Sample
Dorn/cuda/65
Human rating uses the original dataset scale.
Scores
| Method | Score | Scale | Dataset rank (0–100) |
|---|---|---|---|
| Posnett | 0.427 | 0–1; higher is more readable | 79.4Δ 8.9 |
| Dorn | 0.451 | 0–1; higher is more readable | 39.3Δ 49.0 |
| Mi | 0.731 | 0–1; higher is more readable | 89.4Δ 1.1 |
| LLOC | 5 | Logical lines; fewer is more readable | 78.0Δ 10.3 |
| DeepSeek V4 Pro | 12.33 | 0–20; mean of three runs | 55.8Δ 32.5 |
| GPT-6.1 Sol | 15 | 0–20; mean of three runs | 52.8Δ 35.5 |
| BTTF | 0.515 | 0–1; higher is more readable | 48.5Δ 39.8 |
| Human | 3.92 | 1–5; original rating | 88.3 |
Views
Select the downloaded ZIP or its extracted folder. Files stay in your browser.
Loading feature diagnosis…
Source regions
Feature contributions
Feature contributions
| Feature | Value | Contribution (log-odds) |
|---|---|---|
| Byte entropyHow varied the bytes in the source text are. Repeated bytes give a lower value; a more even mixture gives a higher value. | 4.795 | −7.192 |
| Halstead volumeHow much symbolic information the code contains. It combines the total number of operators and values with the number of distinct ones. | 180.942 | −5.971 |
| Source linesThe number of non-empty source lines. | 10 | +4 |
Intercept 8.87 + feature contributions = -0.293 (log-odds).
Score = sigmoid(-0.293) = 0.427.
Feature contributions
| Feature | Value | Contribution (log-odds) |
|---|---|---|
| Comment layout (horizontal)How rapidly the visual pattern of comment characters changes from left to right. | 105 | +0.89 |
| Identifier layout (horizontal)How rapidly the visual pattern of identifier characters changes from left to right. It measures spatial frequency, not the meaning of the names. | 105 | −0.782 |
| Comment layout frequencyHow frequently the layout switches between lines containing only comments or whitespace and lines containing code. | 9 | +0.718 |
| Indentation frequencyHow rapidly indentation changes from line to line. | 1 | −0.703 |
| Operator visual areaThe share of the code’s visual area occupied by operators such as +, =, and *. | 0.131 | −0.332 |
| Identifier layout (vertical)How rapidly the visual pattern of identifier characters changes from top to bottom. | 9.745 | −0.274 |
| Number layout frequencyHow rapidly the number of numeric values per line changes through the code. | 9 | +0.214 |
Intercept 0.07 + feature contributions = -0.198 (log-odds).
Score = sigmoid(-0.198) = 0.451.
| Score | 0.731 |
|---|
| Score | 5 |
|---|
Run 1
The code is generally readable with clear intent, flat structure, and good action granularity. The loop break condition is explicit and logical. Minor issues include a slightly dense printf format string, a leading stray closing brace suggesting a larger context, and a trailing blank line that could be cleaned up. Overall, the snippet is concise and easy to follow.
Run 2
The code snippet is partially readable but suffers from structural and formatting issues. It begins with a dangling closing brace, making it unclear what block is ending and harming reading flow. The nesting appears shallow, but the lack of context for the outer construct reduces clarity of intent. The verbose print format mixes integers with floating-point casts, slightly reducing readability. Action granularity is acceptable—one core action per line—but the floating-point conversion for kilobytes could be simplified or clarified. The `if (rbytes == 0 && !online)` condition is understandable but context-dependent. Overall, the fragment is not deeply nested but is ambiguous without surrounding code, and its indentation and formatting detract from clarity.
Run 3
The code snippet is a small fragment that shows basic structure but has readability issues. Variable names like 'rbytes' and 'total_rbytes' use shorthand that is not immediately clear. The use of 'online' as a boolean is ambiguous without context. The verbose logging line is overly long and mixes conversion logic with output formatting, making it harder to parse. The if statement is simple and flat, but the comments are minimal and could state purpose more clearly. Overall, the intent is somewhat discernible but requires context to fully understand, and some lines combine multiple actions, reducing clarity.
Run 1
Code Structure: The byte-count update, progress output, and termination check form a logical sequence, though the leading unmatched brace limits context. Nesting: Mostly flat, with only a simple conditional block. Clarity of intent: The variable names and termination comment make the purpose understandable, although the meaning of online depends on surrounding code. Code length: Concise, with no unnecessary logic. Action granularity: Most lines perform one action, but the verbose condition and fprintf call share a long line. Reading flow: Natural execution order and blank lines help scanning; inconsistent indentation, tight argument spacing, and mixed conditional formatting reduce readability.
Run 2
Code Structure: The sequence logically updates the byte count, reports progress, and checks for completion, but the leading unmatched brace obscures context. Nesting: Minimal nesting makes the snippet easy to scan. Clarity of intent: Variable names and the completion comment convey purpose, although the meaning of online depends on surrounding code. Code length: Concise, without unnecessary logic. Action granularity: Most lines contain one action, but the inline verbose condition and fprintf call compress control flow and output formatting. Reading flow: Generally natural, with useful spacing, though inconsistent indentation, brace placement, and the dense fprintf expression reduce readability.
Run 3
Code Structure: The update, progress report, and termination check are logically ordered, though the leading closing brace lacks context. Nesting: Shallow conditional logic is easy to follow. Clarity of intent: Byte-count tracking and termination are clear, aided by the comment, but the meaning of online is not explained. Code length: Concise without unnecessary operations. Action granularity: Most lines perform one action, but the verbose condition and fprintf call share a long line containing formatting and conversion. Reading flow: Natural overall, with blank lines separating steps; inconsistent indentation, spacing, and brace placement slightly reduce readability.