Home to date time #600
Home to date time #600
Conversation
| ImageFloatBox system_settings(0.685, 0.69, 0.05, 0.03); | ||
| ImageFloatBox other_setting1(0.615, 0.69, 0.05, 0.03); | ||
| ImageFloatBox other_setting2(0.545, 0.69, 0.05, 0.03); | ||
| if (!is_setting_selected(stream, context, system_settings, other_setting1, other_setting2)){ |
There was a problem hiding this comment.
This sounds like it might slow it down quite a bit? An alternative that I'm thinking about is to keep the A press to enter the settings, then run run_until() with the down-scroll mash until you detect that we're at the bottom of the settings menu.
| ssf_press_button(context, BUTTON_A, 3); | ||
|
|
||
| context.wait_for_all_requests(); | ||
| context.wait_for(Milliseconds(250)); |
There was a problem hiding this comment.
For some reason, wait_for_all_requests() alone doesn't actually wait for all the buttons to finish pressing. This is why I have the extra 250ms wait here. Any ideas on why this might be the case?
There was a problem hiding this comment.
That's not supposed to happen. Needing a delay before inference is rather common since the video + Switch's UI delay is longer than the time to process return trip for a finished command.
Is the following wait_until() triggering early without this wait?
There was a problem hiding this comment.
The program still works as intended without the wait.
But I can see that the wait_until inference is starting a bit earlier than I want. I want the button presses to be completely done before we start inference, to reduce the probability of early false positives. However, this is a theoretical concern, since I haven't seen any issues in my testing, even without the 250ms wait.
2 participants
Using inference for better reliability.
Right now, I created a second version of
home_to_date_time, which includes VideoStream. If you think it looks good, we can delete the oldhome_to_date_timeand update all the references.'