Read the Solana network and activated stake
Separate the SOL price from network state, voting transactions from other transactions, and activated stake from a promise of yield. Trace each value in its original units.
1. Start with the network, units and supply
The SOL detail shows a market price; the network view reads the public mainnet-beta RPC. Before using network data, it compares the genesis hash with the fixed identity of that network. If verification fails, other sections are not presented as actual Solana state. A price quote and a network sample have different sources and capture times.
Total, circulating and non-circulating supply are originally integer lamport counts. One SOL equals one billion lamports; the view converts exactly to nine decimal places. The original integer is in the tooltip and JSON. The two supply partitions must sum to the total. Circulating supply alone does not establish free float, an unlock schedule or staked volume.
2. Epoch, slot and block mean different things
An epoch has a source number, length in slots and current zero-based index. Absolute slot and block height are separate fields and cannot be interchanged. Displayed progress is index divided by epoch length, rounded to two decimal places. It is neither a count of produced blocks nor a prediction of the epoch end time.
Epoch, supply and vote accounts request finalized commitment. Individual RPC calls can still return different contexts. Independent capture times help identify input age; they do not promise an atomic shared snapshot.
3. Read TPS together with sample duration
Each sample supplies a transaction count, duration in seconds, slot count and ending slot. TPS is the original count divided by the original duration, rounded half-up to six decimal places. Total transactions include votes; the second series uses only a supplied non-voting transaction count. Missing values remain gaps, and a known zero remains zero.
The horizontal axis uses ending slots, not inferred UTC times. Recent performance samples accept no commitment parameter and do not establish finality or user transaction success. Choosing 30 or 60 requests at most that many recent points. The source may return fewer or an empty window; this is not long history or arbitrary backfill.
4. Refresh, inspect and download the same result
An actual read runs on opening and explicit refresh. A short shared cache limits repeated calls, so the button need not issue a new source request. Each method has its own capture and any unavailability reason. Unavailable sections are never replaced with zeros or synthetic data.
The sample chart and table use the same query and manifest. Full source JSON contains that exact bundle: all supplied samples, all vote accounts, epoch, supply and original integer values. A local validator filter does not restrict the export. A result identity or hash alone does not establish a retained historical archive. Preserve the selected limit, source clocks and methodology for further work.
This is a guide to Datisto. Private features require sign-in and access to your own data. Personal models use your assumptions; availability and permitted use of source data are checked with the particular result. This is not investment advice.