Mining overview
The first application of Oreloom’s intelligence platform: measuring mining work and developing evidence-led decision models.
The initial application
The mining workspace brings together Zcash network readings, pool observations, rental orders, receipts, and illustrative calculations. Each source has its own availability and timestamp.
Intelligence & operations
| Mining Operations | Mining Intelligence |
|---|---|
| What was ordered, delivered, or received? | What can those observations tell us? |
| Rental orders, accepted pool work, balances, and receipts. | Performance analysis, economic scenarios, difficulty dynamics, and model uncertainty. |
| Supported by dated operational records. | Requires stated assumptions and validation before it supports decisions. |
Available & research scope
- Available: descriptive source readings, performance reports, an effort calculator, and exports.
- Research: profitability models, capital versus operating cost analysis, difficulty and hashpower dynamics, and mining-gap analysis.
- Planned: a decision-support engine and monitored execution workflows.
Future roadmap: verifiable GPU compute
Oreloom's longer-term roadmap includes exploring an Oreloom-developed GPU worker on existing GPU hardware. The intended direction is a verifiable compute workflow: assign a workload, check the submitted result, and retain an auditable record of accepted contributions.
This is a future research direction. Implementation has not started, and no Oreloom GPU worker, compute network, or GPU reward program is currently available. The current mining workspace continues to report Zcash operations; those records do not demonstrate GPU compute capabilities.
- Evaluate candidate workloads, including matrix computation, and determine hardware requirements before selecting an algorithm or supported GPU models.
- Design worker registration, job assignment, result verification, and contribution accounting with an explicit trust model.
- Publish runnable worker releases, reproducible GPU benchmarks, and independently repeatable validation before reporting the capability as implemented.
- Evaluate reward accounting and settlement only after compute validation; any GPU reward policy would require separate design and review.