sf2000/tools/bootLogoChanger.htm

310 lines
16 KiB
HTML
Raw Normal View History

<!DOCTYPE html>
<html lang="en">
<head>
2023-05-09 20:05:19 +02:00
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<title>Data Frog SF2000 Boot Logo Changer</title>
<meta name="viewport" content="width=device-width">
<link rel="stylesheet" href="tools.css">
</head>
<body>
<h1>Data Frog SF2000 Boot Logo Changer</h1>
<p>This tool can be used to alter the boot logo on an SF2000 handheld gaming console. Please note this tool is provided as-is, and no support will be given if this corrupts your device's bios; make sure you have backups of anything you care about before messing with your device's critical files! 🙂</p>
<hr>
<div id="steps">
<section id="bisrvSection">
<h2>Step 1: Select Original <code>bisrv.asd</code></h2>
<p>Select the <code>bisrv.asd</code> file whose boot logo you want to modify. You should probably make a backup of the file first, just in case! You can find the <code>bisrv.asd</code> file in the <code>bios</code> folder on your microSD card.</p>
<div id="bisrvMessages"></div>
<div class="controlContainer">
<label class="control">Open <code>bisrv.asd</code>: <input id="bisrvSelector" type="file" accept=".asd" onchange="bisrvLoad(event.target.files[0])"></label>
</div>
</section>
</div>
<script src="tools.js"></script>
<script type="text/javascript">
// Global variables...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
let bisrvData; // Used to store the binary data from the bisrv.asd file
let logoOffset; // Will contain the offset of the boot logo within the bisrv.asd file
let newLogoData; // Used to store raw ImageData of the new boot logo
let ditherEnabled = false; // Holds whether the user wants to enable dithering for RGB565 mode or not
let ditherStrength = 0.2; // Holds the selected dither strength value - 0.2 seems to be a good general balance
// This function is triggered when the person selects a file in Step 1...
function bisrvLoad(file) {
// Read in the contents of the selected file as array buffer...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
const frBisrv = new FileReader();
frBisrv.readAsArrayBuffer(file);
frBisrv.onload = function(event) {
// Clear out any HTML that might already exist after Step 1...
while(document.getElementById("bisrvSection").nextSibling) {
document.getElementById("bisrvSection").nextSibling.remove();
}
// Clear any old messages...
document.getElementById("bisrvMessages").innerHTML = "";
// Read the provided file's data into an array...
bisrvData = new Uint8Array(event.target.result);
// We'll do a hash-check against it...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
const hashResult = getFirmwareHash(bisrvData);
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// The result could be either a Promise if it had a bisrv.asd-like
// structure and we got a hash, or false otherwise... let's check!
if (hashResult instanceof Promise) {
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// We got a Promise! Wait for it to finish so we get our bisrv.asd
// hash...
hashResult.then(function(dataHash) {
// Check the hash against all the known good ones...
switch (knownHash(dataHash)) {
// Mid-March BIOS...
case "03.15":
logoOffset = 0x9B9030;
setMessage("info", "bisrvMessages", "INFO: Mid-March <code>bisrv.asd</code> detected.");
break;
// April 20th BIOS...
case "04.20":
logoOffset = 0x9B91D8;
setMessage("info", "bisrvMessages", "INFO: April 20th <code>bisrv.asd</code> detected.");
break;
// May 15th BIOS...
case "05.15":
logoOffset = 0x9BB0B8;
setMessage("info", "bisrvMessages", "INFO: May 15th <code>bisrv.asd</code> detected.");
break;
// May 22nd BIOS...
case "05.22":
logoOffset = 0x9BB098;
setMessage("info", "bisrvMessages", "INFO: Version 1.5 <code>bisrv.asd</code> detected.");
break;
// August 3rd BIOS...
case "08.03":
logoOffset = 0x9B3530;
setMessage("info", "bisrvMessages", "INFO: Version 1.6 <code>bisrv.asd</code> detected.");
break;
// October 7th BIOS...
case "10.07":
logoOffset = 0x9B1FE8;
setMessage("warning", "bisrvMessages", "WARNING: Version 1.7 <code>bisrv.asd</code> detected; this version has known issues with SNES save states and is not recommended for use.");
break;
// October 13th BIOS...
case "10.13":
logoOffset = 0x9B3618;
setMessage("info", "bisrvMessages", "INFO: Version 1.71 <code>bisrv.asd</code> detected.");
break;
default:
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// Huh... wasn't false so had bisrv.asd structure, but didn't
// return a known hash... a new BIOS version? Unknown anyway!
console.log(dataHash);
setMessage("error", "bisrvMessages", "ERROR: While the file you've selected does appear to be generally structured like the SF2000's <code>bisrv.asd</code> BIOS file, the specifics of your file don't match any known SF2000 BIOS version. As such, this tool cannot safely modify the selected file.");
return;
}
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// If we're here we've got a good file, so onwards to Step 2
// (image selection)...
stepTwo();
});
}
else {
// We got false, so whatever it was, it wasn't a bisrv.asd...
setMessage("error", "bisrvMessages", "ERROR: The file you've selected doesn't appear to have the same data structure as expected for a <code>bisrv.asd</code> file.");
return;
}
};
}
// This function triggers when the user has selected a known `bisrv.asd`
// variant...
function stepTwo() {
// We're going to build up the HTML for Step 2, which has the user
// browse for a PNG or JPEG image file. We need to validate the chosen
// file is an image, and that it has the correct dimensions of 256x100
// pixels; if it is, we convert the image data to an SF2000 format and
// move on to Step 3. First, let's start building our HTML...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
let html = "<section id=\"imageSection\"><h2>Step 2: Select New Logo Image File</h2><p>Select the image file you want to use as the new boot logo. It must be 256 pixels wide and 100 pixels tall. Only PNG and JPEG image types are supported. For PNG files, transparency is handled by drawing the image on a black background; also, you might want to make sure your PNG image has any embedded gamma or colour profile information removed first using a metadata scrubbing tool, otherwise the colour output from <i>this</i> tool might be incorrect. On the SF2000, the boot logo is displayed centred on a black background, so you may want to factor that into your design as well.</p><div id=\"imageMessages\"></div>";
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// Add our file chooser and dithering controls...
html += "<form id=\"imageForm\" class=\"controlContainer\"><div class=\"control\">";
html += "<label class=\"control\"><input id=\"inputImage\" type=\"file\"></label>";
html += "<div class=\"control\"><label for=\"ditherEnabled\">Dithering: <input type=\"checkbox\" id=\"ditherEnabled\" name=\"ditherEnabled\"" + (ditherEnabled == true ? " checked" : "") + "></label> <label for=\"ditherStrength\">Strength: <select id=\"ditherStrength\" name=\"ditherStrength\"><option" + (ditherStrength == 0.1 ? " selected" : "") + ">0.1</option><option" + (ditherStrength == 0.2 ? " selected" : "") + ">0.2</option><option" + (ditherStrength == 0.3 ? " selected" : "") + ">0.3</option><option" + (ditherStrength == 0.5 ? " selected" : "") + ">0.5</option><option" + (ditherStrength == 0.75 ? " selected" : "") + ">0.75</option><option" + (ditherStrength == 1.0 ? " selected" : "") + ">1</option><option" + (ditherStrength == 2.0 ? " selected" : "") + ">2</option></select></label></div>";
html += "</div></form>";
// Next we'll add our image preview...
html += "<div id=\"previewContainer\" class=\"controlContainer\" style=\"display: none;\"><div class=\"control\"><canvas id=\"processFilePreview\" width=\"256\" height=\"100\"></canvas></div></div>";
// Close our section...
html += "</section>";
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// Finally, add a <hr> separator after the last step, and append the new
// step...
document.getElementById("steps").insertAdjacentHTML("beforeend", "<hr>");
document.getElementById("steps").insertAdjacentHTML("beforeend", html);
// Attach our event handler to our new file input control...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
const userImageInput = document.getElementById("inputImage");
userImageInput.addEventListener("change", function() {
// The user has chosen a new file; it should be either a PNG or JPEG
// image for converting to an SF2000 binary image format...
// First let's clear any old messages...
document.getElementById("imageMessages").innerHTML = "";
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// ... and make sure our preview image is hidden...
document.getElementById("previewContainer").style.display = "none";
// ... and clear our newLogoData object ...
newLogoData = new ImageData(1, 1);
// And clear up any HTML already added after the current step...
while(document.getElementById("imageSection").nextSibling) {
document.getElementById("imageSection").nextSibling.remove();
}
// Now let's read in the file...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
const frImage = new FileReader();
frImage.readAsDataURL(event.target.files[0]);
frImage.onload = function(event) {
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// The file is loaded; let's check to make sure we got a PNG or
// JPEG...
if (!event.target.result.includes("data:image/png;") && !event.target.result.includes("data:image/jpeg;")) {
// Whoops! Doesn't look like the user provided a PNG or a JPEG!
setMessage("error", "imageMessages", "ERROR: The selected file does not appear to be either a PNG or JPEG image; please make sure you're selecting an appropriate image file.");
return;
}
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// Now we're going to load the file's data into an Image object, so
// we can access the raw image data...
const img = new Image;
img.src = event.target.result;
img.onload = function(event) {
// Check to make sure the image has the correct dimensions for a
// boot logo...
if (img.width != 256 || img.height != 100) {
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
setMessage("error", "imageMessages", "ERROR: The selected image does not have dimensions of 256 x 100!");
return;
}
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// If we're here, the image is good - store its data in our
// global...
newLogoData = imageToImageData(img, false);
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// And finally, let's show our preview image...
document.getElementById("previewContainer").style.display = "flex";
renderPreview();
// On to Step 3!
stepThree();
}
}
});
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// And finally, attach an event handler to the overall form - we'll use
// this to get the dithering settings, and as the trigger for updating
// the preview image accordingly...
const imageForm = document.getElementById("imageForm");
imageForm.addEventListener("change", function() {
// Get the state of all of our form controls...
const formData = Object.fromEntries(new FormData(imageForm));
// Set our general globals from the current form state...
//if (formData.ditherEnabled !== undefined) {ditherEnabled = true;}
//else if (imageFormat == "RGB565") {ditherEnabled = false;}
ditherEnabled = formData.ditherEnabled ? true : false;
ditherStrength = formData.ditherStrength? parseFloat(formData.ditherStrength) : ditherStrength;
// Finally render an updated preview...
renderPreview();
});
}
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// This function is where we display our download button, which patches
// the provided bisrv.asd with the updated logo data, and sends it to the
// user's browser as a download...
function stepThree() {
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// As usual, start building up our HTML...
let html = "<section id=\"downloadSection\"><h2>Step 3: Download Updated <code>bisrv.asd</code></h2><p>Click the download button for the updated <code>bisrv.asd</code> file; use it to replace the one in the <code>bios</code> folder on your SF2000's microSD card.</p><div id=\"downloadMessages\"></div>";
// Add our download button...
html += "<div class=\"controlContainer\"><div class=\"control\"><input id=\"userDownload\" type=\"button\" value=\"Download\"></div></div>";
// Close our section...
html += "</section>";
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// Finally, add a <hr> separator after the last step, and append the new
// step...
document.getElementById("steps").insertAdjacentHTML("beforeend", "<hr>");
document.getElementById("steps").insertAdjacentHTML("beforeend", html);
// Let's add the event handler for our Download button...
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
const dButton = document.getElementById("userDownload");
dButton.addEventListener("click", function() {
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// So, we should have the original bisrv.asd data in bisrvData; and we
// should have the offset to the original logo data in logoOffset; and
// finally, we should have our new logo's binary data in newLogoData.
// All we need to do is convert the new logo to RGB565 format (using
// whatever dithering options are set), and then replace the old data
// bytes with the new bytes, re-calculate the CRC32 bytes for the
// modified file and set them in the data, and send the data to the
// user's browser. Easy!
// First, get the RGB565 version of the logo; this is a *bit* messy,
// as we take the user's raw input image, convert it to binary RGB565
// data, then convert it back to an ImageData object so we can upscale
// it with a 2x nearest neighbour upscale, and then convert the result
// *back* to an RGB565 binary array... dumb, but it works!
const convertedImage = imageDataToRgb565(scaleImage(rgb565ToImageData(imageDataToRgb565(newLogoData, ditherEnabled, ditherStrength), 256, 100), 512, 200, "Nearest Neighbour"));
// Now replace the old logo data in the bisrv.asd file...
for (let i = 0; i < convertedImage.length; i++) {
bisrvData[logoOffset + i] = convertedImage[i];
}
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// Next, patch the CRC32 value in the bisrv.asd to account for the new
// logo data...
patchCRC32(bisrvData);
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// And finally, send the updated bisrv.asd data to the user's browser
// as a file download...
downloadToBrowser(bisrvData, "application/octet-stream", "bisrv.asd");
});
}
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
// This utility function renders a preview of the final image...
function renderPreview() {
if (newLogoData.data.length > 4) {
let convertedImageData = imageDataToRgb565(newLogoData, ditherEnabled, ditherStrength);
convertedImageData = rgb565ToImageData(convertedImageData, 256, 100);
const previewCanvas = document.getElementById("processFilePreview");
const previewContext = previewCanvas.getContext("2d");
previewContext.clearRect(0, 0, 256, 100);
previewContext.putImageData(convertedImageData, 0, 0);
}
}
</script>
<hr>
Tool updates... This was just supposed to be a new feature for the generic image tool (dithering support), but while working on that things kind of snowballed 😅 * Added dithering support to the generic image tool and the boot logo changer, when converting images to RGB565 format. This uses a Bayer 8x8 matrix, and the overall "strength" of the dither can be controlled - it defaults to what I feel is a sane value. Dithering can help reduce banding effects due to the low colour depth in RGB565. * Made "fix" scaling mode more flexible in the generic image tool - now there's checkboxes beside the width and height dimensions - if you un-check one, the other dimension will be calculated automatically to keep the input's aspect ratio intact * Improved downscaling quality in the generic image tool. While working on the dithering feature, I discovered the previous "gaussian resampling" downscaling method I was using introduced distortion in certain situations. I had a lot of fun playing around with possible replacements (I tried 10 new downscaling functions!), and finally settled on a hybrid function that mixes powers-of-two downscaling (with some custom mipmap style cross-blending) with Hermite interpolation. This new method is reasonably quick, gives clean results with no great distortion, and works well with alpha channels (doesn't introduce any dark fringing) * Generic image tool now shows the dimensions of the output image * Added a max width on the main tool page bodies, so that they don't get so wide on full-screen desktop browsers * Fixed a few edge-case logic bugs here and there (e.g., when upscaling only a single dimensions with the generic image tool, or places where I thought I was copying objects, but was only creating references to them, etc.) * Switched from using "var" declarations across all tool codebases to "let" or "const" instead * Switched out the SVG alert icons to just use emoji instead * Various other nips and tucks (fixed up my arbitrary 80-column comment wrapping on the tools that had previously just been eye-balled, fixed a few comment typos, etc.)
2024-05-14 21:20:57 +02:00
<p><a rel="license" href="https://creativecommons.org/publicdomain/zero/1.0/">CC0</a>: public domain. Version 1.6, 20240514.1</p>
</body>
</html>