5 ms·
I just played around with this in the emulator for a bit and I discovered that the feature has some subtle nuances. Clicking on the same pixel can produce eithe
by BrianHV 14y ago
I just played around with this in the emulator for a bit and I discovered that the feature has some subtle nuances. Clicking on the same pixel can produce either the visible letter or the invisible letter. From what I've been able to ascertain, within a certain range of the visible key it favors the visible key if you're typing the first letter of a word, and favors the invisible key otherwise.
- maxmcd 14y agoDoesn't the the touch area for individual keys change size depending on what is being typed? The behavior you are describing should be the same on almost any key.
- BrianHV 14y agoIndeed, you are correct. And the heuristic clearly isn't as simple as whether or not the cursor is at the start of the word. Clicking between O and P produces a seemingly pseudorandom alternation that probably has some basis in statistics.
- kevinchen 14y agoI remember that the iOS 1 videos Apple made to introduce the soft keyboard mentioned that the size of the keys change based on the actual word that you're typing. (It's very unlikely for a user to want to type an S after typing P-I-Z-Z)
- adpowers 14y agoYeah, I used to demonstrate this my typing "Americ" and then touching down directly over the 's'. Even though I pressed the 's', it was the 'a' that appeared. You could clearly see I wasn't even touching 'a'. I think they got rid of this feature a few versions ago because I haven't been able to reproduce it. I suspect (but have no evidence for) that this is related to the typing lag that was really bad in ios 2 or 3. From my perspective, it seems they weren't able to fix the realtime correction so they made it asynchronous and just fix up words after you hit space. This is how it seems to perform now (and let's them offer more features like multi word correction).
- kevinchen 14y agoI just tested that, and it seems like you're correct. I recall that simply typing stuff quickly could increase CPU usage by ~30%.