forfeited, and the player will need to recreate a character and start the game again. Currently, there isn't a suitable Chinese translation that can vividly express this mode, so I tentatively translate it as "No Death, No Rest" or "Fight to the End."

Compared to the "Play to Earn" model that attracts greedy players through wealth effects, the "Play to Die" model is more sustainable and can foster a thriving creator ecosystem. In this model, if the adventurer dies in the game, the game's creator will receive a share of the funds initially used to mint that adventurer.

Thus, in the "Play to Die" model, the quality and popularity of the game will determine the developers' rewards. If the game is not fun, players will not mint on-chain characters, and game developers will not receive rewards. This model will help create a positive, prosperous, and bilateral market that includes both players and creators.

After Loot Survivor goes live on the mainnet, it may use its native token $Lords to mint on-chain characters, empowering its token.

Influence

Influence is currently the most popular large multiplayer online space strategy game on Starknet, also running on the testnet. According to the game's design, players can colonize asteroids, build infrastructure, accumulate resources, and battle with other players. However, overall, this game is not as easy to operate as Loot Survivor, and I encountered many issues during my experience.

• Team Situation

Founder Chris Lexmond has served as CTO and VP of technology at several tech startups, and other team members also have backgrounds in producing AAA games. As a startup team, their capabilities are promising.

The project team has also stated that when their native token SWAY is issued, 1% will be used to reward testnet participants.

Briq

Briq can be likened to an on-chain Lego system, where players can freely use blocks to build their desired structures and can mint their models into NFTs. If they wish, they can also recombine or dismantle the minted NFTs, offering high combinability.

Briq actually has significant potential; it resembles a foundational primitive protocol that could fully integrate with a full-chain game in the future, allowing users to use NFTs minted on Briq within that game.

Cartridge

Cartridge is a full-chain game integration platform on Starknet, and players often compare it to the Steam platform for full-chain games. However, it goes beyond just serving as a game integration platform; its developed Cartridge Controller not only helps full-chain games easily go on-chain but also improves players' gaming experience; it participated in the development of the full-chain game engine Dojo; and it is creating its own full-chain game Roll Your Own (RYO).

Cartridge completed its seed round financing in June 2022, with investors including Fabric Ventures, Valhalla Venture, and Chapter One. I believe that the multifaceted Cartridge will play an indispensable role in the future Starknet full-chain game ecosystem.

Of course, in addition to the four mentioned above, there are many other potential full-chain games on Starknet. Interested readers can refer to the table below; due to space constraints, I won't elaborate further here.

Summary and Reflection

At this year's ETHCC, Will Robinson from AllianceDao raised four issues constraining the development of full-chain games in his speech: user experience and user interface problems (UX/UI Problems), scalability issues (Uncertain How to Scale), lack of game design standards (No Standard Design Patterns), and uncertain value (Uncertain Value-Add).

Based on the advantages of Starknet in developing full-chain games discussed earlier, it can be seen that the first three issues raised by Will Robinson can be effectively addressed in the future. However, regarding the last issue, do full-chain games have value? It is well known that blockchain addresses issues of asset ownership and economic system establishment in the virtual world. So for games, aside from putting in-game assets on-chain, do we really need to move the entire operation of the game onto the chain? Does this truly matter to players? This is not a problem faced solely by Starknet; it is a common issue for all participants in the full-chain game ecosystem.

At the end of the article, I want to step outside of Starknet and attempt to answer this question. First, we cannot simply assume that full-chain games are reasonable just because they are conducive to innovation; such reasoning has clear logical flaws. While I was pondering this, a story about Vitalik that I stumbled upon inspired me.

As a child, Vitalik was obsessed with playing "World of Warcraft," but he eventually completely abandoned the game. The reason was that Blizzard modified some character skills in an update, claiming it affected game balance, which made Vitalik quite angry. After multiple unsuccessful attempts to contact Blizzard's engineers via email, he stopped playing the game altogether and realized that there was severe centralization in traditional games, where players had very little say.

Of course, in recent years, players' voices and freedoms in traditional games have been gradually increasing, thanks to the deepening understanding of user-generated content (UGC) concepts. However, this freedom is clearly still insufficient; players need true game autonomy. I believe this is the opportunity and significance of full-chain games.

Full-chain games can elevate UGC to a whole new level; they build an immutable core gameplay on the blockchain and maximize the design of the game to be handed over to players, allowing for infinite extensibility in content creation. For example, in the full-chain game Dark Forest, players can do anything they want as long as they follow the most basic rules. They can create their own asset trading markets in the game, develop tools to help automate their gameplay, and even create new games within the game. Currently, there are nearly a hundred Dark Forest plugins available on the market, allowing players to visualize opponents' territories, automate wars, facilitate trades, and more.

However, this is not the final state of full-chain games, as the aforementioned operations can only be realized by those with coding knowledge. But I believe that in the future, ordinary players will also be able to build everything they imagine within full-chain games.

Because we believe, we see.