iPhone touch screen trumps Droid, Nexus One, Eris?

[Cleverboy]

http://labs.moto.com/diy-touchscreen-analysis/

Just read this from Electronista:

MOTO Development Group in a study published Saturday has given the iPhone top marks for the quality of its touchscreen. Comparing it against multiple Android phones -- including the Nexus One, Droid and Droid Eris -- the hardware consultants found Apple's screen is the most accurate and lets users draw smooth, straight lines regardless of how quickly the finger moves or how much of it is in contact with the display. The Droid Eris and Nexus One were next with some erratic behavior with partial contact, while the Droid produced a similar effect no matter what contact has been made.

While all four phones have capacitive touchscreens, the iPhone's victory is credited to it having a small sensor size, a high sampling rate and intelligent-enough processing to determine where the finger is actually moving. The Droid by contrast may have a too-coarse sensor that can incorrectly guess the intended input and produce "stair steps."

Touchscreen performance hasn't been a major complaint for the Android phones in the test, but MOTO suggests that quality should be a greater focus and that companies making touch phones should devote more resources to them and take much of the hardware and software design in-house and as early as possible.

http://www.electronista.com/articles/10/01/09/iphone.more.accurate.than.droid.nexus.one/

There are a number of painting programs on my iPhone, as well as any number of programs that rely on touch precision (Eliminator Pro for one). When the iPhone first came out in 2007, I'd picked up "Sketches" (I'd previously been a fan of the jailbroken version). Years later, I've happily purchased Brushes and Sketch Mobile and they've both been fantastic programs I look forward to using more.

Looking at these simple examples of poor-precision from these Android programs is the nightmare scenario I'd always been concerned about with the iPhone... that when it was finally tested, the iPhone's touchscreen would show that its version of "touch" sensitivity wouldn't be useable for painting programs. Instead, it sounds like the rush to release advanced hardware has made iPhone's competitors release products with the deficiencies I thought iPhone would have had (but never did).

~ CB

 

[kdarling]

The biggest interference maker for a capacitive touchscreen is the LCD behind it.

Size, resolution, materials, distance... all play a part.

Having 800x480 active transistor pixels is not a big help for either the Droid or Nexus.

They need to do some electronic/hardware tweaking and/or add software smoothing.

 

[ToroidalZeus]

The biggest interference maker for a capacitive touchscreen is the LCD behind it.

Size, resolution, materials, distance... all play a part.

Having 800x480 active transistor pixels is not a big help for either the Droid or Nexus.

They need to do some electronic/hardware tweaking and/or add software smoothing.

Uhm NO

Resolution is nothing if the screens ntsc accuracy is off or the size is thinner like the droids or the touchscreen technology is inferior like all those phones.

 

[OneWhoTweets]

Nope

Although the screens on the Droid and Nexus One have greater resolutions, the iTouch and iPhone have the best response to touch. They also have a great feeling which none of the android phones to date can match (I have used all the U.S. versions).

 

[Cleverboy]

They need to do some electronic/hardware tweaking and/or add software smoothing.

That's the interesting thing. These SIMPLE tests, really show that they could clearly work on their "smoothing" and not be exposed so embarassingly for poor quality results. I don't think smoothing however would address a lower sampling rate (which might be hardware dependent, accuracy perhaps lost in the attempt to lower component cost or lower battery consumption?)

It would be encouraging to think they could update their drivers and resolve the problems.

~ CB

 

[iphones4evry1]

I have a draw app on my iPhone 3G and I can draw perfectly straight lines (near perfect, not absolute perfect, of course). The lines look perfectly straight to the eye. App is Sketchbook.

In my opinion, it's dam good!

 

[ntrigue]

I have a draw app on my iPhone 3G and I can draw perfectly straight lines (near perfect, not absolute perfect, of course). The lines look perfectly straight to the eye. App is Sketchbook.

In my opinion, it's dam good!

Mind you, it's a slow drawing rate to ensure all data points are contacted.

I can rapidly draw straight lines.

 

[gloss]

This is not at all a scientific test.

 

[kdarling]

This is not at all a scientific test.

That's the problem with press releases like this one... they're intended to garner publicity for that company or group, more than anything else.

The testers themselves say they have more scientific methods they use, but those aren't what they released. They also said that most tests showed little problem, but the worst photos are what are posted.

There are things to consider:

1) The iPhone has the least PPI, and so cannot show as much stair stepping.

2) We don't know the resolution of the touchscreens on the other devices. If they actually have 850x480 touchable points vs the iPhone's 480x320, that's a factor because the higher touch ppi means it's MORE sensitive to a stroke's wobbles.

3) The silliness about this affecting the keyboard is beyond ridiculous, when you think about the different size of your finger pads, plus the Droid has a hardware keyboard anyway.

4) Way too much brouhaha over this, since very, very few apps would be affected by such a little waviness. After all, the whole point of touch UIs is that they accept gross sized touches. If you want to worry about accuracy, get a stylus driven system.

 

[LIVEFRMNYC]

It's not that silly of a test when you think of all the nit picking that's done to the iPhone.

 

[Cleverboy]

The iPhone has the least PPI, and so cannot show as much stair stepping.

Interesting. So you're dismissing the assertion that the iPhone has a higher sampling rate for touch input, and simple going with the fact that the screen resolution is higher on the other devices.

We don't know the resolution of the touchscreens on the other devices. If they actually have 850x480 touchable points vs the iPhone's 480x320, that's a factor because the higher touch ppi means it's MORE sensitive to a stroke's wobbles.

You know... in the amount of times each pixel is polled is LOWER on the Android devices, because it has more pixels per area... it would make a lot of sense, and I'd truly expect the quality of the touchscreen accuracy to be degraded.

I mean... I see what you're thinking, but considering the VISUAL DISPLAY and the CAPACITIVE TOUCH layer are two separate components (last I read anyway), I'd think one would not conflate the two things.

~ CB

 

[gloss]

A major issue here is that the quality of the software used to perform the test is not guaranteed to be identical on all platforms. More importantly, the finger gestures themselves, by nature, will differ between tests, as they are done 'live' by an individual.

This test could only be done properly using identical programs which boast identically accurate results vis a vis the touch screen's input and a mechanism able to produce the exact same inputs on each of the four screens, both in terms of pressure and track.

In other words, the whole thing is sort of bollocks.

 

[Cleverboy]

A major issue here is that the quality of the software used to perform the test is not guaranteed to be identical on all platforms. More importantly, the finger gestures themselves, by nature, will differ between tests, as they are done 'live' by an individual.

Well, not the "quality of the software". Let's assume that the "quality" is the same and that the software is custom on each device. I think its more of a concern that the finger-swipe is not mechanically driven using Pogo styluses as necessary. Disclosing the methodology of the software routines would help, but assuming them standard, wouldn't be too hard given the basic nature of the function.

This test could only be done properly using identical programs which boast identically accurate results vis a vis the touch screen's input and a mechanism able to produce the exact same inputs on each of the four screens, both in terms of pressure and track.

Maybe you're thinking of "benchmarks" where the software could run by itself. Unfortunately, touchscreens are real-world mechanisms that need to be interacted with. I think they main shortfall is in not drafting a mechanical test. I can't see how it would have been hard either.

In other words, the whole thing is sort of bollocks.

I think its useful and even relevant, just not strictly scientific OR transparent. I'd be willing to bet however, that the results are indicative of real disparities across the displays however.

~ CB