Passt nicht? Macht nichts! Sie können Artikel bis zu 30 Tage zurückgeben
Mit einem Geschenkgutschein können Sie nichts falsch machen. Der Beschenkte kann sich im Tausch gegen einen Geschenkgutschein etwas aus unserem Sortiment aussuchen.
Bis zu 30 Tage Rückgaberecht
A single main memory access costs roughly three hundred cycles. An integer addition costs one. That means one cache miss is worth more than a thousand arithmetic operations on the same core, and it explains why two loops with identical instruction counts can differ in cost by a factor of ten. Most optimization work fails before it begins because the programmer is counting instructions while the processor is waiting on memory.
If that number interests you, this is the book for you. If you would rather know why your frame time spikes at ninety seconds instead of why your renderer is slow, keep reading.
You already know the pain. The profiler says the loop is expensive and the loop looks innocent. The frame is fine on your workstation and collapses on the minimum specification device. A regression appears three weeks after the change that caused it, and nobody can find the change. Your team optimizes the renderer for a month before discovering the bottleneck was animation, or an allocation fallback, or a debug overlay left enabled in the shipping configuration.
This book replaces guesses with evidence. You will learn to build a frame timing harness, profile before touching code, classify a loop as compute-bound, bandwidth-bound, or latency-bound, and apply the remedy that matches the regime instead of the one that feels satisfying.
What sets it apart is that every claim is measurable and every chapter ends with something you can run. This is not a survey of techniques. It is a working method.
Inside you will find:
Downloadable source code for every tutorial.
Step-by-step tutorials that build a complete toolkit across the book.
Practical exercises that apply each technique to your own project.
Real-world programming examples drawn from shipping titles.
Troubleshooting sections for the moments when a technique does not behave as described.
End-of-chapter projects that produce measurements, not just understanding.
Stop guessing where the milliseconds go. Start measuring them.
Hallo! Ich bin Libroamiko, dein Buchberater.
Wie kann ich dir helfen?