While transaction fees at Ethereum are skyrocketing, the scene is arguing about the scaling of the blockchain. The old drama we've had with Bitcoin for so long is repeating itself. Dapps are therefore taking refuge in other blockchains or in rollups.
Have you tried an Ethereum transaction today? Hopefully not. Because the fees there are currently so expensive that it can spoil the whole day.
Expensive transactions weigh on the mood:
The gas prices currently amount to 155-230 Gwei, and if you do not understand what that means, you are not alone. The Metamask wallet currently charges these fees for the following actions:
- an ETH transaction: 8 dollars
- a token transaction: $ 16
- the exchange of tokens through UniSwap : 42 dollars
- I could not test the provision of liquidity on UniSwap, the permission to use coins alone cost 15 dollars. I wouldn't be surprised if the fees were 3 digits here
This is of course nice for the miners: yesterday they received around 18,350 ethers in transaction fees, which corresponds to a good 27 million euros and is a lot more than the 12,500 ethers that the block reward earned. Ironically, Ethereum finances itself more sustainably the worse the network works for the user.
Because for them all of this is rather frustrating. Pretty frustrating. At any given time, around 100,000 transactions are waiting for confirmation, and those who have not paid enough fees will have to wait a long time. The network is full and that spoils the day for many.
The core developers of Ethereum are making the same lame excuses for not allowing Ethereum to scale onchain. Like Bitcoin Core 5 years earlier ". Back then, Nikita campaigned in vain for onchain scaling in the “wars of scaling” over Bitcoin. In a way, Ethereum was a ray of hope for him that things could be different.
So now the thing is repeating itself. We see the same strange, inconclusive exchange of arguments.
Core developers don't want to scale:
The core developers Nikita attacks here are primarily Marius van der Wijden and Péter Szilágy. Van der Wijden has previously explained why he does not want to increase the gas limit, even though this would allow more transactions and thus lower the fee pressure.
Basically, he thinks raising the limit is dangerous and you have to consider several factors. We summarize them briefly, what - attention! - something gets technical:
- The uncle rate measures the "orphaned" blocks, i.e. the blocks that were produced by miners but did not make it into the blockchain because they were overtaken by a block from another miner. If the uncle rate increases, it indicates difficulties in distributing new blocks.
- The most important factor for Marius is the size of the state . The state is the "state" of the blockchain - all currently valid entries in it, i.e. all credits and all states of smart contracts. This is the biggest limiting factor. The nodes must keep the current state. You can delete the old one, but it is difficult, which is why a node accumulates unnecessary data over time - Marius speaks of 500 gigabytes.
- Synchronization: It currently takes several days to synchronize a full node. If you increased the gas limit, this time would continue to increase.
- The execution time : when there is a new block, all nodes must execute all transactions in it and update the state. The higher the gas limit, the longer it will take the nodes to execute blocks; once they take more than 13 seconds - the interval between the blocks - the network will collapse.
- Archive nodes : These nodes, which store all past states, are useful for block explorers, for example. A higher gas limit increases the pressure on these nodes significantly.
Marius lists these bottlenecks, but does not give any specific values as to the extent to which they are already narrowing. Instead, he presents a list of solutions to widen these bottlenecks. He hopes that Layer2 solutions will sprout the situation and concludes with his prioritization: “Personally, I prefer a congested network to a broken network, as users are no longer able to operate their own nodes to verify the network . "
Yes, for those who know the scaling debate about Bitcoin, these tweets should evoke memories: The core developers demand that every user can verify the blockchain; they present not just one, but a large number of bottlenecks, but it is vague how tight they are these are, but present a lot of solutions that they have to implement before we can talk further about scaling.
No technology - ideology:
Nikita, who as the operator of a block explorer of course knows how exhausting an Ethereum node is , reacts with anger. 500 gigabytes of storage? That certainly costs millions of dollars, he says, and of course it's better for users to pay $ 1,000 for fees than for node operators to wait a few days for the node to sync. In none of the factors listed did his knots even come close to a load limit. Even weak hardware can handle 20-30 times as many operations. And so on.
We know this discussion. You should also know Nikita well enough to know that no technical arguments will help him. The technology here is just the facade of a debate that is more about ideology. Perhaps it is because core developers work for idealistic motives, Nikita, on the other hand, from more commercial ones, and that core developers do not run a company that finances strong servers like Nikita's, but rather as private individuals the node on a (good) PC let run. But that's speculation.
In the end, it is the core developers who decide, and their head, Péter Szilágy, does not seem willing to compromise: He accuses Nikita of either not understanding the consequences of increasing the limit or of intentionally wanting to damage the network. He is against increasing the limit "because all harmful effects fall back on the same four people who have been running the network for six years" - by which he means his three co-developers to us.
Escape to Layer-2:
So Ethereum won't scale any further, and ETH2 will take a while to be operational. So what options do users and dapps have left?
One option is to switch to a "Layer 2 solution". With the rollups, as Christoph Jentzsch explains here , Ethereum has a range of well-functioning, albeit somewhat centralized, methods. The decentralized exchange Loopring, for example, uses zkRollups to enable largely free, private and real-time decentralized trading. This offer is increasingly used: Loopring reached around February 8th in trading volume of nearly 30 million dollars; other L2-DEX are also following suit.
Of course, all of this is still very little in comparison with UniSwap, which has turned over more than a billion dollars in the last few days. But it's a promising start, and it seems like the Ethereum community has agreed on “their” Layer 2 solution.
But there is no less demand for other alternatives.
Migration to another blockchain:
A Layer-2-Dex is complex to set up, centralizes the operation on the operator of the rollup and affects the interoperability. Some dapps therefore decide to leave the Ethereum blockchain or at least to open up to alternatives. Some examples:
- Equilibrium is a “cross-chain version” of the widely used Dapp Curve.Finance, which will also run in parallel on Polkadot.
- SingularityNET, a Dapp for creating and sharing AIs, begins the migration from Ethereum to Cardano
- PancakeSwap on the Binance blockchain is seeing skyrocketing trading volumes
- DEX Pangolin starts on the Avalanche Blockchain (AVA). Allegedly, $ 6.5 million liquidity has already migrated from ETH to AVA.
With Polkadot, Cardano, Binance and now also Avalanche, four blockchains are ready to welcome users and dapps with open arms when they are frustrated by Ethereum looking for a new home.
Overall, the volume of the escape seems to be very manageable. The DeFi ranking at Defillama still shows an absolute dominance of Ethereum. Presumably, all other blockchains simply lack network effects. Binance can try to build this up through its own large exchange, Polkadot got off to a strong start , Cardano seems to be of little relevance after the long period of development without results, and Avalanche is considered technically mature, but in terms of personnel it is a rather cerebral birth bitcoin cash scene is somewhat isolated. And we don't even want to talk about Julian Hosp's DeFiChain.
If four platforms compete to take the users from a still overpowering blockchain, this is usually a guarantee that the overpowering blockchain will remain overpowering. If there was a competitor that the market had already agreed on, it would look different. But no alternative should generate enough network effects to become a serious threat to Ethereum.
After all, Polkadot has good prospects of positioning itself parallel to Ethereum by primarily relying on coexistence and bridges instead of competition. But even this blockchain still has to achieve a lot of network effects before it becomes relevant for DeFi.