Hello and welcome to Dark Tunnels, a newsletter dedicated to exploring the emerging ecosystem of fully onchain games. Dark Tunnels is published by Always Scheming, a research consultancy focused on the future of interactive entertainment.

The following essay is Issue #15 of Dark Tunnels. Enjoy!

Regular readers of this newsletter will know that I largely avoid technical deep dives here. Frankly, they tend to be a bit outside my comfort zone as a writer. Where I do try to explore technical topics, however, is in areas that provide us with some indication of how gaming might evolve in the future.

Despite the silly name, I believe Proto-Danksharding is one such development.

Sometimes, I find it helpful to remind myself that this whole fully onchain games thing is a long term bet. If you pull back and project the trends of technological advancement out into the future, it becomes much easier to envision what’s going to be possible in this space.

Part of that optimism for the future stems from the promise that onchain games will eventually be capable of the same type of user experience that has long been table stakes in the games industry: fast, performant, high-fidelity, polished games that require no prior technical understanding to start playing.

This latest EIP du jour represents an enormous step towards leveling the playing field with “traditional” gaming.

Only when fully onchain games can be evaluated by the same criteria as all other games will the true differentiators of web3 technologies — decentralization, composability, permanence, and so on — really shine through.

While Proto-Danksharding doesn’t solve everything, it does meaningfully ramp up the speed and capacity of the Ethereum ecosystem, enabling faster, cheaper, and more seamless gameplay experiences to emerge on L2s.

Now that we’ve properly hyped up EIP-4844, let’s break down what it all means.

To do that, let’s first untangle some of the jargon:

EIP = Ethereum Improvement Proposal.

EIPs are the primary means by which Ethereum changes take place, acting as a source of truth for the community.

Anyone within the Ethereum community can create an EIP, and each EIP receives an identifying number (thus, “EIP-4844”).

Dencun = A portmanteau of the “Deneb” and “Cancun” projects.

“Deneb” and “Cancun” are actually two separate upgrades happening simultaneously as a part of EIP-4844.

“Deneb” impacts Ethereum’s consensus layer, while “Cancun” affects its execution layer.

Sharding = A type of horizontal scaling solution aimed at improving network capacity and speed.

Sharding solutions propose to divide the Ethereum mainnet into a series of smaller, interconnected networks, or “shards.” These shards would each handle a set of nodes that process transactions, with validators assigned at random to verify those transactions and maintain the state.

There have been several sharding proposals throughout Ethereum’s history. Regardless of each proposal’s specific implementation, the intent of sharding has always been to increase the overall speed and efficiency of the Ethereum blockchain via parallel processing.

Rather than relying on all Ethereum nodes to validate every single transaction (and in turn requiring an ever-increasing amount of resources to do so), sharding deduplicates the verification work and opens up node operation to a wider variety of stakeholders.

Crucially, sharding proposals in recent years (including Danksharding, which we’ll return to in a moment) have begun to follow Ethereum’s rollup-centric roadmap, opting to provide more space for data blobs, rather than transactions. These data blobs are used by Layer-2 rollups, which “contract out” their security and decentralization to the Ethereum Layer-1, gaining speed and scale in the process.

To provide some context around the potential impacts of sharding, some of the more optimistic projections have estimated that post-sharding throughput could increase to 100,000 transactions per second or more. This could potentially allow Ethereum to rival the speed of traditional financial rails, all while preserving the added security and decentralization of blockchains.

I’ll leave it to the experts to diagnose how realistic those forecasts are, but the intent is clear: to prepare Ethereum for mass adoption.

Danksharding =  a sharding design that implements a “merged fee market,” wherein a single block proposer chooses all transactions and data for a given slot.

Named after Ethereum researcher Dankrad Feist (what, did you think it was named for something else?), Danksharding is the name for a specific sharding architecture that represents a significant simplification as compared to earlier sharding alternatives.

A key differentiator of Danksharding is the introduction of the merged fee market. Without getting too deep in the weeds on the differences between block proposers and block builders, the important function of slots, the need for proposer/builder separation, and a bunch of other techno-babble, we can simply state for the purposes of this exploration that the merged fee market streamlines block verification, thereby increasing Ethereum’s capacity for handling data. Every block in the chain is verified at faster speeds and lower costs.

It’s also important to note that the full implementation of Danksharding is still several years away. While the potential improvements are clear, the path to achieve them remains long and complex, with many intermediate steps along the way.

In order to achieve full Danksharding, Ethereum must first implement…

Proto-Danksharding = An early step implementing much of the logic and scaffolding required for Danksharding.

Ah, we’ve arrived at last.

Proto-Danksharding implements the majority of the logic and standards required to achieve Danksharding, without actually doing any of the sharding.

The primary feature introduced with Proto-Danksharding is the aforementioned data blob (or, more specifically, data blob-carrying transactions). These blobs carry large amounts of data from L2s, but are much cheaper than similar amounts of calldata.

Unlike normal transaction data, blobs are stored using cryptographic commitments in a separate layer of Ethereum, removing them from direct processing and thereby reducing costs.

By combining this new form of transaction and data storage with the aforementioned merged fee market designed specifically for blob data, gas costs on Layer-2 networks are expected to drop precipitously.

We’re already starting to see evidence of this trickle out on social media:

What this all implies for fully onchain games is an increase in abundance.

Transactions are meaningfully cheaper. Gasless gameplay is achievable. Truly free-to-play experiences are possible for onchain games.

Increased speed and lower costs also allow for greater throughput, meaning even more game transactions can be put onchain. This further expands the design space for onchain game developers. Where concessions were previously forced due to low throughput, builders can now attempt more ambitious designs.

It’s entirely possible that the lower fees and increased performance resulting from EIP-4844 (and subsequent improvements) could lead to a surge in demand, gobbling up the new slack faster than expected. After all, onchain games won’t be the only customers of blockspace.

This could be one of those so-called “good problems,” as it likely means that web3 is reaching larger audiences. However, it may be equally likely that the demand comes from an increase in bot transactions, which is probably less desirable.

Think of it like when the government adds a new lane to a crowded highway: congestion might subside in the short term, but over the long haul people just end up driving more and everyone still ends up stuck in traffic.

Whether that new traffic is real humans or autonomous vehicles remains to be seen, but I feel confident in saying that the additional blockspace will not go to waste. Enterprising developers will take advantage of this newfound abundance, and onchain gaming (as I’ve written about previously) should be at the forefront of that emerging wave.

1. If you’re interested in reading more about the specifics behind Proto-Danksharding, I’ve compiled a few of the resources I used to write today’s newsletter:

   1. EIP-4844 official website

   2. CoinDesk

   3. Investopedia

The blobs are here, folks.

Ethereum is scaling. Transaction costs are dropping, speeds are increasing, and builders are eager to take advantage.

Day by day, the possibilities of what can be built onchain become increasingly comparable to what can already be built offchain.

The real question for fully onchain developers is: when the playing field is truly leveled, how will your game leverage the unique capabilities of web3 to stand out from the pack?

Thanks, as always, for reading. We’ll be back in a couple weeks.

Until then, come find me at GDC!

Hello and welcome to Dark Tunnels, a newsletter dedicated to exploring the emerging ecosystem of fully onchain games. Dark Tunnels is published by Always Scheming, a research consultancy focused on the future of interactive entertainment.

The following essay is Issue #15 of Dark Tunnels. Enjoy!

Regular readers of this newsletter will know that I largely avoid technical deep dives here. Frankly, they tend to be a bit outside my comfort zone as a writer. Where I do try to explore technical topics, however, is in areas that provide us with some indication of how gaming might evolve in the future.

Despite the silly name, I believe Proto-Danksharding is one such development.

Sometimes, I find it helpful to remind myself that this whole fully onchain games thing is a long term bet. If you pull back and project the trends of technological advancement out into the future, it becomes much easier to envision what’s going to be possible in this space.

Part of that optimism for the future stems from the promise that onchain games will eventually be capable of the same type of user experience that has long been table stakes in the games industry: fast, performant, high-fidelity, polished games that require no prior technical understanding to start playing.

This latest EIP du jour represents an enormous step towards leveling the playing field with “traditional” gaming.

Only when fully onchain games can be evaluated by the same criteria as all other games will the true differentiators of web3 technologies — decentralization, composability, permanence, and so on — really shine through.

While Proto-Danksharding doesn’t solve everything, it does meaningfully ramp up the speed and capacity of the Ethereum ecosystem, enabling faster, cheaper, and more seamless gameplay experiences to emerge on L2s.

Now that we’ve properly hyped up EIP-4844, let’s break down what it all means.

To do that, let’s first untangle some of the jargon:

EIP = Ethereum Improvement Proposal.

EIPs are the primary means by which Ethereum changes take place, acting as a source of truth for the community.

Anyone within the Ethereum community can create an EIP, and each EIP receives an identifying number (thus, “EIP-4844”).

Dencun = A portmanteau of the “Deneb” and “Cancun” projects.

“Deneb” and “Cancun” are actually two separate upgrades happening simultaneously as a part of EIP-4844.

“Deneb” impacts Ethereum’s consensus layer, while “Cancun” affects its execution layer.

Sharding = A type of horizontal scaling solution aimed at improving network capacity and speed.

Sharding solutions propose to divide the Ethereum mainnet into a series of smaller, interconnected networks, or “shards.” These shards would each handle a set of nodes that process transactions, with validators assigned at random to verify those transactions and maintain the state.

There have been several sharding proposals throughout Ethereum’s history. Regardless of each proposal’s specific implementation, the intent of sharding has always been to increase the overall speed and efficiency of the Ethereum blockchain via parallel processing.

Rather than relying on all Ethereum nodes to validate every single transaction (and in turn requiring an ever-increasing amount of resources to do so), sharding deduplicates the verification work and opens up node operation to a wider variety of stakeholders.

Crucially, sharding proposals in recent years (including Danksharding, which we’ll return to in a moment) have begun to follow Ethereum’s rollup-centric roadmap, opting to provide more space for data blobs, rather than transactions. These data blobs are used by Layer-2 rollups, which “contract out” their security and decentralization to the Ethereum Layer-1, gaining speed and scale in the process.

To provide some context around the potential impacts of sharding, some of the more optimistic projections have estimated that post-sharding throughput could increase to 100,000 transactions per second or more. This could potentially allow Ethereum to rival the speed of traditional financial rails, all while preserving the added security and decentralization of blockchains.

I’ll leave it to the experts to diagnose how realistic those forecasts are, but the intent is clear: to prepare Ethereum for mass adoption.

Danksharding =  a sharding design that implements a “merged fee market,” wherein a single block proposer chooses all transactions and data for a given slot.

Named after Ethereum researcher Dankrad Feist (what, did you think it was named for something else?), Danksharding is the name for a specific sharding architecture that represents a significant simplification as compared to earlier sharding alternatives.

A key differentiator of Danksharding is the introduction of the merged fee market. Without getting too deep in the weeds on the differences between block proposers and block builders, the important function of slots, the need for proposer/builder separation, and a bunch of other techno-babble, we can simply state for the purposes of this exploration that the merged fee market streamlines block verification, thereby increasing Ethereum’s capacity for handling data. Every block in the chain is verified at faster speeds and lower costs.

It’s also important to note that the full implementation of Danksharding is still several years away. While the potential improvements are clear, the path to achieve them remains long and complex, with many intermediate steps along the way.

In order to achieve full Danksharding, Ethereum must first implement…

Proto-Danksharding = An early step implementing much of the logic and scaffolding required for Danksharding.

Ah, we’ve arrived at last.

Proto-Danksharding implements the majority of the logic and standards required to achieve Danksharding, without actually doing any of the sharding.

The primary feature introduced with Proto-Danksharding is the aforementioned data blob (or, more specifically, data blob-carrying transactions). These blobs carry large amounts of data from L2s, but are much cheaper than similar amounts of calldata.

Unlike normal transaction data, blobs are stored using cryptographic commitments in a separate layer of Ethereum, removing them from direct processing and thereby reducing costs.

By combining this new form of transaction and data storage with the aforementioned merged fee market designed specifically for blob data, gas costs on Layer-2 networks are expected to drop precipitously.

We’re already starting to see evidence of this trickle out on social media:

What this all implies for fully onchain games is an increase in abundance.

Transactions are meaningfully cheaper. Gasless gameplay is achievable. Truly free-to-play experiences are possible for onchain games.

Increased speed and lower costs also allow for greater throughput, meaning even more game transactions can be put onchain. This further expands the design space for onchain game developers. Where concessions were previously forced due to low throughput, builders can now attempt more ambitious designs.

It’s entirely possible that the lower fees and increased performance resulting from EIP-4844 (and subsequent improvements) could lead to a surge in demand, gobbling up the new slack faster than expected. After all, onchain games won’t be the only customers of blockspace.

This could be one of those so-called “good problems,” as it likely means that web3 is reaching larger audiences. However, it may be equally likely that the demand comes from an increase in bot transactions, which is probably less desirable.

Think of it like when the government adds a new lane to a crowded highway: congestion might subside in the short term, but over the long haul people just end up driving more and everyone still ends up stuck in traffic.

Whether that new traffic is real humans or autonomous vehicles remains to be seen, but I feel confident in saying that the additional blockspace will not go to waste. Enterprising developers will take advantage of this newfound abundance, and onchain gaming (as I’ve written about previously) should be at the forefront of that emerging wave.

1. If you’re interested in reading more about the specifics behind Proto-Danksharding, I’ve compiled a few of the resources I used to write today’s newsletter:

   1. EIP-4844 official website

   2. CoinDesk

   3. Investopedia

The blobs are here, folks.

Ethereum is scaling. Transaction costs are dropping, speeds are increasing, and builders are eager to take advantage.

Day by day, the possibilities of what can be built onchain become increasingly comparable to what can already be built offchain.

The real question for fully onchain developers is: when the playing field is truly leveled, how will your game leverage the unique capabilities of web3 to stand out from the pack?

Thanks, as always, for reading. We’ll be back in a couple weeks.

Until then, come find me at GDC!