简体中文界面。帖子、指南及政策正文保留原文；未对原文作自动翻译。
MinaChan1120 · 2026年3月6日 09:12 UTC
Have them time the read, file write and display update separately, then exercise controlled delays in each offline. Folder activity is a clue, not enough to choose a repair or a replacement laptop.
14分The first timing run showed long display updates as the chart history grew. File writes were short in that run. Developer is checking the chart and repeating with a deliberately slow writer.
15分Do fresh readings still arrive while the chart is stuck?
9分Yes. The stored readings keep their normal sequence while redraw gets slower. A bounded visible history fixes that in the test version; the full recorded history remains available separately.
13分Then keep the two histories clearly named. Also test a genuinely delayed read and the slow writer, so a responsive chart does not imply that its latest value is fresh or every reading was recorded.
25分Developer tested both delays and the recording limit. Display age and recording-gap indication remain separate. Installed read-only comparison ran through the normal session without the growing redraw delay. We closed the chart issue and kept the timing evidence with the handover.
17分