Hi, I'm investigating a "quirk" I have noticed related to VIDEOCONLOCK .
I've found a consistent 5-second gap between the display powering off and Modern Standby entry when VIDEOCONLOCK > 5s, and I'm trying to understand what causes it.
Environment: Windows 11 laptop, Intel, Modern Standby
STANDBYIDLE is 0 on AC, so standby is driven entirely by VIDEOCONLOCK.
Results across different VIDEOCONLOCK values
| VIDEOCONLOCK | Display off | Standby | Offset |
|---|---|---|---|
| 1s | 1s | 1s | 0s |
| 2s | 2s | 2s | 0s |
| 3s | 3s | 3s | 0s |
| 4s | 4s | 4s | 0s |
| 5s | 5s | 5s | 0s |
| 6s | 1s | 6s | 5s |
| 7s | 2s | 7s | 5s |
| 10s | 5s | 10s | 5s |
| 20s | 15s | 20s | 5s |
| 60s | 55s | 60s | 5s |
During the offset, pressing a key brings the lock screen back instantly (no standby wake delay), so the system is still fully awake; just the display is off.
Event log findings
- Kernel-Power Event 506 fires at the standby moment, not the display-off moment.
- Kernel-Power Event 566 skips a session state on wake (e.g. 145→147, 142→144), suggesting an intermediate "display off, system active" state that never gets its own 566 event.
- No events from any Windows provider have been found at the display-off moment.
Questions
- Is this 5-second offset hard-coded in the power state machine?
- Is there any way to configure it?
- Has anyone seen documentation on this?
/Michalis