Nov 14 2021
Greetings from your engineering team. The last few weeks and months have been very busy, as we are now supporting two complementary L2 solutions. Boba is our solution for L2 applications that need EVM compatibility. Plasma is our transfer-focused L2, which trades off EVM compatibility for extremely cost effective token transfer (e.g. with as many as 64,000 token transfers per batch). We are now the only team that covers two main L2 use cases.
The main challenges for us, and indeed the entire Ethereum ecosystem, are the variable (and high) values for gas and the increasingly high cost of ETH. For example, on Nov. 10, gas briefly spiked to 405 Gwei and ETH spiked to $4800. Compared to when OMG was launched in 2017, the average gas has increased from 16 to 136 Gwei and ETH has increased from $320 to $4600. That's a 122x overall increase on a USD basis (136/16 * 4600/320 = 122). Gas in Nov. 2021, compounded by the price of ETH, present great difficulties for everyone. Since we are trust-rooted in Ethereum, every single transaction we do requires us and our users to pay ETH directly or indirectly to miners on L1. However, the design of Boba, an Optimistic Rollup, allows significant cost savings compared to Ethereum. For example, a token swap on Boba is generally 13x less expensive than on Ethereum, turning a $150 fee into a $12 fee. Moreover, with increasing transaction volumes and users on Boba, the costs will come down further, since everyone shares the gas savings from processing more transactions per batch.
Aside from all the positive feedback, for which we are grateful, users raised three main issues, namely, (1) the long waiting times for cross-chain bridging, (2) the cost of classic deposits, and (3) intermittant reverts seen when fast-bridging.
2.1 Increasing speed of Fast Exits. Classic exits take 7 days, and, previously, fast exits took several minutes to several hours depending on L1 gas. We initially used an algorithm that waited, sometimes for 3 or 4 hours, for L1 gas to dip before relaying messages. However, you have told us that you want to be able to make that choice. Starting now, the gas for the L2->L1 bridging will float with current L1 gas. This means that if you need a bridge to execute quickly, regardless of gas, you can now do that. Or, if you have more time, you can decide to wait for L1 gas to dip before starting a bridge, saving you money. So now, you can choose the best approach for your needs. Related to these changes, as of now, essentially all bridging operations should complete in 55 minutes or less, regardless of L1 gas.
2.2 Reducing cost of Classical Bridge to L2. Code improvements have allowed us to reduce the average cost of classic L1->L2 bridging by about 30%.
2.3 Avoiding Fast Bridge reverts. Occasionally, the fast bridges reverted. The difficulty was that pending transactions ahead of you in the queue could deplete the bridge pools while you were waiting. This then led to a revert and an annoying delay for your funds to be returned to you from the bridge contracts. As of now, the gateway considers the current and predicated future balance of the liquidity pools before allowing you to initiate a fast-bridging operation.
We would like to draw your attention to the availability of the community fraud detection, as described here. As noted previously, the overall security of the system depends on independent verification of the state roots we submit. The more users inspect our operations, root-by-root, the safer the overall system is. At present, the only incentive for you to run a fraud detector is to help secure your funds. However, additional mechanisms to incentive fraud-detection will soon be announced.
Teams wishing to manage their treasuries on Boba, and large DeFi and NFT projects, need a multisig solution. Gnosis Multisig will be available within the next 10 days on Boba.