What is signal-to-noise ratio in AI coding?

SNR = output / (input + output). It measures the share of fresh conversational traffic represented by model output. High SNR means most of your fresh traffic is output; low SNR means most is input. It is a bounded (0-1) view of Velocity: S = V/(1+V). SNR measures the compression/transmission structure, not the semantic quality of the output.

What counts as signal vs noise in AI coding?

In SigRank, signal is model output and noise is fresh input. Output is the work the model generates for you; input is the fresh context you provide. SNR measures what fraction of the fresh traffic (input + output) is actually output. It does not independently judge the semantic quality of that output — a brilliant response and a mediocre one both count as signal. The distinction is about transmission structure, not content quality.

How does SNR relate to the Conservation Law of Commitment?

The Conservation Law of Commitment states that commitment content survives transformation only with an enforcement gate. In token-cascade terms, signal (the commitment content of your context) survives across turns only when you maintain a stable, structured context — the enforcement gate is your context discipline. Without it, signal decays into noise with each transformation.

How do I increase my signal-to-noise ratio?

Increase signal by maximizing cache reuse (stable context, structured prefixes) and requesting substantive output. Decrease noise by trimming fresh input, avoiding re-pasted context, and building on prior turns instead of restarting. The highest-leverage move is conversation continuity — every turn that reuses cache adds signal without adding noise.

◈ Signal Density

Signal-to-Noise Ratio (SNR)

SNR in AI coding measures the share of your fresh token traffic that is output versus input. A high signal-to-noise ratio means most of your fresh traffic is productive output; a low SNR means most is input. A bounded view of Velocity.

The SNR formula

SNR = output / (input + output)

Where output is tokens the model generates and input is fresh context you provide. The ratio ranges from 0 to 1. An SNR of 0.8 means 80% of your fresh conversational traffic is model output; an SNR of 0.2 means 80% is fresh input.

SNR is a bounded transformation of Velocity (V = output / input). The relationship is exact: S = V / (1 + V). This places the unbounded velocity ratio onto a 0-1 scale, making it easier to compare, threshold, and visualize. SNR does not independently judge the semantic quality of the output — it measures transmission structure, not content quality.

What signal vs noise means in AI coding

Signal (output)

output — tokens the model generates in response to your input. This is the productive work: code, explanations, analysis, refactors. Output is what you want from the interaction.

Noise (input)

fresh input — tokens you send to the model: prompts, instructions, pasted code, context. Input is necessary but not the goal. High input relative to output means you’re spending more than you’re getting.

The distinction is not about content quality — it’s about transmission structure. A brilliant prompt and a redundant one both count as input. A correct response and an incorrect one both count as output. SNR measures what fraction of the fresh traffic is output, not whether that output is good. Quality is a separate question that requires external evaluation.

Why high SNR matters

High SNR means most of your fresh conversational traffic is output. You send relatively little input and get relatively much output back. This is the signature of an operator who writes efficient prompts — concise instructions that produce substantial responses.

Low SNR means most of your fresh traffic is input. You’re sending large prompts, pasting extensive context, or re-explaining things — and getting relatively little output back. The model processes your input, gives a short response, and you need to send more input to continue.

SNR is also a cost indicator. Output tokens are typically more expensive than input tokens, but high SNR means you’re getting more value per token of input you provide. Low SNR means you’re paying for input that doesn’t produce proportional output.

How to increase signal density

1. Maintain conversation continuity. The single highest-leverage move. Every turn that reuses cached context adds signal without adding noise. Don’t restart sessions unnecessarily — the cached prefix from turn 5 is free signal on turn 50.
2. Use stable, structured prefixes. Put project conventions, file layout, and coding standards at the top of your context — in a consistent order. Stable prefixes get cached; chaotic, re-ordered context breaks the cache and forces fresh processing.
3. Reference instead of re-pasting. If the model already has a file in context, reference it by name rather than pasting its contents again. “Add tests to the auth module we discussed” is signal; re-pasting auth.ts is noise.
4. Request substantive output. More output tokens means more signal in the numerator. Ask for complete deliverables — functions, modules, test suites — rather than one-line answers. Output compounds: today’s generated code becomes tomorrow’s cached context.
5. Avoid context-switching mid-session. Jumping between unrelated tasks in one session dilutes the cache. Each topic switch forces fresh input to establish new context. Group related work into focused sessions where the cache stays relevant.

Relationship to the Conservation Law of Commitment

The Conservation Law of Commitment states that C(T(S)) ≈ C(S) with enforcement — commitment content survives transformation only when an enforcement gate is present in the pipeline. Without the gate, commitment decays: C(T(S)) < C(S).

In token-cascade terms, the “commitment content” of your context is the signal — the accumulated understanding, conventions, and code that should survive across turns. The “enforcement gate” is your context discipline: stable prefixes, conversation continuity, structured prompts. With discipline, signal survives transformation (each new turn) and compounds. Without it, signal decays into noise — you re-explain, re-paste, and re-establish context every turn.

High SNR is the empirical signature of the conservation law holding: the enforcement gate (your context discipline) is present, and signal survives. Low SNR is the signature of the law failing: no gate, signal decays, noise dominates. The law is published under CC-BY-4.0 (DOI: 10.5281/zenodo.20029607).

FAQ

What is signal-to-noise ratio in AI coding?: SNR = output / (input + output). It measures the share of fresh conversational traffic represented by model output. High SNR means most of your fresh traffic is output; low SNR means most is input. It is a bounded view of Velocity: S = V/(1+V).
What counts as signal vs noise in AI coding?: Signal is model output; noise is fresh input. SNR measures what fraction of the fresh traffic (input + output) is output. It does not independently judge the semantic quality of the output — a brilliant response and a mediocre one both count as signal. The distinction is about transmission structure, not content quality.
How does SNR relate to the Conservation Law of Commitment?: The Conservation Law states commitment content survives transformation only with an enforcement gate. In token-cascade terms, signal survives across turns only when you maintain stable, structured context — the gate is your context discipline. Without it, signal decays into noise. High SNR is the empirical signature of the law holding.
How do I increase my signal-to-noise ratio?: Increase signal by maximizing cache reuse and requesting substantive output. Decrease noise by trimming fresh input and avoiding re-pasted context. The highest-leverage move is conversation continuity — every turn that reuses cache adds signal without adding noise.