Exploring Apple's 3D Touch | Hacker News Reader

Exploring Apple's 3D Touch

medium.com

91 points_rknLA30 comments

@freino made a web-based 3D-touch demo that prints the value in question, however the javascript DOM attribute has values between 0 and 1, not 0 and 6.666667.

Check it out with an iPhone 6s/6s+ at http://freinbichler.me/apps/3dtouch/

Yep, I'm seeing the same thing -- a DOM attribute with values between 0 and 1. Interesting.

Wondering how the 3d touch changes the cross platform development!?

What happens when a developer has 2 different actions for a long press and pressured (3d touch) press!?

I have trouble understanding the usability pov for the 3d touch! Do we really need it?

Not much in the short to medium terms. Apple are still selling lots of iOS devices that don't have 3D touch hardware support, and will probably continue to do so for several years to come, so even iOS only apps will still need to offer alternative UI options for the lifetimes of those devices.

Apple very strongly advises that actions available via 3D Touch must also be available via some other mechanism. 3D Touch right now is used for shortcuts (e.g. on the home screen), or previewing things (e.g. Peek) where the Commit behavior is the same as what you get when you actually tap on the item instead of 3D Touching it.

This is important because many users, especially the elderly, do not have the physical ability to press hard on the screen.

It doesn't really require that much force. Subjectively, it feels to me like Peek requires less force than pushing the Home button, and Pop requires a bit more, but not significantly more.

Also, under Accessibility, there is a way to control the 3D Touch sensitivity, so it requires less (or more) force to activate.

That said, for people who use alternative input mechanisms (whether it be various forms of assistive touch, or just using VoiceOver), AFAIK they can't 3D Touch (well, you can 3D Touch with VoiceOver, but only if you can see where to push; there's no way to trigger 3D touch on the highlighted element if it's not under your finger).

Liken it to a right-click on a two-button mouse interface. It just ads shortcuts (as stated below). You might cut/copy/paste with right-click in a spreadsheet, but it's also available on the File menu, keyboard shortcuts, etc. Need it? No. But I can see it being incredibly convenient and time-saving in many contexts.

and to think there was an api to derive pressure from contact size back in 2.0

https://developer.apple.com/library/ios/documentation/UIKit/...

anyone knows any app usage of this? I remember some games had pressure calculated from the accelerometer, don't know if any did use this one instead.

GarageBand also uses accelerometer data to try get the velocity of taps on drums it works reasonably well, not like 3D Touch of course.

I would love to see how closely/consistently the output from "contact size" increases with "force"

I don't think is gonna be consistent among different fingers/people, sadly. but synaptic did pressure sensitivity on contact area long ago, wouldn't be surprised if they got some nice normalization heuristic already going.

edit this is funny coincidence: the dev guide for synaptic says 255 level but better badn in 7 levels at most

(2005) http://arstechnica.com/civis/viewtopic.php?f=9&t=352584

Edit: I know those are not really comparable teches, force touch for one is transparent

This sort of thing really depends on the application, and the nice feature provided with the new "force" property is that you can actually distinguish between a "whole finger pad" light touch, and a "thin, firm, almost just the finger nail" hard touch, and most things in betwee.

That was a private API -- not sure what Apple used it for, since GarageBand was using accelerometer from what I could tell.

In iOS 8 it became a public API. But the problem is that this didn't actually directly measure pressure, it just measured touch radius. Pressure affects touch radius, but so do a lot of other things (finger size, what part of the finger is used, which finger is used, etc) and so it's not particularly effective at extracting pressure.

I'm wondering: can it measure force also for different presses simultaneously? And how far apart do the presses need to be?

In my experience so far, yes, it will provide unique force values per-touch, though I'm not sure if the quality deteriorates, or at what point it might deteriorate.

I don't know how far apart the presses need to be, as I haven't done a "two fingers as close as possible" test yet.

I really would like to see google maps to make use of 3d touch to make zooming easier. Moving the finger up/down while pressing would zoom in/out. That would make one-finger usage of the maps so easy.

On my Android phone, at least, this can already be achieved with a double-tap of sorts.

Yes, here too. But pressing might be more intuitive then double tapping.

You can zoom in and out with one finger on Google Maps by tapping and holding (1.5 taps?) the map and then sliding your finger up and down.

You may already know this but you can double-click, hold and slide up/down on google maps to zoom in and out.

I have the 3D thing on my macbook trackpad (force touch) and I have it turned off. Unfortunately the click on the trackpad is triggered by the same mechanism, which means detection of the click is effectively handled by software.

The net result is that, unless you make distinct actions when moving between a zoom and a click, the OS can't tell when the scroll ends and the click begins. I work in CAD software where it's really common to zoom out and roll straight into the click and drag (pan) but it's an absolute nightmare on this trackpad.

The laptop I have right now has a similar pressure sensitive Synaptics ForcePad, and I've never used a more annoying trackpad before. Just simply moving the cursor on the trackpad can be ridiculously error prone, with random clicks triggering whenever you let your guard down and push a little too hard.

The worst part is, I can't even use it as a regular trackpad with tap-to-click only, because the pressure sensitive clicks can't even be completely disabled.

google maps used to have the +/- icons in the corner of a screen.. I found this to be very useful for single hand operation of maps... double tap + sliding doesn't work too well because your slide action is constrained to the range of your finger movement, and typically results in a small amount of zoom.

I'd like to know how precisely the force sensors are calibrated during the manufacturing process, i.e. what is the variation in maximum detected pressure across iPhones?

For more on 3dTouch, I've catalogued a dozen other articles here:

http://www.h4labs.com/dev/ios/swift.html?age=10000&q=3dTouch

By the way, over the weekend I added a Swift Weekly view, which should be a nice way to keep up on new Swift blogs:

http://www.h4labs.com/dev/ios/swift.html?week=0

How does pressure sensitive touch essentially differ from a long touch?

Well... there's pressure

Er, you press harder? A long touch is already a recognised gesture(1).

(1) https://developer.apple.com/library/prerelease/ios/documenta...

A long press is a function of time: "How long has the user pressed?"

3D Touch is a function of pressure: "How hard has the user pressed?"

In practice, things that your application previously accomplished with long presses may be more suitably accomplished with 3D Touch — things like contextual menus or shortcuts come to mind. Additionally, iOS has new APIs for presenting content using 3D Touch. The article briefly references them.

Reply on news.ycombinator.com