-
Is there any startup speed difference among
In practice, it seems no difference, all takes 1s, measured by eye. But theoretically, shouldnt there be a speed difference between "normcap is already running" vs "normcap is not already running (cold start)"? ( Enhancement Suggestion - Add Customizable Hotkeys (Keyboard Shortcuts) · Issue #590 · dynobo/normcap normcap/adr/004-do-not-implement-hotkey.md at main · dynobo/normcap (Im currently using terminal + AHK to run Normcap.) |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
TLDR; Capturing with NormCap from tray icon is faster compared to a fresh start, BUT the difference is so small, it barely matters in practice. NormCap was always designed to start fast to not require running it as daemon in tray. NormCap barely needs to load state (mainly the application settings), so there is little to win by running in tray. But, as every Python application, there is always a startup overhead for loading & running the Python interpreter and importing all modules. How much time this takes depends on your system, last time I measured it on my machine, it was like 300ms. The processing time for text recognition also depends on the system, but mostly on the selected region's pixel size. On my system, even for a really small region, it takes 1-2 seconds. That means, the startup delay is always quite small compared to the rest of the processing time. For most use cases, I would recommend to not use the tray. Possible reasons for still using the tray might be:
Does this answer your questions? 🙂 |
Beta Was this translation helpful? Give feedback.
TLDR; Capturing with NormCap from tray icon is faster compared to a fresh start, BUT the difference is so small, it barely matters in practice.
NormCap was always designed to start fast to not require running it as daemon in tray. NormCap barely needs to load state (mainly the application settings), so there is little to win by running in tray. But, as every Python application, there is always a startup overhead for loading & running the Python interpreter and importing all modules. How much time this takes depends on your system, last time I measured it on my machine, it was like 300ms.
The processing time for text recognition also depends on the system, but mostly on the selected region…