What are micro jerks?
Let’s briefly recall how each successive frame is created. Since the consistency we are aiming for cannot be achieved in practice, and since it is not possible to render 30 frames in advance as a buffer, the following scenario is most likely to occur in general:
We can see here the clip from one second of Metro 2033 and the work of two Radeon HD 6850s, which seemed suitable because we could move exactly in the window around 30 frames per second, where the micro-stutters, if they occur, are supposedly felt most annoyingly. What we notice first when looking at the diagram is the very uneven length in ms needed to create each frame.
All 30 partial images of the documented second are recorded in the diagram. We see in the sequence mainly the delay of some frames, which are perceived as stalls. It doesn’t help if the next frame makes up for everything in terms of time. You can compare it to a sputtering engine. You can drive with this just as well e.g. 30 km/h, as with a functioning one. Only that one is properly shaken through.
When do you notice micro jerks?
Short and crisp: always. The lower the average frame rate is, the more you perceive the actual average value as a perceived lower frame rate due to the jerks. Thus, a frame rate with 30 fps quickly looks like only one with 20-25 fps with a bit of bad luck. However, the respective differences in the creation times for the frames are still clearly noticeable to a trained eye even at frame rates beyond 60 fps. This is also one of the reasons why we prefer a test with high frame rates in the following GPU scaling tests. It is quite astonishing how the horror can spread even above the often quoted limit of about 40 fps when the rendering gets out of sync.
- 1 - Einführung und Übersicht
- 2 - Wir funktionieren Crossfire und SLI?
- 3 - Mikroruckler - so klein und doch so ärgerlich
- 4 - Stufe 1: CrossfireX mit 2 GPUs
- 5 - Stufe 1: CrossfireX mit 2 GPUs versus SLI
- 6 - Stufe 2: CrossfireX mit 3 GPUs
- 7 - Stufe 3: CrossfireX mit 4 GPUs versus SLI
- 8 - Testsystem und Benchmarkauswahl
- 9 - Benchmarks: 3DMark11 und Sanctuary
- 10 - Benchmarks: DirectX 11 Spiele
- 11 - Benchmarks: DirectX 10 Spiele
- 12 - Benchmarks: DirectX 9 Spiele
- 13 - Leistungsindex und Leistungsaufnahme
- 14 - Zusammenfassung und Fazit
9 Antworten
Kommentar
Lade neue Kommentare
Mitglied
Urgestein
Veteran
Urgestein
Urgestein
Mitglied
Veteran
Urgestein
Urgestein
Alle Kommentare lesen unter igor´sLAB Community →