Since the beginning of time, man has been asking many seemingly unanswerable questions. What is vertical synchronization (V-Sync) and how does it work? What exactly does temporal anti-alias (TAA) really mean? How does triple buffering (TB) operate? How are all of these mysteries connected? Like the crop circles that once confounded, observances of such terms brought conjecture and reasoning…but also some uncertainty. But also like the crop circles, there is a purely logical solution behind them all; these solutions are merely obscured by misdirection.
Sure, that seems a little bit dramatic and perhaps a tad exaggerated. But yet I see these terms popping up on hardware forums everywhere, often questioned and just as often answered. But sometimes the answers are incorrect, or they’re correct but with the wrong foundation…or they’re perfectly correct but without intricate detail or validation. And of course, the terms aren’t always reasonably linked together. So my intention here is to offer a defining presentation that hits all of the basics in one swoop, covering the most obvious ramifications.
These things may seem pretty simple to those who understand them, but they quickly roll into complication for those that are just learning how to tweak and use their video card. All of these choices have side-effects and it is important not only to realize what they are, but how and why they exist. And there are peripheral questions, as well. Does v-sync really decrease performance significantly? Why does tearing seem less noticeable on LCDs? Does triple buffering increase performance, and does it function without v-sync?
So the first priority is to define all these terms, then look at how they operate and what changes they entail. After that there will be an analysis of performance using all of the combinations. And, throughout this, perhaps we’ll take a look at the more subjective issues.
_______________
Read More/Source:
Team Radeon
Guide to Vertical Synchronization, Triple Buffering, and Temporal Anti-Alias
