Improve formatting of frame time values and enhance performance reporting in RTIOW.html

Improve frame time and performance reporting

Add spaces between numeric values and units for clarity and consistency.
Expand detailed performance notes into bullet points for improved readability.

Improve formatting and performance reporting

Adds spaces between numeric values and units for clarity. Expands performance details into bullet lists for enhanced readability and better communication of frame time data.

Author: karimsayedre

RTIOW.html

@@ -103,18 +103,29 @@ <h2>Introduction</h2>
                             <td class="spec-value">Vulkan</td>
                             <td class="spec-value">RTX acceleration</td>
                             <td class="spec-value">Procedural sphere tracing + triangle modes</td>
-                            <td class="spec-value fps-highlight">~33ms</td>
+                            <td class="spec-value fps-highlight">~33 ms</td>
                             <td class="spec-value fps-highlight">~30 FPS</td>
-                            <td class="spec-value">Cornell Box, Lucy, etc. - complex scenes</td>
+                            <td class="spec-value">
+                                <ul>
+                                    <li>No acceleration structure compaction</li>
+                                    <li>Using procedural AABBs per sphere</li>
+                                    <li>Using ray tracing pipeline (no inline ray tracing)</li>
+
+                                </ul>
+                            </td>
                         </tr>
                         <tr>
                             <td class="spec-value">Mine</td>
                             <td class="spec-value">CUDA</td>
                             <td class="spec-value">No hardware RT cores</td>
                             <td class="spec-value">Procedural spheres only</td>
-                            <td class="spec-value fps-highlight">~8ms</td>
+                            <td class="spec-value fps-highlight">~8 ms</td>
                             <td class="spec-value fps-highlight">105 FPS</td>
-                            <td class="spec-value">same resolution and settings</td>
+                            <td class="spec-value">
+                                <ul>
+                                    <li>Same resolution and settings</li>
+                                </ul>
+                            </td>
                         </tr>
                     </tbody>
                 </table>
@@ -907,9 +918,9 @@ <h2>Optimization #6 — Structure of Arrays (SoA)</h2>
                     <tbody>
                         <tr>
                             <td>Frame Time</td>
-                            <td>140ms</td>
-                            <td>65ms</td>
-                            <td class="improvement">-75ms (-53.6%)</td>
+                            <td>140 ms</td>
+                            <td>65 ms</td>
+                            <td class="improvement">-75 ms (-53.6%)</td>
                         </tr>
                         <tr>
                             <td>L1 Cache hit rates</td>
@@ -1236,9 +1247,9 @@ <h3>Global Memory Performance</h3>
                                     </tr>
                                     <tr>
                                         <td>Frame Time</td>
-                                        <td>~10ms</td>
-                                        <td>~8ms</td>
-                                        <td class="improvement">~ -2ms ~(-20%)</td>
+                                        <td>~10 ms</td>
+                                        <td>~8 ms</td>
+                                        <td class="improvement">~ -2 ms ~(-20%)</td>
                                     </tr>
                                 </tbody>
                             </table>
@@ -1497,7 +1508,8 @@ <h3>Case Study: Ray-AABB Intersection</h3>
                     done per
                     frame. Switching from the generic <code>std::fma</code> and <code>std::max</code> to the intrinsic
                     float versions
-                    led to a frame time drop from <strong>12ms</strong> to <strong>9ms</strong>, and reduced instruction
+                    led to a frame time drop from <strong>12 ms</strong> to <strong>9 ms</strong>, and reduced
+                    instruction
                     count.
                 </p>
 
@@ -1550,8 +1562,8 @@ <h3>Performance Breakdown</h3>
                         </tr>
                         <tr>
                             <td>Performance (in hot path)</td>
-                            <td><strong>9ms</strong> total frame time</td>
-                            <td><strong>12ms</strong> total frame time</td>
+                            <td><strong>9 ms</strong> total frame time</td>
+                            <td><strong>12 ms</strong> total frame time</td>
                         </tr>
                     </tbody>
                 </table>
@@ -1592,7 +1604,7 @@ <h3>Best Practices</h3>
                     intrinsics is
                     not a micro-optimization—it's a major win in performance-critical kernels. In our case, it shaved
                     off
-                    <strong>3ms per frame</strong> and greatly simplified the PTX output.
+                    <strong>3 ms per frame</strong> and greatly simplified the PTX output.
                 </p>
             </section>
 

style/style.css

@@ -666,7 +666,7 @@ code {
   border-collapse: collapse;
   width: 100%;
   max-width: 100%;
-  table-layout: fixed;
+  table-layout: auto;
   margin-top: 1.5rem;
   margin-bottom: 1.5rem;
   border: 1px solid rgba(255, 255, 255, 0.1);