Preface: Usually, for my blog posts, I just write stream-of-consciousness rambles, and don’t even bother proofreading. This is different - this is the first time I actually put in some effort. Why? To celebrate Mirror launching Writing NFTs on Optimism. I have no idea why anyone would buy NFTs for this post, but if you do: 100% of the proceeds will be directed straight to the Gitcoin multi-sig for public goods funding. I’d love to hear feedback - should I repeat this for my older posts, and future posts?

Blockchains have found their niche as a store-of-value, in DeFi applications, speculation (kwonzis?) and NFTs. Certain identity & reputation protocols like ENS are gathering steam too. Everyone’s asking - what’s next? Many are convinced its gaming. In this post, I’ll argue why blockchain gaming isn’t what many crypto people think; but also, how it can find an interesting niche.

Games are diverse

Before we begin, there are all sorts of games which couldn’t be more different from one another. Likewise, different gamers have different preferences. It’s impossible to write a post and cover all areas. So, when I talk about “games” and “gamers”, I’ll take a broad view.

Why we play games

In said broad view, I define games as a form of entertainment. Most gamers play games to entertain themselves, challenge themselves, in one way or another.

A great quote from Todd Howard that captures the spirit of games:

That’s one of the things people who don’t play a lot of games never quite understand. When you play a game and you accomplish something, that’s real. It’s a real accomplishment in your life, or it has been to me, and to other people I meet who love gaming. You can finish the week and say, ‘I saved the world’, and you legit feel that way. That’s the magic.

In almost all cases, it’s an escape from the hardest game of all - real life.

Money is the antidote to fun

Arguably, the most toxic element of real life is money. Picture this - you’re having a ton of fun in a game, and you need to acquire an item that’ll give you some abilities to - you guessed it - have more fun. Sure, the game world has its own rules with their own consequences, but you can be sure a mistake will never cost you irl. This liberating feeling is essential to gaming.

Now - if you add real money to the picture… I need say no more. There’s nothing that’ll suck away immersion in a more soul-crushingly debilitating manner than financializing games. It is critical that the game world’s economics and mechanics have a hard separation from the real world.

Play-to-earn is unsustainable, pay-to-win is loathed

By now, it should be clear that inflationary mechanisms like play-to-earn are economically unsustainable. Some would argue they’ll “find an equilibrium”, but in reality, it’s just a matter of time - there are a limited number of greater fools. Play-to-win systems are an improvement because it could be economically sustainable, but it’s also no fun. See: separation of game world and real world, as mentioned above. Pay-to-win already exists in gaming - in the form of microtransactions - and it’s safe to say they are heavily loathed by core gamers, as proven by the Star Wars: Battlefront 2 debacle. Since then, the games industry has done a sharp U-turn away from play-to-win mechanics, for very good reason. At most, there are cosmetic microtransactions, which have no impact on gameplay. Indeed, EA’s follow-up to Battlefront 2, Star Wars Jedi: Fallen Order, had no microtransactions of any kind, and turned out to be EA’s most commercially successful non-FIFA game in recent years. People paid money to play the game, and that’s it.

All that said, financial mechanics will always exist in some form, and they are normalized in some forms of gaming, particularly in the mobile and MMO space. Financialization is a great fit there - but it’s important to note that it will not work for every game. You can also have purely financial games, more akin to gambling or trading - and blockchains have three key advantages: borderless, regulation-circumventing, and most importantly, DeFi composability. But this is not what I think of as “gaming” - call it “GameFi” if you will.

Cross-game asset sharing and interoperability is hard - needs new infrastructure

Another common idea is that game assets from one game can transfer to another, and continue to function if a game shuts down. While a nice idea, in reality game assets from one game are not compatible with each other. You have to get developers of two different games to agree to the same standard. Besides, assets are of no use if the game shuts down anyway. Yet, common infrastructure that many games can use does have potential.

So, to enable these types of use cases in a decentralized manner, we need to build extremely complex infrastructure, a lot of it from the ground up. This is an incredibly challenging task, which may take several Lubins (note, 1 Lubin = 1,136 days or ~3.1 years) to build. The brave folks at Topology are taking on this very ambitious challenge.

Someone like Epic Games is better placed to accomplish this, by adding these features to Epic Online Services & Unreal Engine, and creating a standard SDK which different games can use. However, right now, there isn’t any demand for cross-game interoperability, so it’s a low priority. It’s much harder to accomplish this in a decentralized manner.

Decentralized gaming infrastructure is more than blockchains

Let’s assume that the decentralized gaming infrastructure above happens - a big assumption, as noted above. Putting it “on blockchain” is simply impossible. Games require several orders of magnitude greater compute & storage than the fastest blockchains will ever provide. For example, most “crypto games” currently are totally centralized, and just put NFTs on-chain.

The wider decentralized gaming infrastructure will need peer-to-peer game clients (run by the players), decentralized storage layers, dedicated execution layers etc.

Zero-knowledge proofs (& related tech like MPCs/NMCs) will play a key role. The idea that you can have massive computations but only need to verify a succinct proof changes the game. The infrastructure will need to be modularized. For example, certain things can be computed transparently at the game client end, certain things can use traditional anti-cheat, certain things may require the game clients to generate zero-knowledge proofs, storage may be on things like Filecoin/IPFS, while others may need consensus on specialized execution layers, i.e. application-specific recursive rollups/volitions or L3s, and composability & interoperability mechanisms between them. In the end, the blockchain part - the L3s - may only be used for very specific purposes. State channels on top of L3s also show promise for certain types of multiplayer games. Of course, certain financial games can still use smart-contract chains (L1s or L2s) for high-value assets but I expect these to be a niche - albeit a highly valuable one - going forward.

My general recommendation to whoever is building these is to ask yourself - how would Epic Online Services build this? And then progressively decentralize. Also, don’t bother building the entire gaming stack - just the bits that are unique and valuable to distributed gaming, with the ability to plug in to existing gaming engines and middleware.

Game development studios as DAOs?

Developing games is a creative endeavour. As Francis Ford Coppola put it, “filmmaking is the last dictatorial medium”. Games are no different - it’s very important for the game to have a coherent, creative voice, so bazaar-style development simply does not work. A well-ordered structure with creative directors and department leads is essential to producing a high-quality game. Indeed, some studios like Valve have tried a mild variant of this - and failed. Smaller indie developers work differently, but they also have no need for DAOs. A more promising avenue is using blockchain platforms for fundraising - there’s potential here to do better than Kickstarter.

(As a side-note: crypto people vastly underestimate how difficult making a game is. I have seen roadmaps which claim to make “AAA games in 6 months” or general statements like “hundreds of AAA crypto games in development”. This is simply not how it works. The average AAA game costs $100 million, requires >100 people and 3 years to develop. Oh, did I mention that most game developers dislike crypto games? The talent is simply not available to make so many high-quality AAA games as you may have been led to believe. Sure, indie games can be made for much less, but crypto people have a bad habit of overhyping “AAA crypto games”.)

Nevertheless, there are some types of games that can genuinely be better as DAOs - online game platforms with creative communities. Roblox is an example that’d be better decentralized, with better commissions and rights for creators. Indeed, Briq is building this.

Another possibility is evolving multiplayer games. Once the game is developed, it can be open-sourced. Players who buy the game can propose and vote on changes to the protocol. In reality, though, it’s impossible to satisfy everyone’s needs, and game developers are very well incentivized to do their best. By and large, game developers of these types of games are very pragmatic, and all of them listen closely to community feedback. The ability to fork away can be interesting, though, so I’d definitely like to see this pursued.

Game distribution

Decentralizing game distribution is another legitimate usecase. Unfortunately, it is extremely difficult for a new game distribution service to gain adoption in 2022. Epic Games Store has burned through hundreds of millions (if not billions) of dollars, yet they remain a distant second to Steam. Others like GOG, Itch, Microsoft Store are relegated to small niches. So, really, decentralization is best achieved in the form of private sector players competing fiercely, than a decentralized platform that lacks focus and resources. Still, this could form a niche - I can see it being popular for retro games, or perhaps for trading cards only.

Novel mechanics using smart contracts

So far, basically, the conclusion is blockchain has no place in a vast majority of games, but they can be used to enhance certain aspects, in specific cases.

Things get interesting when you consider novel mechanics that can only be accomplished by smart contracts. Dark Forest is a great early example. Imagine: an artificially intelligent agent that can write and design stories, puzzles, mechanics, worlds etc in a digital world. Using zero-knowledge cryptography, no one knows the solution. Yet, people from around the world can cooperate and coordinate to solve these. Granted, we are Lubins away from something complex like this, but it shows there’s potential for emergent gameplay that’s only possible with blockchain tech.

It is imperative for actually novel games like this to avoid financialization. It can simply be a game that people pay for, to be entertained. What a concept! Some element of tokenization and rewards can help certain types of games, but we must be very, very cautious against making these tradeable.

Concluding

In general, most of “crypto gaming” today and in the near future is nonsense and is justifiably despised by the wider gaming community. Most games do not need blockchain tech - indeed, it will be detrimental to most types of games.

However, we can do better and form niches with novel paradigms. I’d recommend keeping a close eye out for the StarkNet gaming community - this is where real innovation is being attempted.