Step 3 of 9
Who keeps the list
Miners, stakers, nodes, and why the hardware differs.
Try it: could your computer check the list?
Slide to your machine. Networks whose official minimum you meet light up.
Published minimums for a full or validating node, from each project's docs. Marked est.: no official figure, so a sourced estimate. Meeting a minimum is not the same as running smoothly under load.
Two ways to pick the next page writer
What it takes to check the list yourself
bitcoin.org · rusty-kaspa crescendo guide · ethereum.org · docs.anza.xyz
Tap a bar for the requirement and its source.
Minimums or recommendations from each project's docs, checked Nov 16, 2026. Marked est.: no official figure, so a sourced estimate. Bar length uses a square-root scale so small values stay visible.
Why does hardware differ so much?
Three things drive it: how many transactions the chain packs in (more data per second needs more bandwidth and disk), how much live state it keeps (every account and contract a node must remember), and whether it keeps old history. Solana chose huge throughput, so validators need data-center machines. Bitcoin chose small blocks, so a laptop can verify it. Kaspa sits between: fast blocks, but old block data is pruned.
Why Kaspa nodes stay relatively light (the honest reasons)
A Kaspa node keeps the recent DAG (about 30 hours of blocks) plus a compact proof and the set of unspent coins, then prunes old block data. A new node syncs from that proof instead of downloading years of history. Bitcoin Core can also prune to 7 GB, but every new node still downloads and checks about 740 GB once. The trade-off: a pruned Kaspa node cannot hand you the full history, and archive nodes do that job. Kaspa still asks for 8 cores and 16 GB of RAM, more than a minimal Bitcoin node, because it processes ten blocks a second.