Blog

DAGKnight vs Bitcoin's Nakamoto Consensus: Same 50% Security, Very Different Speed

The DAGKnight paper's full title is A Parameterless Generalization of Nakamoto Consensus. In other words, DAGKnight is designed as an extension of the same idea that secures Bitcoin, not a replacement for it. This article compares the two: what they share, where they differ, and why it matters.

Published 15 November 2026

Bitcoin keeps one chain and discards blocks found at the same time, while Kaspa's blockDAG keeps parallel blocks and orders them all.

A quick refresher: how Nakamoto consensus works

Bitcoin's consensus, named after its creator Satoshi Nakamoto, is simple:

That design is very secure, but it forces Bitcoin to keep blocks slow. If blocks came much faster, more would be found at the same time, more would be orphaned, and security would suffer. Merchants also wait for several confirmations; a common convention is six, or about an hour.

What DAGKnight keeps from Bitcoin

What's different

Bitcoin (Nakamoto)Kaspa with DAGKnight (planned)
StructureSingle chainblockDAG: blocks can have many parents
Parallel blocksOrphaned and discardedKept and ordered
Block rateAbout 1 every 10 minutes10 per second today, with higher rates discussed
Latency assumptionSafe because blocks are slow relative to network delayNo fixed delay bound in the protocol; adapts to real conditions
ConfirmationsPaced by the 10-minute block timeDesigned to follow actual network speed
Security thresholdBelow 50% of hashrateBelow 50% of hashrate (per the paper)

The key shift is how parallel blocks are handled. Bitcoin avoids them by being slow. A blockDAG embraces them: every block can reference all the tips its miner has seen, so honest blocks mined at the same moment all count. The consensus protocol's job is then to put them in one agreed order.

Why that's hard, and how DAGKnight does it

Keeping every block creates a new problem: an attacker's blocks are kept too. The protocol needs to tell honest parallel blocks apart from an attacker's hidden chain.

Kaspa's current protocol, GHOSTDAG, does this with a fixed setting (k) based on an assumed worst-case network delay. DAGKnight goes further: it works out the right setting from the DAG itself, by looking for the group of blocks backed by a majority of the work that is most tightly connected. Honest miners publish immediately, so their blocks are well connected; an attacker mining in secret can't reference blocks it hasn't seen. Our parameterless consensus explainer covers this step by step.

One way to think about it: Bitcoin is the special case where only one block at a time is allowed in. DAGKnight lets many in and still recovers a single history.

What this means in practice

For a broader side-by-side, including supply and mining, see our Kaspa vs Bitcoin comparison.

Frequently asked questions

Is DAGKnight more secure than Bitcoin?

The paper claims the same headline threshold: security against attackers with less than 50% of hashrate. Real-world security also depends on the implementation, which is still being built, and on how much hashrate secures each network.

Could Bitcoin use DAGKnight?

In principle the ideas are general, but switching Bitcoin's consensus would need a major change that its community has shown no sign of pursuing.

Is DAGKnight live?

Not yet. Kaspa runs GHOSTDAG today. See our DAGKnight status update.

Not financial advice. This article is for education only. dagknight.com is a community project and is not affiliated with the Kaspa Foundation.

Sources