This Ethereum Glamsterdam state gas guide explains the end of the flat 21,000 gas rule for basic ETH transfers. Sending to a brand-new address will include a one-time state storage charge, so wallets, explorers, and fee tools must adjust. Test these changes on upcoming testnets to avoid misquotes, rejections, or stuck payments. Ethereum is changing a rule that many people consider a constant: every simple ETH transfer costs 21,000 gas. With the Glamsterdam upgrade, that will no longer be true in every case. Sending ETH to an address that has never appeared on-chain will cost more than sending to an address that already exists. The network will add a new “state gas” fee because it must create and store a new account record. If your product or script assumes “21,000 gas always,” you need to update it now.

Why the 21,000 gas rule is ending

The old model in one sentence

For years, wallets and services treated a basic ETH transfer as 21,000 gas, full stop. It did not matter if the recipient was a well-used wallet or a fresh address. The network charged the same.

The new model: first-time addresses cost more

Glamsterdam changes this. A transfer to an existing account still uses 21,000 execution gas. A transfer to an address that has never appeared on Ethereum’s state will include an extra fee for creating and storing that account. The proposal sets this extra “state gas” at 183,600 units. The goal is simple: storage is not free forever, and the price should reflect the work the network must do to add a new permanent record.

Ethereum Glamsterdam state gas guide

What is state gas?

State gas is a new meter that charges for permanent state growth. Creating a brand-new account grows blockchain state. That is different from updating balances for accounts that already exist. With Glamsterdam: – Sending to an existing address: 21,000 execution gas. – Sending to a never-seen address: 21,000 execution gas plus 183,600 units of state gas. You still pay in ETH. Wallets should show users the total cost and, if possible, label the part that comes from state gas. This helps explain why two “simple sends” may have different fees.

How wallets should estimate

Update fee estimation so 21,000 is the floor, not the fixed amount. Before building the transaction: – Check if the recipient exists on-chain. – If it exists, estimate 21,000 execution gas. – If it does not exist, include the 183,600 state gas in your fee logic. How to check existence: – Query balance (eth_getBalance), code (eth_getCode), and nonce (eth_getTransactionCount). If all are zero or empty, the account likely does not exist yet. – Cache recent lookups to reduce RPC calls, but always handle race conditions. Another transaction could “pre-warm” the address between your check and your send. UI tips: – Warn users when they send to a brand-new address for the first time. – Show a clear fee breakdown or a short note like “Includes one-time state fee for new address.” – Do not reject a send that asks for more than 21,000 gas. That behavior will break valid payments.

How explorers and mempool tools should adapt

– Stop assuming 21,000 gas means “simple ETH transfer” in all cases. – Update fee charts and labels to include state gas. A “21,000 transfer” and a “first-touch transfer” will now look different. – Adjust spam filters and sorters that used 21,000 as a hard cap. Otherwise, you may hide or misclassify valid transactions.

Who is affected and real-world examples

Wallets and fee estimators

– Fixed 21,000 estimates will underquote fees for first-time recipients. – Transactions could be stuck or dropped if the attached fee is too low. – Auto-retry logic might loop, wasting time and confusing users.

Exchanges and payment processors

– Bulk withdrawals often go to fresh customer addresses. Expect more first-touch sends. – Per-withdrawal costs may rise when many recipients are new. – Payout queues and batch logic need to detect new addresses and price correctly.

Dapp developers and bots

– Airdrops to untouched addresses will cost more per address. – Referral rewards or faucet payouts to new users will see the state gas charge. – Trading, arbitrage, and sweeping bots must handle variable fees for “simple sends.” Everyday users will notice this when they send ETH to a brand-new cold wallet for the first time. The first deposit costs more. Later transfers to the same wallet will look normal again.

Action checklist before mainnet

For wallet teams

• Replace hardcoded 21,000 with dynamic detection that adds state gas for new addresses.
• Run a preflight check: balance, code, and nonce for the recipient.
• Update fee UIs to explain first-touch costs in plain language.
• Add robust error handling if state changes between estimation and submission.
• Test on upcoming testnets with both new and existing addresses.
• Log metrics: percentage of sends that include state gas, average extra cost, failure rates.

For infrastructure providers and explorers

• Expose state gas in APIs, estimators, and dashboards.
• Update classification logic for “simple transfers.”
• Re-tune spam and anomaly filters that assume a hard 21,000 cap.
• Document the difference between execution gas and state gas for your users.
• Monitor client versions and network readiness across testnets.

For power users and businesses

• Expect higher costs when paying first-time recipients at scale (payrolls, refunds, grants).
• Consider “pre-warming” addresses with a tiny deposit to pay the one-time state fee in advance.
• Update budgets and fee ceilings in scripts and bots.
• If you use allowlists or new-address onboarding, bake the first-touch fee into your flows.
• Educate support teams so they can explain why two simple sends have different fees.

Testing timeline and networks

Glamsterdam will activate on a practice network called Platåberget first. That testnet uses worthless tokens so teams can experiment without risk. After that, the upgrade will roll out to Sepolia and Hoodi, and then to mainnet. Developers should: – Spin up nodes on Platåberget as soon as activation happens. – Run end-to-end tests for new-address sends. – Try stress cases: batched payouts, airdrops, and high-volume retail flows. – Report edge cases and update tooling before Sepolia and Hoodi. Nobody sending day-to-day ETH needs to change habits right now. The early warnings are for teams who build the software that users depend on.

Risk management and communication

If you do nothing, you risk: – Underquoting fees for first-touch recipients and causing stuck transactions. – Rejecting valid transactions because your code treats 21,000 as a ceiling. – Mislabeling transactions in your explorer or analytics. To reduce risk:

• Default to a safe “extra buffer” when you cannot determine address status. Make that buffer visible and explain why it exists.
• Implement a clear retry strategy that reads the latest state, adjusts fees, and resubmits with a single replacement transaction.
• Alert on unusual failure rates and fee outliers after Glamsterdam hits testnets.
• Share a short user-facing note or banner about “first-time address” fees before mainnet day.

What users should do now

– Keep using your wallet as normal. Most sends to known addresses still look like 21,000 gas transfers. – When you move funds to a brand-new address, expect a higher fee one time. After that, transfers to the same address will be normal again. – If your wallet rejects a simple send or estimates a strange fee, update to the latest version or contact support. The app may need a fix to handle state gas. As testnets switch on, watch for announcements from the Ethereum Foundation, client teams, and your wallet provider. The process is staged so everyone can find and fix issues before mainnet activation. The goal of this change is fair pricing for permanent storage. It should also encourage better state hygiene. By planning now, your product will handle first-touch transfers smoothly on day one. Ethereum is still Ethereum. You still pay gas in ETH. The main difference is that the network will charge extra when it must create a lasting record for a new account. That is work, and the fee reflects it. If you build software that moves ETH, this is the time to test, measure, and communicate. In short, treat 21,000 gas as the floor, not the rule. Add new-address detection. Include the state gas charge when needed. Test on Platåberget, then Sepolia and Hoodi. If you follow the steps in this Ethereum Glamsterdam state gas guide, your users will see clear fees, fewer errors, and smooth first-time transfers.

(Source: https://www.coindesk.com/tech/2026/08/18/ethereum-s-next-upgrade-breaks-the-21-000-gas-rule-wallets-rely-on)

For more news: Click Here

FAQ