Tool AT-W34REV. 2026-A

Gas calculator: what does a transaction cost on Ethereum and layer 2?

A transaction costs gas × (base fee + tip). A token transfer needs about 41,300 gas, a swap on an exchange about 133,500. On Arbitrum, OP Mainnet and Base there is an additional fee for the data that ends up on Ethereum; Polygon PoS is calculated in POL. The gas calculator fetches the current fees, works out each action for all five chains in euros and shows the calculation.

  • free
  • no wallet needed
  • live or with your own values

Choose an action, read the cost per chain

Calculator · transaction costs compared

Guide values for planning, not investment advice and not a price forecast. The rate is used only for conversion into euros. The tool does not ask for keys and does not trigger any transaction.

Step 1What do you want to do on the chain?

Step 2How urgent is it?

Calculated with the figures as of 14 November 2026 at 12:55.

Gas usage

—

Choose an action.

Gas amounts checked on . How we plan fees for your own application is described under Web3 development; the easiest way to send us the copied result is via the contact form.

View or change fees and rates

Gwei are billionths of an ETH or POL. Base fee and tip apply per unit of gas; the tip follows the speed you choose.

Ethereum ETH
Arbitrum One ETH
OP Mainnet ETH
Base ETH
Polygon PoS POL
Rates for conversion

The scalars for OP Mainnet and Base are in the chain's gas oracle; they rarely change.

Calculates in your browser. For the live values the page only contacts aureum-tech.com; your inputs are not sent.

How the fee is made up

Since the London upgrade in 2021, every transaction on Ethereum has two price components per unit of gas. The network sets the base fee for each block and burns it. The tip (known as the priority fee) goes to whoever builds the block. The calculation is therefore simple: gas × (base fee + tip). Bidding more than that gains you nothing, except in sudden demand spikes, when it turns back into an auction for the tip Roughgarden 2024.

The base fee follows utilisation. If a block is more than half full, the base fee rises by up to 12.5% in the next one; otherwise it falls. Fees therefore arise from scarcity: those who do not want to wait pay for priority, and without a bottleneck they tend towards zero Huberman et al. 2021. On 14 November 2026 at 12:55 the base fee on Ethereum was 0.116 gwei, so a simple transfer cost €0.005. Even before EIP-1559, fees depended mainly on the number of waiting transactions and hardly at all on the exchange rate Pierro and Rocha 2019. The calculator therefore keeps the gas price in gwei separate from the conversion into euros: if the rate rises, the same transaction becomes more expensive in euros, even when the network is empty.

EIP-1559 shortened waiting times and evened out prices within a block, but hardly lowered fees Liu et al. 2022. In the short term, blocks swing widely between full and half empty, and when demand surges the base fee follows only slowly Reijsbergen et al. 2021. This comes down to the step size: if it stays small enough, the base fee settles; larger steps make it unstable Leonardos et al. 2021. For you this means: the base fee of the next block is a good estimate, but not a fixed price.

Fee cap and speed

A wallet shows two numbers: the cap per unit of gas (maxFeePerGas) and the tip. The cap is a reserve. You pay the base fee of the block in which the transaction lands, plus the tip. Because the base fee rises by at most 12.5% per block, it is possible to work out how long a cap will last. The calculator gives the base fee × 2.03 plus the tip, which covers six full blocks in a row, a good minute.

The speed sets the tip as a percentile of the last 20 blocks: the 10th, the 50th or the 90th. A suitable fee can thus be seen as the price for a desired probability of being included within a given time Laurent et al. 2022. These values come from the blocks themselves. What flows to the block builders through private channels is only partly visible in public, and those who use such channels sometimes pay more with no guarantee Messias et al. 2023.

Base fee after consecutive full blocks
full blocksapprox.Factor
112 s× 1,13
224 s× 1,27
448 s× 1,6
672 s× 2,03
10120 s× 3,25

One block every 12 seconds, at most 1/8 more per block (EIP-1559). Status of the rules: Ethereum after Fusaka (03.12.2025) and BPO2 (07.01.2026).

How much gas an action needs

For a simple transfer the protocol sets 21,000 gas. Everything else depends on the contract that the transaction calls. We have therefore counted real transactions. Sample from 14.11.2026: about 30,000 successful transactions from 250 Ethereum blocks, plus Base and Arbitrum One for the data size. We counted only what could be clearly assigned to an action by function and events. The table gives the median and the usual range from the 10th to the 90th percentile.

Gas usage on Ethereum, sample from 14.11.2026
ActionGas (median)usual rangeCount
Send ETH or POL21.000fixed11.517
Send token (ERC-20)41.30035,100 to 46,1002.712
Send token to new address (ERC-20)63.20057,500 to 63,2001.605
Grant approval (approve)46.60029,100 to 55,900681
Swap on an exchange133.500102,200 to 370,0002.432
Transfer NFT (ERC-721)68.10064,900 to 86,70039
Mint NFT190.700112,600 to 244,00017

Only successful transactions that could be clearly assigned by function and events were counted. NFT transfer and mint: small sample.

Two patterns stand out. Token transfers fall into two groups: to recipients who already hold the token, at about 41,300 gas, and to new recipients, at about 63,200. The difference is storage: a new value costs 20,000 gas, changing an existing one 2,900. And a swap varies widely, because every exchange and every route calculates differently. In a swap, though, gas often is not what matters. Bots bid for the order of transactions in order to front-run swaps Daian et al. 2020, and the price disadvantage from slippage can far exceed the gas. The wallet should therefore have a tight slippage limit.

How much gas a contract needs is decided when it is programmed. As early as 2017, 80 to 93.5% of 4,240 contracts examined contained costly patterns that the compiler did not optimise away Chen et al. 2017. That is why the table gives ranges, and why, for your own application, it is worth looking at the contract before it is deployed. What an approval allows exactly is shown by the calldata decoder.

Deploying a contract

A deploy costs 21,000 gas base fee, 32,000 for creation, the calldata for the initcode, 2 gas per 32 bytes of initcode and 200 gas per byte of stored code. On top comes what the constructor computes and stores. The calculator assumes that the initcode is 500 bytes larger than the code and consists of 10% zero bytes, and it allows 110,000 gas for the constructor, the median of our sample. For 8,000 bytes this gives:

21,000 + 32,000 creation + 125,800 calldata (8,500 bytes of initcode) + 532 initcode words + 1,600,000 code (8,000 bytes × 200) + 110,000 constructor = 1,889,332 gas

Ethereum does not accept more than 24,576 bytes of code per contract, and since Fusaka a transaction may use at most 16,777,216 gas. Larger applications are split across several contracts.

Writing data to the chain: the floor under EIP-7623

Some businesses only want to anchor a proof, such as the SHA-256 checksum of a contract or certificate. A transaction with 32 bytes of data to an address without a contract is enough for that. Since Pectra (May 2025), a floor applies to such pure data: each byte counts as a token, a non-zero byte as four, and at least 10 gas per token is due instead of 4. For 32 bytes that is 22,280 gas, €0.0054 on Ethereum and €0.00023 on Base.

Whether a blockchain is the right tool at all, or a signed database is enough, is something the blockchain suitability check clarifies. The document itself never belongs on the chain, only its checksum.

Pure data, no zero bytes
BytebeforetodayEthereum
3221.51222.2800,0054 €
25625.09631.2400,0075 €
1.02437.38461.9600,01 €
10.000181.000421.0000,10 €

“Before” = 16 gas per byte (EIP-2028), “today” = floor under EIP-7623. Euros as of 14 November 2026 at 12:55.

Arbitrum, OP Mainnet, Base and Polygon: how fees are calculated there

Arbitrum One, OP Mainnet and Base are rollups. They execute transactions themselves and post the data to Ethereum in bundles so that anyone can verify the result Thibault et al. 2022. Their fee therefore has two parts: execution, calculated as on Ethereum but with its own, much smaller base fee, and a share for the space on Ethereum. Polygon PoS is built differently: its own chain with its own validators, which only writes checkpoints to Ethereum. There is no data share there, but you pay in POL instead of ETH.

Fee models, as of 14 November 2026
ChainTypeExecutionShare for Ethereum
EthereumLayer 1Gas × (base fee + tip), in ETHnot applicable, calldata is included in the gas
Arbitrum OneOptimistic rollup (Nitro)Gas × (base fee + tip), in ETH; base fee at least 0.02 gweiBrotli size of the transaction × price per byte (ArbGasInfo)
OP Mainnet, BaseOptimistic rollup (OP Stack)Gas × (base fee + tip), in ETH; plus an operator share, currently 0estimated size (at least 100 bytes) × (16 × scalar × L1 base fee + blob scalar × blob fee) ÷ 10⁶
Polygon PoSown chain with checkpoints on EthereumGas × (base fee + tip), in POLnot applicable

Sources: Arbitrum documentation and Nitro source code, OP Stack specification Fjord and Jovian, Polygon documentation. We measured the data size per action on Base and Arbitrum One.

Since EIP-4844 in March 2024, rollup data goes into blobs, a separate storage area with its own base fee. Rollups have used Ethereum more heavily since then, and their transactions have become cheaper. Because operators bundle data to save money, some of it only reaches an Ethereum block later Park et al. 2026. Confirmation on the layer 2 is instant; the security from Ethereum follows afterwards. A blob costs the same per block however full it is; rollups with little traffic therefore often do better with ordinary calldata Crapis et al. 2024. That is why Arbitrum and the OP Stack do not work out data costs directly from the blob price, but through a price per byte that the operator keeps up to date.

Cheap does not mean stable. In the first months of the blob market, badly packed blocks at times lost up to 70% of fees, and in busy periods blob fees jumped sharply Heimbach and Milionis 2026. Since Fusaka there is a reserve price: if a blob costs less than 8,192 gas at the base fee, or one sixteenth of the base fee per byte, the blob fee stops falling and rises again with every blob used (EIP-7918). On 14 November 2026 at 12:55 it stood at 0.00726 gwei, close to this reserve price (0.00725 gwei).

One byte on Ethereum, as of 14.11.2026
Storage locationPrice per byte
Calldata (16 gas × base fee)1.86 gwei
Blob (1 blob gas per byte)0.00726 gwei
One full blob, 131,072 bytes0,0016 €

Calldata was therefore about 256 times as expensive as blob space. Currently up to 21 blobs fit in each block, with a target of 14 (BPO2).

0,000001 €0,0001 €1 cent1 €Ethereum0,0099 €0,45 €Arbitrum One0,0014 €0,07 €OP Mainnet0,0000029 €0,00012 €Base0,00043 €0,02 €Polygon PoS0,001 €0,05 €

Fig. 1 · Upper bars: token transfer. Lower bars, hatched: contract of 8,000 bytes. Logarithmic scale, each division is a hundredfold increase. As of 14 November 2026 at 12:55, normal speed.

The figure shows how far apart the chains are: between Ethereum and OP Mainnet there was a factor of more than 1,000 on the reference date. On OP Mainnet, execution was so cheap that the share for Ethereum made up almost the whole fee; on Base and Arbitrum it was the other way round. For an application with many small transactions, the choice of chain therefore matters more than any detailed optimisation. Which chain suits your project also depends on users, exchanges and security, not just on the fee. More tools around blockchain are in the overview of Web3 tools.

All actions on all chains

The table works out every action with the figures as of 14 November 2026 at 12:55 and at normal speed. It is a snapshot: for current values, use the calculator above.

Cost in euros, as of 14.11.2026
ActionGasEthereumArbitrum OneOP MainnetBasePolygon PoS
Send ETH or POL21.0000,005 €0,00074 €0,0000029 €0,00022 €0,00052 €
Send token (ERC-20)41.3000,0099 €0,0014 €0,0000029 €0,00043 €0,001 €
Send token to new address (ERC-20)63.2000,02 €0,0022 €0,0000029 €0,00066 €0,0016 €
Grant approval (approve)46.6000,01 €0,0016 €0,0000029 €0,00049 €0,0012 €
Swap on an exchange133.5000,03 €0,0046 €0,0000043 €0,0014 €0,0033 €
Transfer NFT (ERC-721)68.1000,02 €0,0024 €0,0000029 €0,00071 €0,0017 €
Mint NFT190.7000,05 €0,0066 €0,0000048 €0,002 €0,0047 €
Anchor data (e.g. a checksum)22.2800,0054 €0,00078 €0,0000029 €0,00023 €0,00055 €
Deploy contract (8,000 bytes)1.889.3320,45 €0,07 €0,00012 €0,02 €0,05 €

Gas prices of the snapshot: Ethereum 0.139 gwei, Arbitrum One 0.02 gwei, OP Mainnet 0.000000943 gwei, Base 0.006 gwei, Polygon PoS 328 gwei. Rate 1 ETH = €1,727.75, 1 POL = €0.08 (Chainlink, for conversion only).

Where the live values come from and where the calculator stops

Our server fetches the live values at most once a minute from public access points of the chains: the fees of the last 20 blocks via eth_feeHistory, the blob fee, the price per byte on Arbitrum and the scalars from the OP Stack gas oracle. Your browser contacts only aureum-tech.com. The providers see our server, not you, and what you enter does not leave your device. If one provider does not respond, the server asks the second; if a chain fails completely, the calculator uses the snapshot figures for it and says so.

Public access points (RPC), fixed list
ChainProvider, in order
EthereumPublicNode (Allnodes), then dRPC
Arbitrum OneOffchain Labs, then PublicNode (Allnodes)
OP MainnetOptimism Foundation, then PublicNode (Allnodes)
BaseCoinbase (Base), then PublicNode (Allnodes)
Polygon PoSPublicNode (Allnodes), then dRPC

The rate comes from Chainlink feeds, read directly from the contracts on Ethereum and Polygon. It is used only for conversion into euros and is shown with a timestamp. Price oracles also deviate regularly from the current market rate and have had outages Liu et al. 2021. The server therefore discards any rate that is older than allowed and then uses the snapshot figures. EUR/USD is idle at weekends, like the foreign exchange market.

Rates for conversion
FeedNetworkupdated fromat the latest
ETH / USDEthereum0.5% deviation1 h
EUR / USDEthereum0.15% deviation24 h
POL / USDPolygon PoS0.25% deviation27 s

Where the calculator stops

  • Gas amounts are medians from a sample of 14.11.2026 or protocol values. Your contract may need considerably more or less; eth_estimateGas or your wallet's estimate is authoritative.
  • The layer 1 data fee on Arbitrum, OP Mainnet and Base depends on the compressed size of your transaction. The tool assumes a typical size for each action.
  • Slippage, exchange fees in the pool, bridge fees and the costs of zkRollups are not included.
  • The Chainlink rate is used only for conversion into euros. It can differ from the market by the deviation threshold of the feed and is not a price forecast.
  • Status of the protocol rules: Ethereum after Fusaka and BPO2, OP Stack after Jovian, Arbitrum Nitro. The planned Glamsterdam changes the base fee, storage costs, contract size (EIP-7954), state access (EIP-8038) and the calldata floor.

If you are planning your own application, you need more than an average: the gas usage of your own contracts, the number of transactions per user and a decision on who bears the fees. How we approach this is described under Web3 development, and how users see what they are approving before they sign, in the section Making transactions understandable.

What the research says

The assumptions in the calculator are based on 15 peer-reviewed papers. Five findings support the most important decisions; the full list with review notes is further down.

  • Base fee plus a small tip is enough. Bidding more achieves nothing outside demand spikes Roughgarden 2024. That is why the speed changes only the tip.
  • The base fee lags behind when demand is high. When demand surges, it follows only slowly Reijsbergen et al. 2021. That is why the calculator gives a cap with a buffer for six full blocks.
  • Tip as a probability. The fee needed is a price for a desired chance of inclusion within a given time Laurent et al. 2022. That is why the speed uses percentiles rather than a fixed value.
  • Blobs made layer 2 cheap, but not calm. Rollup fees fell Park et al. 2026, but the blob market had jumps and losses Heimbach and Milionis 2026. That is why the calculator shows the blob fee live.
  • In a swap, more than gas matters. Front-running by bots is measurable and large Daian et al. 2020. That is why the calculator points to slippage for a swap.

Using this tool from an AI assistant

Assistants such as Claude or ChatGPT that speak the Model Context Protocol (MCP) can call this page’s calculation directly instead of rebuilding it. The server is public, needs no sign-in and stores nothing. The answer always includes the link to this page, the date of the data and the limits of the result.

Endpoint
https://aureum-tech.com/mcp
Tool
werkzeug_gas_rechner
Transport
Streamable HTTP, read-only
Inputs
NameRequiredMeaning
aktionyesWhat is to be paid for: eth = send ETH or POL; erc20 = send token (ERC-20); erc20neu = send token to new address (ERC-20); approve = grant approval (approve); swap = swap on an exchange; nft = transfer NFT (ERC-721); mint = mint NFT; daten = anchor data (e.g. a checksum); deploy = deploy contract; eigen = custom gas usage.
temponoTip as a percentile of the last 20 blocks: langsam (slow) = 10th, normal = 50th (default), schnell (fast) = 90th.
chainsnoWhich chains are compared: eth = Ethereum, arb = Arbitrum One, op = OP Mainnet, base = Base, pol = Polygon PoS. All if omitted.
code_bytesnoOnly with aktion=deploy: size of the runtime bytecode in bytes (at most 24576). 8000 if omitted.
konstruktor_gasnoOnly with aktion=deploy: gas for the constructor. 110000 if omitted (median of a sample).
daten_bytesnoOnly with aktion=daten: number of bytes to be put on the chain as calldata, e.g. 32 for a SHA-256 checksum.
gasnoOnly with aktion=eigen: known gas usage, for example from eth_estimateGas or the block explorer.
calldata_bytesnoOnly with aktion=eigen: size of the calldata in bytes, for the layer 1 data fee of the layer 2. 0 if omitted.
livenotrue (default): fetch current fees and rates, at most 60 s old. false: calculate with the fixed snapshot; the result is reproducible.
eth_eurnoOptional: your own rate for 1 ETH in euros instead of Chainlink.
pol_eurnoOptional: your own rate for 1 POL in euros instead of Chainlink.

Questions where an assistant should call it

  • “What does a USDC transfer on Ethereum cost right now compared with Base and Arbitrum?”
  • “How expensive is it to deploy a smart contract with 12 KB of bytecode on Ethereum and on Arbitrum?”
  • “What does it cost to anchor the SHA-256 checksum of a document on Polygon or Base?”
  • “Why is the fee on a layer 2 split in two, and how high is the share for Ethereum right now?”

How the connection works and what MCP can do in your own business is explained on the page about MCP integrations.

Common questions about gas and fees

What does a transaction on Ethereum cost right now?

That depends on gas usage and on the base fee, which changes with every block. On 14 November 2026 at 12:55, an ETH transfer (21,000 gas) cost about €0.005, a token transfer such as USDC (about 41,300 gas) about €0.0099 and a swap on an exchange about €0.03. The calculator above fetches the current values and works out all chains side by side.

What are gas and gwei?

Gas is Ethereum's unit of computation: every step of a transaction has a fixed gas price, and a simple transfer needs 21,000 gas. Gwei is the price per unit of gas, one billionth of an ETH. The cost is gas × (base fee + tip). The network sets the base fee and burns it; the tip goes to the block builder.

Why does a token transfer to a new address cost more?

If the recipient has never held the token, the contract has to create a new storage slot for them. That costs 20,000 gas, compared with 2,900 for changing an existing value. In our sample, a transfer to known recipients came to about 41,300 gas and one to new recipients came to about 63,200, so about 21,900 gas more.

Why do you pay two fees on Arbitrum, OP Mainnet and Base?

A layer 2 executes the transaction itself and then posts the data to Ethereum in bundles so that anyone can verify it. The first fee pays for the execution, the second for the space on Ethereum. Since EIP-4844 this data goes into cheap blobs, and the data share is usually small. On OP Mainnet, however, execution is so cheap that the data share makes up almost the whole fee.

What does it cost to deploy a smart contract?

The calculation is based mainly on the size of the bytecode: 200 gas per byte of code, plus calldata, 32,000 gas for creation and the constructor. A contract of 8,000 bytes needs about 1,889,332 gas. On 14 November 2026 at 12:55 that came to about €0.45 on Ethereum and about €0.02 on Base. When the network is busy, it can be many times that on Ethereum.

How high should the fee cap (maxFeePerGas) be?

The base fee rises by at most 12.5% per block. About twice the current base fee plus the tip is enough for six full blocks in a row, so a good minute (factor 2.03). The cap is only a reserve: you pay the base fee of the block in which the transaction lands, plus the tip. You are not charged for the rest.

Is it worth waiting for lower fees?

For a single transfer, usually not; the base fee is low anyway when there is space. Waiting pays off for large operations such as a deploy, or for many transactions at once, when the network is busy. The “Slow” speed in the calculator only lowers the tip; against a high base fee, only a later time helps.

Where does the gas calculator stop?

It works with typical amounts of gas from a sample, with typical data sizes for layer 2 and with a rate that is used for conversion only. It does not calculate slippage, bridge fees or zkRollups. What your own application costs per user and which chain suits it, we work out with you from your contracts and volumes.

What the tool is based on

The guide values and notes above are based on these 15 scientific papers. Each one has been checked via its DOI or the publisher’s page (most recently on 14 November 2026): it exists, it has not been retracted, and its findings still hold for this question today.

  1. Roughgarden, T. (2024): Transaction Fee Mechanism Design. Journal of the ACM 71 (4), pp. 1–25

    Finding The EIP-1559 fee rule, with a burned base fee and a tip, is straightforward for users: bidding more than the base fee plus a small tip gains nothing. The exception is sudden demand spikes, when it becomes an auction again.

    In the tool The calculator takes base fee plus tip; the speed changes only the tip, and when the base fee is rising it points to the fee cap.

  2. Liu, Y.; Lu, Y.; Nayak, K. et al. (2022): Empirical Analysis of EIP-1559: Transaction Fees, Waiting Times, and Consensus Security. Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security (CCS), pp. 2099–2113

    Finding EIP-1559 shortened waiting times and evened out prices within a block, but hardly lowered fees. When the ETH rate fluctuated strongly, transactions waited considerably longer.

    In the tool Explains why the base fee of the next block is a good estimate, and why the text does not promise that Ethereum became cheaper as a result.

  3. Reijsbergen, D.; Sridhar, S.; Monnot, B. et al. (2021): Transaction Fees on a Honeymoon: Ethereum's EIP-1559 One Month Later. 2021 IEEE International Conference on Blockchain, pp. 196–204

    Finding On average, EIP-1559 reaches its goal. In the short term, however, block sizes fluctuate widely, and when demand surges, such as in NFT sales, the base fee follows only slowly.

    In the tool That is why the calculator gives, alongside the current value, a fee cap that covers six full blocks in a row.

  4. Leonardos, S.; Monnot, B.; Reijsbergen, D. et al. (2021): Dynamical analysis of the EIP-1559 Ethereum fee market. Proceedings of the 3rd ACM Conference on Advances in Financial Technologies (AFT), pp. 114–126

    Finding Whether the base fee settles depends on the step size of the adjustment. Below a threshold it is stable; above it, it becomes unstable or even chaotic; the range of the base fee can be bounded.

    In the tool Explains in the text why blocks alternate between full and half empty while the base fee moves in steps of at most 12.5%.

  5. Huberman, G.; Leshno, J. D.; Moallemi, C. (2021): Monopoly without a Monopolist: An Economic Analysis of the Bitcoin Payment System. The Review of Economic Studies 88 (6), pp. 3011–3040

    Finding No participant can set the fees. They arise in a market in which users pay for priority to avoid waiting; without a bottleneck they fall towards zero.

    In the tool Explains why the fee is almost zero when there is space in the block and rises when demand is high, and why the calculator gives current values rather than fixed prices.

  6. Pierro, G. A.; Rocha, H. (2019): The Influence Factors on Ethereum Transaction Fees. 2019 IEEE/ACM 2nd International Workshop on Emerging Trends in Software Engineering for Blockchain (WETSEB), pp. 24–31

    Finding In measurements every 15 seconds, fees depended mainly on the number of waiting transactions and the number of miners, and considerably less on the US dollar rate or on electricity prices.

    In the tool The calculator keeps the gas price in gwei, which follows utilisation, separate from the conversion into euros, which follows the exchange rate.

  7. Laurent, A.; Brotcorne, L.; Fortz, B. (2022): Transaction fees optimization in the Ethereum blockchain. Blockchain: Research and Applications 3 (3), 100074

    Finding The fee needed can be seen as the minimum price for a desired probability of inclusion within a given time, and can be estimated from real block data.

    In the tool Basis of the three speeds: the 10th, 50th and 90th percentiles of tips from the last 20 blocks instead of a fixed value.

  8. Messias, J.; Pahari, V.; Chandrasekaran, B. et al. (2023): Dissecting Bitcoin and Ethereum Transactions: On the Lack of Transaction Contention and Prioritization Transparency in Blockchains. Financial Cryptography and Data Security, Lecture Notes in Computer Science, pp. 221–240

    Finding Private channels to the block builders are widespread. Which transactions compete and what is paid for it is only partly visible in public, and users of private channels sometimes pay more without getting anything in return.

    In the tool The tip suggestions come from the blocks themselves; the text says that they guarantee nothing for private channels.

  9. Daian, P.; Goldfeder, S.; Kell, T. et al. (2020): Flash Boys 2.0: Frontrunning in Decentralized Exchanges, Miner Extractable Value, and Consensus Instability. 2020 IEEE Symposium on Security and Privacy (SP), pp. 910–927

    Finding Bots outbid each other in fee auctions to front-run swaps on exchanges. The value that block builders can extract from the ordering of transactions (MEV) is measurable and large.

    In the tool For a swap, the calculator points out that the price disadvantage from slippage and front-running often costs more than the gas and is not included in the calculation.

  10. Thibault, L. T.; Sarry, T.; Hafid, A. S. (2022): Blockchain Scaling Using Rollups: A Comprehensive Survey. IEEE Access 10, pp. 93039–93054

    Finding Rollups execute transactions outside Ethereum and post the data there in bundles. Optimistic and ZK rollups differ in how correctness is proven; both have their own limits.

    In the tool Divides the layer 2 fee into execution and data share, and explains why zkRollups are not in the calculator.

  11. Crapis, D.; Felten, E. W.; Mamageishvili, A. (2024): EIP-4844 Economics and Rollup Strategies. Financial Cryptography and Data Security. FC 2024 International Workshops, Lecture Notes in Computer Science, pp. 135–149

    Finding A blob costs the same per block however full it is. Rollups with little traffic often do better with ordinary calldata, because a full blob would mean too long a wait; shared blobs do not always help.

    In the tool Explains why Arbitrum and the OP Stack spread data costs over a price per byte instead of working them out directly from the blob price.

  12. Heimbach, L.; Milionis, J. (2026): The Early Days of the Ethereum Blob Fee Market and Lessons Learnt. Financial Cryptography and Data Security, Lecture Notes in Computer Science, pp. 53–71

    Finding In the first months of the blob market, poor block packing at times led to a fee loss of up to 70%; in two busy periods, blob fees jumped sharply.

    In the tool That is why the calculator shows the blob base fee live and says that layer 2 data fees can rise quickly when blobs are in demand.

  13. Park, S.; Mun, B.; Lee, S. et al. (2026): Impact of EIP-4844 on Ethereum: Consensus Security, Ethereum Usage, Rollup Transaction Dynamics, and Blob Gas Fee Markets. IEEE Transactions on Network Science and Engineering 13, pp. 1294–1310

    Finding Since EIP-4844, rollups have used Ethereum more heavily, and their transactions have become cheaper. Because operators bundle data to save money, some rollup transactions only reach an Ethereum block with a delay.

    In the tool Explains the low data fees on layer 2 and the note that security from Ethereum comes later than confirmation on the layer 2.

  14. Liu, B.; Szalachowski, P.; Zhou, J. (2021): A First Look into DeFi Oracles. 2021 IEEE International Conference on Decentralized Applications and Infrastructures (DAPPS), pp. 39–48

    Finding The rates of large price oracles deviate regularly from the current market rate; outages and outliers also occurred.

    In the tool The Chainlink rate is used only for conversion into euros, with a timestamp; if it is too old, the service discards it and uses the snapshot figures.

  15. Chen, T.; Li, X.; Luo, X.; Zhang, X. (2017): Under-optimized smart contracts devour your money. 2017 IEEE 24th International Conference on Software Analysis, Evolution and Reengineering (SANER), pp. 442–446

    Finding The Solidity compiler did not remove costly programming patterns. In 4,240 real contracts, three such patterns occurred in 80 to 93.5% of cases.

    In the tool That is why the table gives ranges instead of a fixed value, and the text says that the same purpose can cost considerably more gas depending on the contract.

Standards, authorities and manufacturer data

Mon–Fri 8:00–17:00

Are you planning an application on Ethereum or a layer 2? Send us the copied result or your contracts. We will work out the fees per user and propose a chain, from Rodewisch.

+49 3744 4399760[email protected]

Mon–Fri 8:00–17:00 · Straße der Jugend 5, 08228 Rodewisch

Discuss your project
+49 3744 4399760Mon–Fri 8:00–17:00
CallCallback

PARTS LIST · SHEET 1/1

What this page is made of

Values measured at build time. With JavaScript, your browser measures them again live.

ITEMPARTTYPESIZE
01index.htmlDocument≈ 29 kB
02styles.cssStylesheet≈ 20 kB
03scripts.jsScript≈ 13 kB
04archivo-variable.woff2Font≈ 58 kB
05ibm-plex-mono.woff2Font≈ 30 kB
06favicon.svgImage< 1 kB

Advertising or analytics cookies: none. Without your consent, the page loads nothing from third-party servers apart from Cloudflare, our delivery service. Fonts, scripts and images are hosted by us. That is how we build your IT, too.

OUTSIDE OFFICE HOURS

Feel free to call anyway

Our regular office hours are Monday to Friday, 8:00–17:00.

In the evening and at weekends the number is forwarded to us, and often someone will pick up. We cannot promise that, though.

If nobody answers, leave a short message on the voicemail: your name, your number and what it is about. We will call back as soon as we can, on the next working day at the latest.

The call costs nothing. If someone has to start work straight away outside office hours, the surcharge in the price list applies.

Call now+49 3744 4399760

Prefer to write? Use the contact form or e-mail [email protected]. We will read it on the next working day.

DIAGNOSIS · AT-K2

What do you need help with?

↵ SELECTESC CLOSECALL: +49 3744 4399760

Custom settings

Which services may load?

Strictly necessary

The choice you make here and your settings for colour scheme and motion. These are kept as small entries on your device and are never sent to a server. They are needed for the site to work and therefore cannot be switched off.