diff options
Diffstat (limited to '_site/log/matrix-digital-rain/index.html')
| -rw-r--r-- | _site/log/matrix-digital-rain/index.html | 162 |
1 files changed, 33 insertions, 129 deletions
diff --git a/_site/log/matrix-digital-rain/index.html b/_site/log/matrix-digital-rain/index.html index 659922d..f47ac5e 100644 --- a/_site/log/matrix-digital-rain/index.html +++ b/_site/log/matrix-digital-rain/index.html @@ -2,12 +2,12 @@ <html> <head> <meta charset="utf-8"> - <title>Recreating the Matrix rain with ANSI escape sequences</title> + <title>Matrix Rain: 2025 refactor</title> <head> <meta charset="utf-8"> <meta name="viewport" content="width=device-width, initial-scale=1"> - <title>Recreating the Matrix rain with ANSI escape sequences</title> + <title>Matrix Rain: 2025 refactor</title> <link rel="stylesheet" href="/assets/css/main.css"> <link rel="stylesheet" href="/assets/css/skeleton.css"> </head> @@ -41,141 +41,45 @@ <main> <div class="container"> <div class="container-2"> - <h2 class="center" id="title">RECREATING THE MATRIX RAIN WITH ANSI ESCAPE SEQUENCES</h2> + <h2 class="center" id="title">MATRIX RAIN: 2025 REFACTOR</h2> <h6 class="center">21 DECEMBER 2025</h5> <br> - <div class="twocol justify"><p>My 2022 implementation of the Matrix rain had too many loose ends. Unicode -support was inflexible: the character set had to be a single contiguous block -with no way to mix ASCII with something like Katakana; Phosphor decay level was -stored in a dedicated array–still don’t understand why I did that when I had -already used bit-packing for the RGB channels; The algorithm was difficult to -decipher. The 2022 version worked, but that’s not the same thing as correct.</p> - -<p>I began by placing the decay factor in the LSB of the 4-byte RGB value. The PD -value plays a somewhat analogous role to an alpha channel in that both -influence transparency. However, they work very differently. So, I avoided -labelling it A so as not to cause confusion:</p> - -<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>enum { - R, /* Red */ - G, /* Green */ - B, /* Blue */ - PD /* Phosphor decay level */ -}; - -typedef union color_tag { - uint32_t value; - unsigned char color[4]; -} color; -</code></pre></div></div> - -<p>The decision to use union over more portable bit twiddling was made three years -ago, as I recall, for readability. Seeing as all my systems are little-endian, -this is unlikely to cause any trouble. Besides, if union is never to be used, -why is it in the language anyway?</p> - -<p>The blend() function, which emulates the dim afterglow of Phosphor by eroding -the RGB channels towards the background, with minor refactoring, remains as -elegant as it did three years ago:</p> - -<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>#define DECAY_MPLIER 2 - -static inline void blend(matrix *mat, - size_t row, size_t col) -{ - unsigned char *color; - - color = mat->rgb[index(mat, row, col)].color; - color[R] = color[R] - (color[R] - RGB_BG_RED) / DECAY_MPLIER; - color[G] = color[G] - (color[G] - RGB_BG_GRN) / DECAY_MPLIER; - color[B] = color[B] - (color[B] - RGB_BG_BLU) / DECAY_MPLIER; -} -</code></pre></div></div> - -<p>While the memory inefficiency of Phosphor decay was a technical oversight I -hadn’t noticed, the limitation around mixing nonadjacent Unicode blocks was a -nagging concern even three years ago. So, a fix was long overdue.</p> - -<p>In the new version, I introduced an array that enables a user to add as -many Unicode blocks as they want. The insert_code() function picks a block -from it at random, and then picks a character from that block at random:</p> - -<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>#define UNICODE(min, max) (((uint64_t)max << 32) | min) - -static uint64_t glyphs[] = { - UNICODE(0x0021, 0x007E), /* ASCII */ - UNICODE(0xFF65, 0xFF9F), /* Half-width Katakana */ -}; - -static uint8_t glyphlen = (sizeof glyphs) / (sizeof glyphs[0]); - -static inline void insert_code(matrix *mat, - size_t row, size_t col) -{ - uint64_t block; - uint32_t unicode_min, unicode_max; - - block = glyphs[(rand() % glyphlen)]; - unicode_min = (uint32_t)block; - unicode_max = (uint32_t)(block >> 32); - - mat->code[index(mat, row, col)] = rand() - % (unicode_max - unicode_min) - + unicode_min; -} -</code></pre></div></div> - -<p>The Unicode blocks are stored in 8-byte containers: the low four bytes form the -first codepoint and the high four bytes the last. Here, I chose bitwise -operations over unions because, first and foremost, the operations themselves -are trivial and idiomatic, and the UNICODE() macro simplifies the management of -charsets.</p> - -<p>The init_term() function is the arbiter of this zero-dependency software. It -prepares the graphical environment so that I can interact with it via ANSI -escape codes instead of unnecessary layers of abstraction:</p> - -<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>static inline int init_term(const struct winsize *ws) -{ - struct termios ta; - - if (tcgetattr(STDIN_FILENO, &ta) == 0) { - ta.c_lflag &= ~ECHO; - if (tcsetattr(STDIN_FILENO, TCSANOW, &ta) == 0) { - wprintf(L"\x1b[48;2;%d;%d;%dm", - RGB_BG_RED, RGB_BG_GRN, RGB_BG_BLU); - wprintf(L"%s", ANSI_FONT_BOLD); - wprintf(L"%s", ANSI_CRSR_HIDE); - wprintf(L"%s", ANSI_CRSR_RESET); - wprintf(L"%s", ANSI_SCRN_CLEAR); - setvbuf(stdout, 0, _IOFBF, 0); - ioctl(STDOUT_FILENO, TIOCGWINSZ, ws); - return 1; - } - } - return 0; -} -</code></pre></div></div> - -<p>insert_code() seeds the Matrix, blend() creates the old monochrome CRT display -nostalgia, and ANSI control sequences paint the screen. The result is a digital -rain that captures the original Matrix aesthetic with high visual fidelity:</p> - -<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>$ cc -O3 main.c -o matrix -$ ./matrix -</code></pre></div></div> + <div class="twocol justify"><p>Fixed the Unicode issue finally. Can now mix ASCII + Katakana.</p> + +<p>Took me 2 hours to decipher how this even works. For future me: mat.col[] +stores shuffled column indices, mat.row[] tracks last updated row per column. +shuffle() randomizes the working set, index i (line 333) and lines 364-370 draw +one column at a time, swap() rotates columns in and out. That’s the rain.</p> + +<p>Moved Phosphor decay level into the LSB of the RGB union - should’ve done this +in 2022 instead of separate array. WTF was I thinking?</p> + +<p>Keeping the RGB/PD union as it is. I know the ‘portability’ nerds hate it, but +I’m on a little-endian machine, and I’m the only one reading this. It’s +cleaner.</p> + +<p>New charset array works. UNICODE(min, max) macro packs the range into uint64. +insert_code() picks random block, unpacks it, and picks random char. Elegant.</p> + +<p>Looks a lot like the original now:</p> <video style="max-width:100%;" controls="" poster="poster.png"> <source src="matrix.mp4" type="video/mp4" /> </video> -<p>There was no cause to measure the program’s performance characteristics -precisely; it’s gentle on the CPU. On my ThinkPad T490 running OpenBSD, which -has a resolution of 1920x1080, it uses about 2-3% of the CPU, with occasional -jumps of up to about 8%; the cores remain silent, the fans don’t whir, the rain -falls in quiet.</p> +<p>Using half-width Katakana (U+FF61-U+FF9F) because full-width characters break +columns.</p> + +<p>blend() is still good, left it alone.</p> + +<p>Tossed the license and automake cruft. Just <code class="language-plaintext highlighter-rouge">cc -O3 main.c -o matrix</code>. Don’t +need the ceremony.</p> + +<p>Performance regressions: none. Runs like a charm on the T490. 2% CPU. No +whirring fans.</p> -<p>Files: <a href="source.tar.gz">source.tar.gz</a></p> +<p>Commit: +<a href="https://git.asciimx.com/matrix-digital-rain/commit/?id=03f8d87ba7c2e46bd3f3cc4c772fb3a2ac740c92">03f8d87</a></p> </div> <p class="post-author right">by W. D. Sadeep Madurange</p> |