EVM Resource # Breakout #3, March 26, 2025

[dcrapis]

EVM Resource # Breakout #3, March 26, 2025

• Mar 26, 2025, 16:00 UTC
• Duration in minutes : 60
• Recurring meeting : false
• Occurrence rate : none
• Zoom link

Agenda

• Empirical analysis of historical opcode and resource usage @misilva73
• Benchmarking to align hash function gas cost with ZK proving cost @ivokub
• EIP-7762 Increase MIN FEE per Blob @dcrapis
• Update on Gas Re#, EIP-7904 @ JacekGlen
• Gas model for memory that is paged @shemnon
• Discussion

Other comments and resources

The zoom link will be sent to the facilitator (please fill in the email and telegram)
Facilitator email: davide.crapis@ethereum.org
Facilitator telegram: davidecrapis

[ivokub]

I would request a short slot (5-7 min) to explain benchmarking I did to align hash function gas cost with ZK proving cost. I have summarized the findings at https://hackmd.io/@ivokub/H1t9SIg2kl

[misilva73]

I want to present some early results of an empirical analysis of historical opcode and resource usage (~10min). This is the continuation of the multidimensional gas analysis Shouqiao presented in the last breakout session.

[poojaranjan]

Recording: https://youtube.com/live/vV2UlSMlr6E

[shemnon]

Notes from notegpt.io

---

Summary

The session focuses on comprehensive research conducted into Ethereum’s multi-dimensional # and the historical use of opcodes for gas metering proposals. Maria begins with an analysis of historical opcode data, assessing how changes to gas metering could influence transaction throughput and block utilization. The project aims to provide empirical insights by breaking down gas costs based on resource types, such as memory and compute, to gauge the effects of various opcode costs under differing metering regimes.

Maria highlights challenges in data collection and the complexities of assigning accurate gas costs to opcodes, which included facing anomalies related to gas refunds. Collaborations with other researchers aim to further elucidate opcode cost distributions and rectify textual discrepancies within the data. Her next steps involve enhancing resource mapping for gas costs while delving into specific opcode anomalies, particularly with CREATE and CREATE2.

Evo then presents on the proving costs of hash functions in Ethereum, emphasizing the discrepancy between native execution speeds and gas costs associated with proving in zero-knowledge systems. He discusses EIP 7667, targeting gas cost modifications for hash functions relevant to native rollups. Testing revealed that the utilization of hash functions is minor, with the uniqueness of functions like RipeMD leading to under# concerns.

Evo proposes adjustments to EIP 7667 to align gas costs with performance, recommending dynamic adjustments based on input sizing and usage patterns. The discussion wraps up with an emphasis on collaborative efforts to address ongoing challenges in opcode # and the proving methodology.

Throughout the presentation, several proposals concerning gas costs were discussed, including gas usage assessments in real-life transactions, re# proposals, and market demand analyses. Suggestions included moving to simpler memory access charge models, adjustments to opcode costs based on complexity, and initiatives for ensuring stability and predictability in transaction costs.

Highlights
🔍 In-depth Analysis of Opcode Usage: Maria’s research evaluates historical opcode data and its implications on Ethereum’s gas metering.
🔄 Collaborative Efforts: Ongoing partnerships among researchers aim to clarify opcode cost distributions and address discrepancies.
⚙️ Proving Costs in Zero-Knowledge Systems: Evo addresses the cost discrepancies related to hash functions and proposes EIP 7667 adjustments.
📊 Gas Usage Insights: An assessment reveals the limited impact of increasing gas costs on overall transaction efficiency.
💸 Re# Proposals: Discussions around static and dynamic cost adjustments are aimed at refining the costs associated with various operations.
🛠️ Memory Access Charge Adjustments: A recommendation to transition to a simpler page costing format to ease memory access computations.
📈 Future Planning Considerations: Participants engage in discussions on long-term adjustments, ensuring # reflects demand and stabilizes the network.
Key Insights

📉 Complexity in Gas Cost Assignment: The intricate nature of gas costs assignment is crucial for accurately understanding transaction fees, and Maria’s work highlights how this can significantly affect network efficiency.

Analyzing opcode usage is essential in proposing shifts to gas metering practices, as seen in Maria’s exploration of how memory and compute types influence costs.

🤝 Importance of Collaboration: Collaborative efforts among researchers are vital in addressing inconsistencies in historical data and opcode cost distributions.

By pooling resources and expertise, researchers can identify anomalies and develop universally accepted gas # models, ultimately benefiting the Ethereum ecosystem.

🔍 EIP 7667 and Hash Function Costs: Evo’s exploration into EIP 7667 reveals the necessity of refining hash function costs to balance operational speeds with gas #.

Linkages between proving performance and gas costs can lead to more efficient resource utilization, ensuring developers maintain competitiveness within the Ethereum environment.

🔬 Market Demand Misunderstandings: Historical demand for Ethereum resources indicates periods of oversupply, suggesting a need for better tracking of market behaviors.

This data underlines the importance of continual adaptation to # models to reflect true market conditions, as evidenced by the proposed adjustments discussed during the session.

🚀 Memory Access Charges Simplification: Changing memory access charges to a page costing format demonstrates a forward-thinking approach to resource allocation.

A simpler model could alleviate developer frustrations regarding costs, enabling smarter use of memory in more complex applications.

⚖️ Efficient Opcode Cost Structures: Adjustments to opcode costs based on computational complexity could promote efficiency and innovation within blockchain solutions.

By recalibrating costs, Ethereum can foster an environment conducive to developing advanced smart contracts without prohibitive expenses.

📅 Long-term # Strategies: The session concludes emphasizing the need for evolving gas # strategies to accommodate future complexities.

Preparing for a more intricate # model can help Ethereum adapt to emerging technological advancements and user demands, securing its position as a leading blockchain platform.