Dream Westward Journey "Treasure Pavilion" Web3 Game Revelation: Marketplace, Asset Value Measurement Dimensions, Meta-Economic Interactions.

Author: Aiko, Researcher, Folius Ventures

It was a very ordinary Sunday. I had about two research projects going on simultaneously: one was a company research report, and the other was research on AI. To find more inspiration and material, I spent about two hours browsing the NetEase Treasure Pavilion, and I have some insights I would like to share. Since I wrote quite a bit, it turned into an article, so I plan to publish it here.

This article will be divided into two parts:

1) An overview of the differences between web3 game marketplaces and the Treasure Pavilion, along with suggestions on what a game asset marketplace should look like.

2) A discussion of the concept of "meta-economic interaction" and its extension into the field of AI + gaming.

Inspiration from the Treasure Pavilion for Game Marketplaces

I am particularly interested in transaction taxation. In fact, the previous deck from the company specifically explained "IAT" (In-app Taxation) as a new commercialization model. However, I have never written about marketplaces in the gaming vertical. Deep down, I have always been skeptical about the possibility of a trading platform created by a third party that only has simple game asset aggregation functions being established in the past, present, and near future.

My research on the Treasure Pavilion reaffirmed my previous thoughts: the methods of centralized marketplaces conflict with the essence of decentralized trading, which pursues high frequency and zero fees.

1) As a first-party self-operated platform, the value of all game assets in the Treasure Pavilion is supported by the player DAU and transaction volume behind it. Thus, the first step in ensuring a marketplace has demand is to ensure that assets have a broad audience willing to pay a premium.

2) Customizable features do not work well in the rapidly shifting hotspots of web3. A customizable marketplace needs to accumulate user data, such as player preferences and demands, which requires a stable user base and long-term operation of the game to effectively customize. However, the ROI-driven web3 gaming hotspots may not be able to keep pace with the speed of shifting trends.

The Treasure Pavilion has a high degree of customization. Each game has its own market and special operations. While the display interface functions are similar, various lottery systems, special events, appraisals, discount areas, servers, and seasons are completely different. For a specific game, there may be a shopping assistant guiding you to choose race, character, profession, and budget step by step. Only long-term operated projects can truly understand what their players need to better customize their offerings.

3) The platform's revenue and user experience come from the transition from inconvenience to convenience, from low turnover rates to high turnover rates. Zero transaction fees and high liquidity do not benefit the platform's revenue or stabilize asset prices.

Players familiar with NFTs may understand how the emergence of Sudo Swap and Blur over the past year has profoundly impacted the NFT market. Zero transaction fees and bulk trading have shaken the foundation of NFTs as luxury goods or symbols of faith, witnessing large holders dumping dozens of monkeys, leading some holders to doubt the significance of community faith in such capital operations.

In our previous two decks, we discussed how to increase revenue by creating transaction friction. The Treasure Pavilion also has many similar designs, such as rapid transaction services: it can shorten the original review time of four days to zero days and reduce the secondary transaction time from eight days to zero days. This service charges sellers an additional 4% fee. In the Treasure Pavilion, both sellers and buyers need to pay transaction fees. Another service that benefits buyers, such as a public reservation period, allows them to pre-purchase game assets, which requires buyers to pay a 5% reservation fee of the total price.

4) Designing gamified mechanisms that satisfy human nature is part of the marketplace delivery.

Many web3 NFT marketplaces focus on transaction forms and matching efficiency, and games are no exception. However, using a crypto asset mindset to create a marketplace for game assets clearly overlooks the strong demand scenarios inherent in games, namely that users have a strong interest in participating in gamified selling activities.

Some games in the Treasure Pavilion have arranged price-increasing lottery systems, where the chances of winning increase with higher bids, which also provides users with a game of expected value. It is speculated that due to regulation, this mechanism cannot be openly designed within the game, but moving it to the trading market makes more sense, increasing user transaction frequency while also improving the platform's profit margins.

In fact, similar suggestions for adding various gamified features to domestic e-commerce platforms have long existed in the NFT trading market, such as Taobao Orchard or lotteries, etc. However, looking back now, there are no major trading markets adopting these ideas, with only Rollbit embedding NFT unboxing as a unique gameplay unit in its large casino, which precisely represents the opportunity for emerging gamified marketplaces.

5) If there is no demand, artificially create demand. The difference between this point and the third point is that the third point can be achieved solely through the platform's trading rules, while the fifth point emphasizes human intervention, where a group of people can be gathered to support a new trading scenario when there is a need.

For example, the Treasure Pavilion has appraisal activities, where the official recruits a group of relatively authoritative players within the game ecosystem to act as appraisers. Items certified by these appraisers sell faster. This group of appraisers can be seen as cost calculation experts, helping sellers assess asset prices and provide certification. After listing, the platform will likely provide more exposure and benefits such as better display positions. The fees paid by sellers to appraisers will certainly be subject to platform fees, and this money is similar to advertising costs for content. The platform has recruited a group of people called appraisers to act as a buffer, making this fee appear more reasonable and softer.

Before understanding "meta-economic interaction," we need to establish a new "value measurement dimension" for game assets.

First, the assets traded in the Treasure Pavilion sparked my thoughts:

The items traded in the Treasure Pavilion are relatively "large," often involving an account that contains multiple characters and skins.

It can be observed that transactions in the Treasure Pavilion are no longer a unit of "asset," but rather a segment of experience. For games with rich layered experiences, a weapon and an equipment may enhance strength but cannot fundamentally change the gaming experience. Why do some people always choose different characters to replay a game? Because the most valuable aspect is the characters and accounts that can fundamentally provide new experiences. Purchasing an account means experiencing the progress within that account, akin to a life restart simulator, allowing one to experience what the world of high-level players is like.

In the deck we released in February, we also mentioned the importance of personal progression. In April of this year, we happened to discuss a team that allows players to package their game character's progression into NFTs for sale, which includes skill unlocks and skin unlocks. Unfortunately, I haven't seen such an interesting design for a long time.

How could those still obsessed with asset units (like 1155, 6551 protocols) understand the nuances? What matters is not how many parts you can divide the asset into or whether you can package it. Trading a worn sword or a grain of sand is irrelevant; what matters is identifying the "data sets" that can generate value perception and recognition for players and selecting these "data sets" as assets to be placed on the chain.

Furthermore, why are "large" assets more important than "fragmented" assets? Because they encompass multiple value measurement dimensions.

In-game asset trading can indeed serve as a form of experience. Many web3 games boast the freedom of in-game asset trading, with popular examples being certain minerals or military supplies on planets, etc., which I also mentioned in my previous blog (Game AMM). Setting aside how large the DAU needs to be to support the operation of such a trading system, pursuing trading in this direction may indeed lead to picking sesame seeds while losing watermelons.

During the subscription period, we used time (or pow) as a measurement standard, focusing on the stability of in-game soft currency. For example, when arguing that Dream of the Red Chamber's economic system is more stable than World of Warcraft's, we pointed out that Dream uses a point card system, while WoW uses a monthly card system. Therefore, Dream can better stabilize the value of every asset in the game and effectively control the inflation rate of gold coins (of course, there are many mechanisms worth referencing, and I recommend reading Frost's articles, salute).

Now, there are not many open economic games left, and not many people pay attention to the price of each item in detail. In other words, players who once cared about whether item prices were stable are now completely different from those who "spent several 648s" to recharge. F2P and Gacha have become mainstream, so the single value measurement dimension of time (pow) can no longer support trading. Instead, it requires proof of capital + proof of luck + proof of work + proof of skill. In simpler terms: I have money to draw, I can draw successfully, and I can maintain this account in a correct and efficient manner, which represents a multidimensional value.

In this complex value measurement dimension, players can roughly calculate the cost of nurturing such an account. However, bringing this account back into the game only yields basic resources and a segment of gaming experience. Even if sold, the price is generally equal to the original price plus the value converted from the amount the new holder has spent, while players earn time saved in cultivation and a brand new experience, rather than any direct economic return. This is indeed a very good commercialization method.

Moreover, this value measurement dimension gradually emerged after F2P. When games became "unfair," and players could spend money to buy value, these economically valuable figures, through randomness and strategy, ultimately acted on the account, forming a new type of asset more suitable for F2P players to trade. However, F2P Gacha has made too much money, leading everyone to compete in art and industrial capacity, without considering that the secondary market can also undergo secondary commercialization, providing an outlet for the piled-up character value.

So, what is meta-economic interaction?

First, let's assume a scenario: which is more suitable as a testing ground for AI gaming, "Rate of the Earth" or "Civilization VI"? (Without considering API interfaces, implementation difficulties, etc.)

To conclude: "Rate of the Earth" is better.

The resources in "Civilization VI" are earned through players' time and skills. Multiplayer connections are often limited by local networks, and participants are usually friends in real life, with entertainment and time-passing as the main goals. Therefore, there are many verbal negotiations or intimidation tactics, such as "let me borrow this" or "give this to me," which exchange emotions for resources. Additionally, the declared diplomatic relationships are limited and distant; countries can form friendly alliances but cannot involve more complex and in-depth class relationships, such as vassalage and tax payments.

In contrast, "Rate of the Earth" has two major features: 1) A general card-drawing system - the military strength and skills in "Rate of the Earth" largely depend on generals, which are obtained through paid card draws. This creates strong economic interactions within the game, such as sending red envelopes to the leaders of vassal regions every month. Players automatically calculate the cost of "drawing a general to initiate a war = paying salaries to five counties," leading to more economic interactions, often resulting in actions like sending red envelopes or salaries outside the game. 2) Alliance relationships - as a multiplayer mobile game with thousands of players on a server, "Rate of the Earth" has more nuanced social designs compared to a game like "Civilization," which primarily relies on single-player and local area network connections. The seamless large map of "Rate of the Earth" requires coordinated expeditions, making road-building and bridge-laying between allies particularly important. Alliances have resource distribution buffs, and siege gameplay requires cooperation among players of various economic strengths and classes, easily forming strong binding relationships around economics and territory.

It is evident that player payments significantly impact the dynamics and overall atmosphere of a multiplayer game, whether positively or negatively, creating an entirely different experience.

Payments open a channel connecting the game and reality, where players' financial resources and luck in the real world greatly influence the game. Moreover, bypassing in-game confrontations, there may exist countless economic exchanges and human interactions outside the game. Since assets are acquired through player spending, in-game values can reflect players' economic strength in reality. This economic strength can be input into the game as game values or used as leverage to negotiate with other players, transferring to other players' real lives, or re-entering the game as values through other players.

At this point, we should envision a scenario to distinguish game interaction vs. meta-economic interaction.

Furthermore, what interesting use cases are there for meta-interaction?

------AI agent meta-interaction

Imagine, within the value measurement standards we just defined, what can allow AI to participate?

• Capital (optimizing capital utilization & external game strategies)
• Time (saving player time)
• Skill (in-game strategies)

Now that AI has bidding and negotiation capabilities, if AI were to replace humans in managing in-game interactions and external interactions with other players, and each player has an agent that can communicate with other agents to negotiate conditions and prices, wouldn't that be fun?

Consider the following scenarios:

1) Each player has an initial agent with consistent game strategies and negotiation abilities; 2) Players need to allocate funds to their agents, allowing them to execute bribery or buy other agents; 3) Agents return selectable strategies to players every six hours, and players choose the optimal strategy for the agent to execute; 4) Agents can learn players' risk preferences from each strategy choice, optimizing the next strategy (e.g., whether to continue raising bids to buy other agents, whether to increase troops for siege under certain resource reserves, etc.); 5) Of course, a player can have more than one agent, independently training multiple agents to collaborate, such as a diplomatic agent and a tactical agent. Perhaps there will be a persuasive agent that can negotiate with other nations without deploying troops or spending a dime. 6) At this point, agents can become entities with "personality," possessing appearance, voice, and character, and the players who train these agents can receive a share of the agent's commercialization profits.

Thus, the original experience of games like "Rate of the Earth," which rely heavily on external social interactions and operate continuously with thousands of players, can transform into a more idle and lightly social experience (similar to single-player), with agents serving as the main interaction objects for players, potentially even commercializing into new game content.

Finally, why do we find AI games involving meta-economic interaction interesting?

Perhaps it's because I've been studying incentive mechanisms in the crypto field for too long, or perhaps influenced by zkml, I believe that future AI natives can manage users' assets for financial activities. Maybe it's due to a lack of interest in the interaction forms between generative/passive agents and humans after "Dwarf Fortress," or perhaps it's my obsession with viewing games as interactive art. I always feel that the current AI + gaming interactivity is still far from sufficient.

Undeniably, almost all AI + games today merely test the extent to which AI can mimic human behavior. Dota trains AI for micro-operations, Minecraft assesses AI's understanding of the physical world, and Stanford Town simulates AI's understanding of human behavior and emotions. So why hasn't AI simulated human economic interactions been placed in game experiments?

Regardless of the era, games must align with human nature. Observing AI's daily behavior can be seen as satisfying a voyeuristic desire. How long can this sense of satisfaction last, and how entertaining is it? Can it compare to the audiovisual experience brought by graphics cards and VR? Can it match the adrenaline and dopamine released when unboxing? Can it satisfy vanity when looking down on others in the entire server? If not, how can we use AI to create games that are more aligned with human nature and better stimulate human pleasure points? This is what I find interesting.

Additionally, I have been contemplating and researching the role of AI in narrative games. These are some insights I gained after completing "Cosmic Wheel Sisterhood" this week. Narrative is also one of the areas where I believe AI can significantly enhance experience and differentiation.

Conclusion

The purpose of writing this article is threefold:

1) To review the interesting mechanisms in the Treasure Pavilion, which may serve as a reference for teams designing marketplaces, highlighting some gaps in feasibility and upper limits between the Treasure Pavilion and web3 game vertical trading markets.

2) To explain the new value measurement dimensions for game assets. With the development of F2P to this day, players' entertainment consumption habits have undergone significant changes. Continuing to conduct game trading based on the logic of the subscription era is less effective than shifting perspectives and exploring other asset trading forms, identifying what the main value carriers and trading forms are, namely trading a segment of gaming experience (including accounts with multiple valuable characters).

3) If you are interested in the "meta-economic interaction" AI + gaming mentioned at the end of this article and are also engaged in research related to AI and gaming, feel free to contact me. We have more research to conduct.