A giga node (GN) is one billion nodes. A node is one position the engine visits. The number is useful because it describes work rather than time: the same 100 GN can take very different amounts of wall-clock time on different hardware.
The ChessUrge conversion
| Unit | Value |
|---|---|
| 1 credit | 200 million nodes (0.2 GN) |
| 5 credits | 1 billion nodes (1 GN) |
| 500 credits | 100 billion nodes (100 GN) |
| $1 | 500 credits = 100 GN |
So 100 GN per dollar is not a separate currency. It is another way of saying 500 credits, with each credit buying 200 million nodes.
What does 100 GN buy?
- 10 MN per position: 10,000 positions. At 40 positions per game, that is 250 full-game reviews at the node ceiling.
- 100 MN per position: 1,000 positions, or about 25 full games at 40 positions each.
- 1 GN per position: 100 positions. Deep analysis is expensive because the node count grows quickly with depth.
These are ceilings, not promises. A quiet position may finish well below its cap. ChessUrge refunds the unused nodes after each position, so the actual cost is usually lower.
Why nodes instead of minutes?
Minutes depend on the CPU, the number of threads, the hash table and the position. Nodes are a stable measure of the work performed. That makes billing reproducible even when the server is busy or the engine is upgraded.
A practical example
A 40-move game has 41 positions. At depth 18, ChessUrge caps each position at roughly 4.5 million nodes. The absolute maximum is about 185 million nodes, or 0.92 credits. If the engine averages 2 million nodes per position, the actual charge is closer to 0.41 credits. The estimate is deliberately conservative.
You can see the current conversion on the pricing page and run a real search on the analysis board.
See the whole game, not just the blunder.
Choose a review depth, see the maximum cost, and get move-by-move accuracy with blunder, mistake and inaccuracy markers.