backtop


Print 146 comment(s) - last by asrey.. on Sep 29 at 12:05 PM

Benchmark reveals why the iPhone's touch gestures feel so fluid, and why Android may feel laggy to iOS users

App maker Agawi has developed a new benchmark, which it hopes will challenge phonemakers and cast light on the performance of their touchscreens.  Dubbed TouchMarks, the benchmark targets devices running Apple, Inc.'s (AAPL) iOS, Google Inc.'s (GOOG), and Microsoft Corp.'s (MSFT) Windows Phone -- the three most used smartphone platforms on the market.

The results of Agawi's preliminary tests are fascinating -- and pretty embarassing for Windows Phone and Android.

In order to measure latencies typically in the millisecond range, Agawi has come up with a set of probing hardware dubbed "TouchScope" and supporting software, all of which it hopes to open source.  

The TouchScope utilizes probes hooked up to an Arduino microcontroller and is capable of measuring Minimum App Response Time (MART) -- i.e. the fastest response you can expected from your smartphone touchscreen.  Agawi warns that actual responses in-app may be slower as the touch processing has to contend with other resources, e.g. graphical rendering or streaming data I/O, where as the TouchMarks app is a bare bones native app.

Touch Scope
The TouchScope is powered by an Arduino microcontroller. [Image Source: Agawi]

In the tests the iPhone 4 -- a three-year old device -- beats every single Android and Windows Phone tested in response time.  Nokia Oyj.'s (HEX:NOK1V) Lumia 928 managed 117 milliseconds (ms), while the fastest Android -- Samsung Electronics Comp., Ltd.'s (KSC:005930) Galaxy S4 -- managed a barely better 114 ms.

Meanwhile the iPhone 4 scored an 85 millisecond response time.  And it's clear Apple isn't resting on its laurels -- the iPhone 5 achieved an even more formidable 55 ms.  Apple was the only OEM to achieve under 100 ms.

TouchMarks
No Android or WinPhone was capable of iPhone-like response times. [Image Source: Agawi]

Fluidity of the interface is often cited as a selling point of iOS/the iPhone.  It's pretty sad to see all Androids/Windows Phones getting beat by a three-year old Apple device.  Agawi raises an interesting point, speculating:

Another possibility is that while the Android and WP8 code are running on runtimes (Dalvik and CLR respectively), the iPhone code is written in closer-to-the-metal Objective-C, which may reduce some latency. In future TouchMarks, we’ll compare C/C++-based Android apps to Java based apps to determine if this is the case.

If so, Google and Microsoft might be wise to incorporate features into their development environments to require native compilation of touch-related secitons of code.

Sources: Agawi, VentureBeat



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Good to know but not practical
By sevin7 on 9/23/2013 2:04:01 PM , Rating: 4
Its good to see that someone made a benchmark, and that iOS is easily 50ms faster depending on the version. However, I really doubt that such a small time is of any significance.




RE: Good to know but not practical
By coondini on 9/23/2013 2:23:30 PM , Rating: 4
It just gives all the Tony Swashes of the world another reason to get all giddy and wet themselves over the love of their "superior" devices.

Tony Swash in 3...2...oh wait, he's already here. That didn't take long...


RE: Good to know but not practical
By Argon18 on 9/24/2013 5:55:44 PM , Rating: 2
you microsofties are all sour grapes. you whine when an apple product performs worse than a microsoft one, and you whine again when it beats it. you guys must have a steve jobs action figure in your rectum or something.

i don't use a single apple product, therefore i don't give two shits whether they perform better or not. same goes for microsoft products, don't use them, don't care about them. but you don't see me moaning and whining like a little girl over every headline. good grief.


RE: Good to know but not practical
By geekman1024 on 9/24/2013 9:21:37 PM , Rating: 1
I didn't see any Microsoft products involved in the benchmark, so why oh why?


RE: Good to know but not practical
By jonm78 on 9/24/13, Rating: 0
By overlandpark4me on 9/26/2013 2:06:02 PM , Rating: 2
Running a Nexus 4 Android and my kid Apple 5. We can open the same app, and the Apple opens it, and loads it quicker, so I can believe this story. His does seem to run smoother. Hopefully they come out with something like that in the future.


RE: Good to know but not practical
By 91TTZ on 9/23/2013 4:25:44 PM , Rating: 4
I have an iPhone 4S and I'm always disappointed when I try out Android phones. They're noticeably slower.

People can pick out that kind of delay pretty easily.


RE: Good to know but not practical
By dgingerich on 9/23/2013 5:16:58 PM , Rating: 5
I had an iPhone 4S forced on me by my employer two weeks ago. I'll admit that the touch response is much faster on the 4S than my Android phone, an HTC OneS, and I get better signal strength on the iPhone than any other phone I've had, but I hate the interface. (I also happen to hate Apple as a company for the elitist and hypocritical way they do business, but that's another matter.) iOS7 has improved things somewhat, by copying the best parts of Android, but it is still awkward to use after using Android for so long.


RE: Good to know but not practical
By Keeir on 9/23/2013 8:41:29 PM , Rating: 1
So... with under 1 week using iOS7 you find that you don't know it as well as Android that you have been using for how long?


RE: Good to know but not practical
By Rukkian on 9/24/2013 9:34:16 AM , Rating: 2
That is the thing, if you ask any apple zealot (I have many in my family), it is so easy, that there is no learning curve, and everybody can just use it.

If a person that can use the supposed difficult, clunky Android is still lost at times after a week, then maybe there is not as much difference as many apple zealots say.

My wife got a new iphone (5c) because she is told Android is too difficult, but then needs me to come show her how to use things.

I don't get the fanboyism on either side. I see benefits to each, and hope both get better to push each other.


By Plazmid19 on 9/26/2013 5:48:53 PM , Rating: 2
It's good to see that there are open-minded people on this thread. I happen to enjoy both Android and Apple, but I've not used a Windows Phone yet, or the Apple 5. There are merits to all of these. I'm glad we have a choice.


RE: Good to know but not practical
By Reclaimer77 on 9/24/2013 8:59:12 PM , Rating: 3
Oh we know iOS7, we've been using it for years...

http://cdn-static.zdnet.com/i/r/story/70/00/016705...


By Cheesew1z69 on 9/25/2013 2:32:06 PM , Rating: 2
Wow...LOL


RE: Good to know but not practical
By KOOLTIME on 9/27/2013 12:52:23 PM , Rating: 2

Sure everyone wants faster, same as cars, but also the fastest cars aka veyron costs few million, most people dont have that cash for it.

So its a mute argument, always. if there is only 1 product brand type in a classification, then stuck with just that, but when many types around, its always cost over performance choices.


RE: Good to know but not practical
By bug77 on 9/24/2013 4:16:24 AM , Rating: 3
At 60fps, a frame takes 16.66ms. 50ms (the difference between iphone and SGS4) is about 3 frames. Namely, the iphone lags about 3 frames, the SGS4 lags 6.


RE: Good to know but not practical
By someguy123 on 9/24/2013 7:51:16 PM , Rating: 2
I honestly don't see how it isn't practical. That's a pretty huge latency difference, especially if you consider that software itself adds a bit of latency. At around 180ms you start to be able to feel and see the delay, which wouldn't be too unrealistic with a 120ms screen and an app that has a 2-3 frame buffer. Ultimately you want every aspect of the phone to be as low latency as possible since each aspect will introduce more latency.

Apple's phones tend to have pretty good hardware, even if I dislike their software model.


By robinthakur on 9/25/2013 9:58:22 AM , Rating: 2
This is interesting because I have always found touch response to be somehow lacking and laggy on Android Phones I've owned, as opposed to iPhones and just assumed it was a difference in the software or that it improved a great deal in the IP5 with the bonded screen/touch sensor. Could never quite put my finger on why they felt less responsive. Have never used a Windows phone since version 5 so can't comment there. Glad to know I'm not going crazy...


RE: Good to know but not practical
By asrey on 9/29/2013 12:05:50 PM , Rating: 2
All the apple fanbois will lose their money to iTunez 2x as fast! XD


By coburn_c on 9/23/2013 4:08:11 PM , Rating: 1
"The company’s team of engineers test responsiveness so they can measure just how acceptable the cloud-based content is to users. The team built a device dubbed Touchscope that can measure response times to a level of accuracy that is plus or minus 4 milliseconds. It then adds the cloud processing response time to calculate the actual delays experienced by users."

no mention of how the device measures, and a vague reference to cloud processing times

the number of variables this throws into the mix.. how is this even considered a touchscreen measurement if the include 'cloud processing response time'




By 91TTZ on 9/23/2013 4:27:12 PM , Rating: 2
What the hell is "cloud processing response time"?


By inighthawki on 9/23/2013 4:30:39 PM , Rating: 2
I suspect they're trying to do calculation of responsiveness for services such as OnLive to measure the latency between when the user presses a controller and how long it takes to see the change happen on screen after the computation is done on the cloud.


Pure Motion HD+
By CyCl0n3 on 9/25/2013 10:15:10 AM , Rating: 3
Im a bit confused because the latest i read to display response time was this when the Lumia 920 got introduced:

quote:
Not that we've noticed much, if any, motion blur on our handsets while scrolling or watching movies, but Nokia claims your average mobile display pixel response rate is 23ms, while a rendered frame lasts 16.7ms, i.e. the pixels don't have time to go with the flow, resulting in blur and artifacts with fast moving objects on the screen, for example.

The way Nokia is overcoming this is applying temporary high voltage to speed up the individual pixels, essentially putting the whole panel in overdrive mode. Lab results show pixel transition taking sub-9ms this way, i.e. plenty of time to do one frame and get ready for another without any lag and blur. This way smooth 60fps rendering is achieved, Nokia says, regardless of how speedy the action is on the Lumia 920 display. Good stuff, but we have to again check with a final unit how much of a difference this makes in reality.


Here the Link to the article:

http://www.phonearena.com/news/The-Nokia-Lumia-920...

And a video to the Technology:

http://www.youtube.com/watch?v=lN-NaHcOeII

And so far i´ve seen and felt, the Lumia series have a way faster display then the Galaxy for example.
So im not sure if the Benchmark done here is quite right.




RE: Pure Motion HD+
By arthur449 on 9/27/2013 4:09:02 PM , Rating: 2
Your quote is referring to the measured delay between a pixel being given a command to change and it changing.

The subject of the article is the measured delay between the touch sensor and an on-screen response.


Wonder how Intel phones compare
By Khato on 9/23/2013 2:10:07 PM , Rating: 2
I seem to recall an article talking about Intel doing similar investigations and tweaking the x86 version of Android in order to decrease the gap. Would be interesting to see what the Razr i results are, or Lenovo K900.




RE: Wonder how Intel phones compare
By Mint on 9/24/2013 10:04:47 AM , Rating: 2
Very interesting. I googled some of those articles:
http://www.tomshardware.com/reviews/surface-window...

http://www.tomshardware.com/reviews/galaxy-tab-and...

However, those articles are measuring keyboard lag, i.e. the time it takes for a pressed key to show a letter typed in the textbox. The OS is doing autocorrect calcs there as well.

If you look at Tom's iPad videos, you can count 70-80 frames (at 1000 fps) from the touch to the time the key changes color. That's similar to the results of this article for the iPhone 4.


Small screen + lower res = faster response time
By ack on 9/23/13, Rating: 0
By TakinYourPoints on 9/23/2013 5:43:36 PM , Rating: 3
You get similar results from a 10" iPad display compared to something like a Galaxy Tab or Nexus 7/10. Display size isn't the issue, the operating system is.


By kenyee on 9/23/2013 5:17:42 PM , Rating: 2
I remember reading the newer screens w/ the touch sensor embedded in the display as opposed to being another layer is faster.

And I'm not sure I believe those numbers. When I last played w/ a Lumia, it felt smoother than my Galaxy S3 when scrolling. According to that chart, it's the same. The BB Z10 feels about the same as the S3 though, so there goes the C++ is better theory... :-)




That is fine but they forgot that
By phatboye on 9/23/2013 11:24:47 PM , Rating: 2
Android may be 55ms slower in response time but iOS is 4 times as sucky.




So where's the Nexus devices?
By douchefree on 9/24/2013 2:39:15 AM , Rating: 2
Also, tomshardware used to perform these tests with a high speed camera. Their findings were that android was a bit faster (something in the neighborhood of 30ms faster). My experience is that ios is more responsive than any other device.
The news here is that the windows devices are no better than android in this area.
Also, so much for the musicians who can perceive 8ms differences but seem to think ios delivers such fast results.
Lastly, android has needed to focus on this problem for a long time rather than the audio issue. Halving their response time would make their relatively worse audio latency a non-issue.




In-cell
By Omega215D on 9/24/2013 5:53:06 AM , Rating: 2
I'm guessing this has to do with the new In-Cell tech (made by LG) which uses one less layer.

I've been getting mixed responses about my LG Spectrum 2 having this technology but it pretty much on par with my brother's iPhone 5S (he traded in his 5) in real world usage. The interface also has no lag unlike those Android phones of yesteryear ( though my Thunderbolt and Rezound were virtually lag free when they were updated).

Hopefully more phones will feature more efficient touch screens (despite being quite imperceptible).




I like...
By ReloadAO on 9/24/2013 8:58:32 AM , Rating: 2
I like turtles, not apples!




They're holding it wrong
By superstition on 9/23/13, Rating: 0
RE: They're holding it wrong
By superstition on 9/23/13, Rating: 0
Humans Aren't Computers
By Reclaimer77 on 9/23/13, Rating: -1
RE: Humans Aren't Computers
By Cloudie on 9/23/2013 2:09:22 PM , Rating: 5
You really can. My old iPhone 4 felt much smoother and more responsive than my Galaxy Nexus. It also registered key taps more accurately. Honestly, if it wasn't for Swiftkey this galaxy nexus would have gone in the bin.

The Nexus 5 can't come out soon enough. And Android touchscreen/UI responsiveness/accuracy cannot be improved soon enough either.


RE: Humans Aren't Computers
By Souka on 9/24/2013 11:29:11 AM , Rating: 1
So low touch response = fast phone?

Hmmm... My company iPhone4 (my first i-device) has always felt slow compared to my Droid X2, S3 (current), or wife's S4. My friend's Nokia Lumina (maybe 10months old) is wayyy faster opening apps than my iPhone.

To me this is like the start-up time for a Windows PC. Faster startup doesn't equate to a fast computer. It's just one factor.

Maybe they need to also benchmark opening and starting Youtube app? I know that "flagship" Android devices will blow past the iPhone4, or even older Android devices...

iPhone5, which my brother has, is pretty slick however... it feels responsive like my/wife's galaxy phones.

My $.02 of ramblings...


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 2:09:52 PM , Rating: 5
Not true. You should watch this video of some Microsoft researchers demoing the latency between various latencies of input:
http://www.tomshardware.com/news/microsoft-touch-s...

The latency when using touch is more significant because there is a physical relationship to the brain between moving your finger and moving the thing on the screen. The disconnect between something like a mouse breaks this connection so the brain has a more difficult time actually seeing the latency.


RE: Humans Aren't Computers
By vortmax2 on 9/24/2013 12:33:56 PM , Rating: 2
That's a great link. It really shows how much improvement there is to be made. Unfortunately, the article didn't mention any technical reasons as to why there's so much latency now. Is it the raw processing power, the actual screen tech, software or all 3? I'm guessing both since Microsoft demos a monochrome-type screen.

Thoughts?


RE: Humans Aren't Computers
By inighthawki on 9/24/2013 12:59:34 PM , Rating: 2
The article did make a claim of:
quote:
but Dietz stated that it was only a research object and a device which Microsoft would like to see becoming available over the next decade


so I would suspect it is primarily a hardware related issue on the touchscreen itself. Although processing power and software will definitely come into play in how long it takes the CPU and OS to handle the interrupts.


RE: Humans Aren't Computers
By zephyrprime on 9/24/2013 1:55:17 PM , Rating: 2
I really doubt it. I wouldn't be surprised if the touch sensor itself was made by the same manufacturer in some cases for some android phones. After all, we know the screen for the iphone used to be made by samsung and of course samsung makes the screens for their own phones. It's probably the inefficiency of the software stack between the ui inputs back out to the gui draw capabilities.


RE: Humans Aren't Computers
By axdirkadirka on 9/23/2013 2:10:13 PM , Rating: 5
What? 50 ms is actually a huge amount of time, and can definitely be felt. For anyone who plays video games competitively, even 10 ms can be felt and can make a massive difference in play. Granted it doesn't make as much of a difference on a phone, but the difference would still be noticeable.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/13, Rating: -1
RE: Humans Aren't Computers
By axdirkadirka on 9/23/2013 2:37:13 PM , Rating: 2
In this context, 50 ms IS a huge amount of time. Also, reaction time is far different than "able-to-tell-a-difference" time. I can tell you for a fact that a jump in ping from 50ms to 100ms changes some applications (FPS games, for example) from usable to unusable. Your example of the 215ms median response time is hilarious...imagine if tried to watch a movie shot at less than 5 frames per second. Touch/visual perception takes far less time than you think it does.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 2:44:17 PM , Rating: 2
quote:
I can tell you for a fact that a jump in ping from 50ms to 100ms changes some applications


SIIIGH...

That's because of how TCPIP works. There are packets and acknowledgement packets going back and forth. So even small changes in latency effect the overall experience. There's a LOT of overhead involved in networking!

Can we just please stop using analogies about FPS or Internet latency here? They are not applicable to this debate, and confuse the issue for lots of people.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 2:50:46 PM , Rating: 1
Most games like FPS use UDP for lower overhead, and so the latency measurement is a one-time trip to and from the server, no acknowledgment necessary. In this case, the OP is correct, the difference between 50ms and 100ms can be unusable for some, especially if the game has poor latency tolerance and interpolation.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/13, Rating: -1
RE: Humans Aren't Computers
By inighthawki on 9/23/2013 3:09:31 PM , Rating: 2
I'm sorry, but how is network latency not related? Network latency in a game is a form of input. The game is required to interpolate across frames in the same way an OS would react to touch input from received network packets which update the game state, thus high latency over network traffic will eventually cause a noticeable effect. Both are forms of input latency.

I also hope you didn't misread "FPS" as frames per second, I was referring to a first person shooter.


RE: Humans Aren't Computers
By Cheesew1z69 on 9/23/2013 3:13:49 PM , Rating: 1
How is NETWORK latency related to THIS article? It's not.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 3:19:22 PM , Rating: 2
It's not directly related to the article, but it is related to the discussion of how input latency affects the end user experience. Network latency in a game is a form of input latency. Once that latency becomes too high, the user notices a perceptible difference in gameplay because the engine is incapable of updating and interpolating the game state in a matter that makes sense to the visual output displayed back to the user.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/13, Rating: -1
RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 3:35:48 PM , Rating: 1
Sorry, getting frustrated. I should at least try to explain.

That latency counter when you play your game is an average. And it's not measuring your response time or any kind of response time, it's a measurement of the QUALITY of the connection between your computer and the game servers. And ALSO the connection quality of every modem link, router, switch and trunk in-between the two.

But it's SO much more complex than that.

Do you understand why using gaming latency in this discussion isn't helpful or valid?


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 4:11:25 PM , Rating: 1
Of course it's an average, but it's an average measurement of latency. This latency is received by the game as input, and the game's game state is updated accordingly. Instead of something like the view matrix adjustment or character position, it simply corresponds with various transforms and such of the objects in the level around you. This is an INPUT LATENCY, and it absolutely IS a measurement of response time. It is a direct measurement of the average time it takes to send and receive a packet to and from the server. All that matters is when you send a packet, and when you receive a packet, nothing else in-between matters.


RE: Humans Aren't Computers
By artemicion on 9/23/2013 4:18:13 PM , Rating: 2
A lot of professional gamers buy monitors based upon input lag. Apparently, it is important enough that there's a webiste with a database for it:

http://www.displaylag.com/display-database/

Note that the site rates input lag > 63 ms as "bad," so apparently, that small timeframe is noticeable to some people.

Personally, I'm not a professional gamer, but even in my limited experience I can notice the impact that very small (< 100 ms) latencies can have. I've heard Kinect latency is 90 ms, and I can definitely notice it. Display latency is generally < 100 ms, but if you play any kind of music/rhythem game you'll notice it.

Granted, input latency in a touch device is not a very significant metric of device performance, IMO. But you are completely wrong in assuming that people can't notice these latency levels.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/13, Rating: -1
RE: Humans Aren't Computers
By inighthawki on 9/23/2013 4:51:43 PM , Rating: 3
You are completely wrong. At best the science would prove that it takes a human >100ms to actually notice the movement due to the complexity of the neural pathways. The movement is still detectable, and so is all the time in-between.

This means that when you swipe your finger across a touchscreen, the finger movement itself already exhibits this 100ms delay, and the display reacts 100ms after, meaning the latency between physical input to the brain processing the image is 200ms.


RE: Humans Aren't Computers
By artemicion on 9/23/2013 4:55:02 PM , Rating: 5
Cite one source.

Here's mine:

Shortest perceivable time division by humans is between 2 and 30 ms. (Nick Herbert, Elemental Mind, Dutton, 1993, p. 50.)

You obviously are not a musician if you think 100 ms is imperceptible, but if you are playing guitar and the time it takes between strumming the guitar and the sound coming out of your speaker is 100 ms, it is VERY noticeable.


RE: Humans Aren't Computers
By 91TTZ on 9/23/2013 5:40:46 PM , Rating: 3
quote:
It's not an assumption. It's provable science!


Yes, and there are numerous studies (that other posters have cited) that prove you dead wrong.


RE: Humans Aren't Computers
By Swamp on 9/23/2013 8:19:24 PM , Rating: 2
lol I feel your pain Reclaimer77.... some of them just dont understand what your saying. it has nothing to do with gaming or networking.

It is an interesting article, its only for the fanboys and haters. I have used apple and Android devices, neither seemed to be faster with UI or any kind of input dectection.... I just basically comes down to being a fanboy or a hater...


RE: Humans Aren't Computers
By TakinYourPoints on 9/23/2013 5:21:00 PM , Rating: 2
The reason he is arguing this is because he is an Android apologist.

Differences in monitor lag are absolutely perceptible, same with touch input lag, but then again I spent lots of time playing Quake and Quake 3 back in the day so I'm very sensitive to those kinds of differences. Reclaimer doesn't strike me as much of a gamer, so perhaps that's why laggy interfaces are more tolerable to him.


RE: Humans Aren't Computers
By Cheesew1z69 on 9/24/2013 10:06:38 AM , Rating: 2
quote:
The reason he is arguing this is because he is an Android apologist.
Gee, the irony...


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 3:22:38 PM , Rating: 1
quote:
I'm sorry, but how is network latency not related?


Are you serious? I'm just...holy crap, this is mind blowing.

You're trying to compare cloud gaming to how fast an OS processes an input. And you ask me why it's not particularly useful to compare the two for the purposes of this topic?

The reason you "feel" a difference between 50ms gaming and 100ms gaming isn't because that's a huge amount of time difference. I can't put it more simply than that without a huge TLDNR about how the Internet and networking works.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 4:14:32 PM , Rating: 2
quote:
You're trying to compare cloud gaming to how fast an OS processes an input. And you ask me why it's not particularly useful to compare the two for the purposes of this topic?

I'm only trying to make a point about latency. Network latency has nothing to do with OS responsiveness, and I never tried to claim so. However the relationship between an OS and how it updates according to touch and mouse input is similar to how a game would update it's game state according to network latency of an online game.

quote:
The reason you "feel" a difference between 50ms gaming and 100ms gaming isn't because that's a huge amount of time difference. I can't put it more simply than that without a huge TLDNR about how the Internet and networking works.

Yes, it is. There is nothing more complicated to it. It doesn't matter what the modem, router, switch, etc is doing between you and the server, it matters how quickly you receive an answer to your request so that the game can update its internal state.


RE: Humans Aren't Computers
By BRB29 on 9/24/2013 8:28:11 AM , Rating: 2
He's not 100% correct but he has a point.

I'm sure almost everyone can tell the difference between 50ms and 100ms. It's not hard and very noticeable actually.

People are using avg response time is over 200ms is irrelevant. The avg response time needs a neural response all the way from your skin, up the spine, to the brain, back down the same way to create a motor response from your muscular cells.

An optical sensor can send a message to your brain much quicker because it bypasses your spine and goes straight to the brain. Then it also doesn't need to travel back down to generate a muscular response.

In conclusion, comparing avg response time to noticing a visual difference in 50ms is night and day. It is a huge time difference. I can tell the difference between 24fps, 30fps and 60fps easily in movies and I'm sure most people could. The differences between these different frame rates are under 50ms. Please don't start with the frame spikes BS because movies are not videogames.

I have never owned a single iphone but it is smoother and more responsive when I was playing with it in the store or from friends. That doesn't really matter much to me. What matters to me more is the fact that Android phones seems to crash more often and gets random hangs here and there that's annoying. However, I only use my phone for text, news, calls, and emails so I'm not a heavy user. That means price means more to me and the iphone is usually off the list.


RE: Humans Aren't Computers
By Cheesew1z69 on 9/24/2013 10:09:01 AM , Rating: 2
quote:
What matters to me more is the fact that Android phones seems to crash more often and gets random hangs here and there that's annoying.
No, they really don't....


RE: Humans Aren't Computers
By axdirkadirka on 9/23/2013 2:38:02 PM , Rating: 2
In this context, 50 ms IS a huge amount of time. Also, reaction time is far different than "able-to-tell-a-difference" time. I can tell you for a fact that a jump in ping from 50ms to 100ms changes some applications (FPS games, for example) from usable to unusable. Your example of the 215ms median response time is hilarious...imagine if you tried to watch a movie shot at less than 5 frames per second. Touch/visual perception takes far less time than you think it does.


RE: Humans Aren't Computers
By Cheesew1z69 on 9/23/2013 2:39:19 PM , Rating: 2
100ms is hardly unusable...


RE: Humans Aren't Computers
By Labotomizer on 9/24/2013 9:39:25 AM , Rating: 2
Network latency is not comparable to touch screen input latency. There is a constant stream of traffic on the network and traffic flowing both directions. So you're compounding 100ms over THOUSANDS of packets per minute. They are very small packets.

You are not making thousands of touches per minute nor is your phone touching you back. Your brain can and will compensate for the additional latency. Sure, you will notice a very slight difference when moving from an iPhone to another phone but within a day, tops, your brain will have compensated and you'll never notice it again. It's not enough of a difference, and it's a consistent difference, to really make an impact in daily use. It would only seem different on first touch.


RE: Humans Aren't Computers
By ClownPuncher on 9/23/2013 2:44:23 PM , Rating: 2
It most certainly is. This is why people used to have issues with wireless mice and LCD displays.


RE: Humans Aren't Computers
By eagle470 on 9/23/2013 3:05:13 PM , Rating: 2
I seem to remember having lost races for swimming in highschool over less time than that.

And yes, you can notice the difference.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 4:05:10 PM , Rating: 1
Another stupid post.

How were those races determined?? Did someone just eyeball the race and go "yup, this guy won!", or did you use accurate measuring devices because the delay was too close to perceive?

Rhetorical question btw.

p.s. not even Olympic races come down to milliseconds, the closest finishes in history have been in hundredths of a second. They don't even USE milliseconds!


RE: Humans Aren't Computers
By artemicion on 9/23/2013 4:31:58 PM , Rating: 4
quote:
That's fifty thousandths of a second. In no context is this a "huge" amount of time. The average median human reaction time is 215 milliseconds.


quote:
I seem to remember having lost races for swimming in highschool over less time than that.


quote:
p.s. not even Olympic races come down to milliseconds, the closest finishes in history have been in hundredths of a second. They don't even USE milliseconds!


FUN FACT:
50 thousands of a second = 5 hundredths of a second.


RE: Humans Aren't Computers
By BRB29 on 9/24/2013 12:26:54 PM , Rating: 2
or 1/20th of a second which is quite noticeable


RE: Humans Aren't Computers
By Scrogneugneu on 9/23/2013 6:23:21 PM , Rating: 2
quote:
p.s. not even Olympic races come down to milliseconds, the closest finishes in history have been in hundredths of a second. They don't even USE milliseconds!


I would like to refer you to the "Qualifying" section of the following wikipedia article, which shows that 3 F1 cars made the exact same time (with a millisecond precision) around the track at Jerez in 1997 : http://en.wikipedia.org/wiki/1997_European_Grand_P...

There are a lot of situations where milliseconds are not only significant, but absolutely HUGE.

While I agree that 50 milliseconds is not a very dramatic change in response time, it is indeed noticeable. You can't clearly detect that one interface is 50ms faster than another one, but over time, one will definitely feel a bit faster than the other one.


RE: Humans Aren't Computers
By Labotomizer on 9/24/2013 9:43:03 AM , Rating: 2
Not worth it. People assume you mean that miliseconds don't matter in everything rather than understand it doesn't matter in a touch screen. It's a lost cause. I mean you just got a comparison to F1... F1 and a smartphone are the exact same thing pretty much, except not at all...


RE: Humans Aren't Computers
By inighthawki on 9/24/2013 12:08:43 PM , Rating: 2
No, he is not claiming that. He claimed it was impossible to feel the difference between the two, to which dozens of people have claimed otherwise in this article. Yes some people in the thread have given examples that aren't really relevant, and they've been called out on it, but Reclaimer has been proven wrong, but refuses to admit it.

This is his original remark:
quote:
The difference in response times here are on such a minute scale, it's impossible for a human being to "feel" the difference


This is 100% wrong. I even replied directly providing a demo by Microsoft researchers showing direct evidence with video footage showing otherwise, where the difference in responsiveness of the UI was night and day between 50ms and 10ms, and even pretty noticeable between 10ms and 1ms. He seems to refuse to acknowledge this.

Let me re-post the video, this time directly to youtube, so there's no excuses not to watch it:
http://www.youtube.com/watch?v=vOvQCPLkPt4


RE: Humans Aren't Computers
By Reclaimer77 on 9/24/2013 2:09:25 PM , Rating: 2
The fact that you keep insisting that video applies here, is the entire problem.

The methods used in that video are specifically designed to lend a VISUAL AID for observing input delay. Something that, in normal use, WOULDN'T BE APPARENT!!!

How in the hell have you convinced yourself that applies to using a smartphone in the wild?? Do you drag around a mouse cursor super-fast on your smartphone or something?

You've taken me entirely out of context here, and insist on making horrible "points" that do not apply to this topic.


RE: Humans Aren't Computers
By inighthawki on 9/24/2013 2:34:35 PM , Rating: 3
You're so completely delusional. Of course it applies. It is exactly what is being discussed. It is the affect of input latency and the resulting latency of the visual output on the screen.
quote:
The methods used in that video are specifically designed to lend a VISUAL AID for observing input delay. Something that, in normal use, WOULDN'T BE APPARENT!!!

What? It is way more than a visual aid. This is exactly how a UI element reacts in an OS. Except instead of a single box, it's the equivalent of scrolling an entire page. There are also plenty of drawing apps on touch device, which is literally identical to the issue shown in the second set of examples in the video.

quote:
How in the hell have you convinced yourself that applies to using a smartphone in the wild?? Do you drag around a mouse cursor super-fast on your smartphone or something?

The same rubber-banding effect shown in the video applies on all smartphones. I've noticed it the first time I've picked up an Android phone. I've noticed it the first time I picked up a Windows phone. I've noticed it the first time I picked up an iPhone. None of them are perfect, they all fall prey to this latency. It is a real issue that real people experience in their UI.


RE: Humans Aren't Computers
By Reclaimer77 on 9/24/2013 2:40:47 PM , Rating: 2
Sigh...

Again, they are NOT measuring the smoothness of the UI. They are JUST measuring the time that the OS takes to process a touch input.

Jesus Christ man, I give up. Fine you win, whatever.


RE: Humans Aren't Computers
By robinthakur on 9/25/2013 10:01:40 AM , Rating: 2
You can definitely feel the difference if you use your phone constantly. It's a bit like people saying that it was not possible to discern refresh rates upwards of 60Hz for video. I can tell the difference all the way upto 150Hzin terms of fluidity and find anything less than 60Hz very distracting. GTA V I'm looking at you...


RE: Humans Aren't Computers
By Reclaimer77 on 9/24/2013 1:51:06 PM , Rating: 2
quote:
Not worth it. People assume you mean that miliseconds don't matter in everything rather than understand it doesn't matter in a touch screen. It's a lost cause. I mean you just got a comparison to F1... F1 and a smartphone are the exact same thing pretty much, except not at all...


Exactly. I mean, it's really sad. F1, swimming races, network lag and everything else is being thrown at this topic and it's just comically poor "bro science".

All I'm trying to do is put this in perspective, and nobody wants to see it. I'm being trolled, called an Android fanboi, and everything else.

Windows Phone is on this list too! Hello? How come everyone universally praises Windows Phone for being smooth and snappy? According to these numbers, Windows Phone is no better off than Android. Yet Windows Phone proponents will say Windows Phone is smooth, but Android "feel" laggy.

Clearly these tiny amounts of time do not equate to an OS "feeling" twice as fast or slow as another. Perception is everything when it comes to how humans view time. Something else is going on here.

I'm being called biased or treated like I have an agenda? I'm one of the few here who ISN'T shoveling something!

I'm running Jelly Bean, fair enough. But when I touch an icon for an app, it's highlighted instantly. Meaning the OS nearly instantly processes that input. I honestly cannot detect any perceptible delay between the touch, and the reaction to that touch input. Yet everyone here is trying to tell me that is a "massive" amount of time, they're so pro they can detect small millisecond delays, etc etc. It's getting absurd!

And you know what TakinYourPoints, it's frankly embarrassing how much of my life I've pissed away playing games. If I don't strike you as a "gamer", then I don't know who does.


RE: Humans Aren't Computers
By inighthawki on 9/24/2013 2:50:37 PM , Rating: 3
quote:
Exactly. I mean, it's really sad. F1, swimming races, network lag and everything else is being thrown at this topic and it's just comically poor "bro science".

A lot of people here are giving super bad examples like those, but you're the only one being delusional enough to believe that you cannot feel 50ms of latency on a touch device's UI.

quote:
Windows Phone is on this list too! Hello? How come everyone universally praises Windows Phone for being smooth and snappy? According to these numbers, Windows Phone is no better off than Android. Yet Windows Phone proponents will say Windows Phone is smooth, but Android "feel" laggy.

Windows Phone is popular for being fast and smooth, and from the start the responsiveness has been pretty good, certainly way better than older Android models. But that has changed. Android is almost as snappy and responsive as any Windows phone. Most of the people here claiming "My GS3 feels so laggy but my iPhone is quick and snappy" have probably also never used Windows Phones or they'd likely agree it's similar to their android experience. Note how nobody has mentioned Windows Phones in their posts for this article.

quote:
Clearly these tiny amounts of time do not equate to an OS "feeling" twice as fast or slow as another. Perception is everything when it comes to how humans view time. Something else is going on here.

Nope. As someone who has used all three phones, I can tell you without a doubt that the iPhone and even iPads have always felt more responsive to the touch for me. That's not bias or fanboism. I wouldn't buy an Apple product if my life depended on it, yet here I am defending them.

quote:
I'm being called biased or treated like I have an agenda? I'm one of the few here who ISN'T shoveling something!

To be honest I'm not really sure why this has come up either. Nothing you've stated really hinted at any level of bias towards a particular device, at least not in this discussion. Probably just the assumption that claiming you can't tell the difference is an attempt to defend the claim that Apple's device is superior in touch responsiveness is all. An illogical conclusion.

quote:
I'm running Jelly Bean, fair enough. But when I touch an icon for an app, it's highlighted instantly. Meaning the OS nearly instantly processes that input. I honestly cannot detect any perceptible delay between the touch, and the reaction to that touch input. Yet everyone here is trying to tell me that is a "massive" amount of time, they're so pro they can detect small millisecond delays, etc etc. It's getting absurd!

This explains a lot. Unlike a swiping motion where the brain's visual centers can see the actual difference between your hand's motion and the resulting display output, you are referring to the difference between you touching something and seeing a result from it. This is where you would get less of an impact due to the brain needing to process electrical impulses and responding to the actual touch of your finger.

This is not the only experience in a UI. In fact it's among one of the least common use case. Swiping and gestures are far more common in touch environment (panning, zooming, rotating, etc), and this is where the noticeable delay is.


RE: Humans Aren't Computers
By Reclaimer77 on 9/24/2013 3:14:46 PM , Rating: 2
Now wait a second, that doesn't tell the whole story at all. Even if the delay was 1ms, the screen would look choppy at low FPS displaying scrolling or fast motion.

This is why Jelly Bean, employing Google's "Project Butter", has eliminated this issue entirely. There is NO noticeable difference between iOS and the latest Android version.

quote:
This is not the only experience in a UI. In fact it's among one of the least common use case. Swiping and gestures are far more common in touch environment (panning, zooming, rotating, etc), and this is where the noticeable delay is.


Seems like you just made this up. Least use case? Lots of people type on their phones and touch icons!

quote:
you are referring to the difference between you touching something and seeing a result from it. This is where you would get less of an impact due to the brain needing to process electrical impulses and responding to the actual touch of your finger.


So in order to be right, you've thrown out use-scenarios that bolster my argument by claiming they aren't relevant, and focus on use-scenarios that highlight your side of the argument? Okay cool...

quote:
To be honest I'm not really sure why this has come up either. Nothing you've stated really hinted at any level of bias towards a particular device, at least not in this discussion. Probably just the assumption that claiming you can't tell the difference is an attempt to defend the claim that Apple's device is superior in touch responsiveness is all. An illogical conclusion.


Well, thank you. That means a lot honestly.

quote:
I wouldn't buy an Apple product if my life depended on it


Hey now we're talking! I take back every bad thing I've ever said to ya :)


RE: Humans Aren't Computers
By inighthawki on 9/24/2013 4:05:47 PM , Rating: 2
quote:
So in order to be right, you've thrown out use-scenarios that bolster my argument by claiming they aren't relevant, and focus on use-scenarios that highlight your side of the argument? Okay cool...

I wasn't implying these aren't use cases (typing is a pretty common one - I probably overlooked quite a bit in my rushed reply, I'm sure there are other common use cases for simple taps as well), but you cannot deny that gestures make up a large percentage of use scenarios on a device. It makes up a very large portion of the user's experience in the device's UI. I own a Windows Phone myself, and I've found the majority of my interaction is through gestures. Scrolling lists, panorama views, start screen. To be honest, from the first time I've used a Windows Phone, even then I always had a strange wonder about what it was that still made the iPhone feel even more responsive to those gestures.


RE: Humans Aren't Computers
By TakinYourPoints on 9/25/2013 5:32:30 PM , Rating: 2
quote:
And you know what TakinYourPoints, it's frankly embarrassing how much of my life I've pissed away playing games. If I don't strike you as a "gamer", then I don't know who does.


Spending years doing raids in WoW doesn't qualify one as an expert on input reaction time


RE: Humans Aren't Computers
By 91TTZ on 9/23/2013 4:36:10 PM , Rating: 2
quote:
That's fifty thousandths of a second. In no context is this a "huge" amount of time. The average median human reaction time is 215 milliseconds.


There's a huge difference between perception time and reaction time. It takes a while for your muscles to react.


RE: Humans Aren't Computers
By PrinceGaz on 9/24/2013 12:14:47 PM , Rating: 3
quote:
That's fifty thousandths of a second. In no context is this a "huge" amount of time. The average median human reaction time is 215 milliseconds. In real world use, sorry, this is simply not a perceptible difference in UI response time.

Even a tiny delay on a touch-screen input will be noticeable, because it is quite different to most other input-device -> display-device usage scenarios.

When using a mouse/keyboard or gamepad, you see something happen on screen (or in some cases hear it through your speakers) and respond to it by sending a signal from your brain to your hand which is holding the mouse/keyboard or gamepad. There's a human-biology latency in that action which the brain has always had to compensate for (we are used to looking at our hands when using them for other purposes so the delay in what they do is expected) and is generally unnoticeable, plus an additional latency caused on the computer side which we perceive as lag.

The lower that perceived lag, the better, and if it is low enough then you will not generally notice it, because your brain is already compensating for the lag caused by sending the signal to the muscles in your hand, so the slight additional delay in on-screen action becomes insignificant.

A touch-screen is a totally different situation because whilst the brain expects and compensates for the delay from sending the signal to your hand's muscles and seeing your hand move, what it does not expect and will always "feel laggy" is any delay between seeing your finger move and the response on the screen beneath said finger, because it is seeing both movements simultaneously through the eyes.

When drawing on a piece of paper with a pen or pencil, there is zero display-lag on the piece of paper. You see your hand move and the results of that action together simultaneously. That's what the brain expects and that is the ideal touch-screen.

Now try using a touch-screen with 110mS lag. Your finger moves but there's now a gap between where you see your finger, and where the screen is showing your finger as being. It is normal to move your finger in excess of 1m/s on a touch-screen. That means that in 110ms, the screen is 11cm (over four inches) behind your finger! That is a lag that anyone who isn't blind will see very easily indeed.

Even the 55mS of the iPhone 5 is still quite poor but at least it is now only half as bad.

Until touchscreen lag is well under 10mS then it will still be noticeable. 50mS lag really is a huge amount of time when using a touchscreen, because the brain is used to seeing zero lag between hand-movement and the effect of it.


RE: Humans Aren't Computers
By schnazzer on 9/25/13, Rating: 0
RE: Humans Aren't Computers
By MozeeToby on 9/23/2013 4:24:57 PM , Rating: 2
50ms isn't huge. Not at all in fact. If you're a particularly tall person it takes nearly half that long for a nerve impulse to travel from your foot to you brain. The visual cortex takes significantly longer than that to process an image from the eye. 50ms is well, well below the time that the brain is capable of "editing out" of your awareness.

Now, 150ms... that's getting pretty close to the limit. If these numbers are accurate, and I consider that a big if, this would be an interesting place to look to improve the perceived performance.


RE: Humans Aren't Computers
By xti on 9/23/2013 4:41:04 PM , Rating: 2
real world - i don't bother messing with my wife's iPhone5. the response time feels light years ahead of my galaxy3, and it just feels extra sluggish when i go back to it.

so, we all have excuses like "smaller window" or "jobs had too much of a beard" but I wish my 4.3 felt as good as iOS6...havent messed with 7 yet. im not talking about customization or other perks of android...i just want it to be smooth...i shouldnt settle for a trade off.

press chrome...wait forever...yay it opened! i should close some apps...oh wait they are all closed. dammit jelly bean.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 4:55:05 PM , Rating: 2
Again, read the article. You're talking about how fast a phone takes to launch an app. This is measuring the response of a touch input, that is all! Not UI lag, not smoothness, not anything other than input delay.

Seriously there cannot be this many stupid people on one tech forum. The article must need to be re-written to be more explanatory or something...


RE: Humans Aren't Computers
By xti on 9/25/2013 8:58:51 PM , Rating: 2
who freaking cares. get us better UI lag, smoothness, butt lag, w/e. thats what this article sparked from the customer base, think outside the box man, nerds love everything so black and white.

android phones need to speed back up, feel like they are just coasting with complacency.

i mean... look at this:
http://blogs.wsj.com/digits/2013/09/25/samsung-wel...


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 5:16:56 PM , Rating: 2
quote:
50ms isn't huge. Not at all in fact. If you're a particularly tall person it takes nearly half that long for a nerve impulse to travel from your foot to you brain. The visual cortex takes significantly longer than that to process an image from the eye. 50ms is well, well below the time that the brain is capable of "editing out" of your awareness.


Shhhhhhh, this is Daily Tech. Facts and reason aren't allowed here.

Daily Tech is chocked full of mutant X-Men who perceive a thousandth of a second as if it were a year in our time.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 5:23:41 PM , Rating: 2
This is not about the delay to process an image in the brain's visual centers. There are two variables into play.

t1 = the time at which the brain fully realizes the finished movement of your finger
t2 = the time at which the brain fully realizes the finished movement of the display

Both variables account for this exact delay he is speaking of in the brain's visual processing. The latency experienced by the end user is (t2 - t1).

It is astonishing that you act like you know everything. The person who agrees with you and is also wrong is the only correct person here, but the other 100 people giving you cited sources, demos, and personal experience are all completely ignorant and stupid.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 5:38:05 PM , Rating: 2
quote:
It is astonishing that you act like you know everything


I know, right? Sometimes I even amaze myself!


RE: Humans Aren't Computers
By troysavary on 9/23/2013 7:27:42 PM , Rating: 2
Get used to it. He is the guy the saying "He's so stupid he doesn't know he's stupid." was made about. It is useless to try to speak logic with reclaimer. All he thinks is "Must defend precioussss".


RE: Humans Aren't Computers
By superstition on 9/23/2013 7:31:24 PM , Rating: 3
Real men use Android. Real men use Windows. Real men use Linux. Only girly men, old people, stupid people, and people scared of technology, use iOS and Mac. Even dumb trendy teenagers have realized that Apple isn't cool anymore.


RE: Humans Aren't Computers
By 91TTZ on 9/23/2013 5:56:08 PM , Rating: 2
quote:
The visual cortex takes significantly longer than that to process an image from the eye. 50ms is well, well below the time that the brain is capable of "editing out" of your awareness.


Wrong.

The mistake that you're making is that you're citing realtime reaction time whereas perception time is post-processed.

People can most definitely tell the difference in time between events that occur less than 50 ms apart. They're not able to take action on this information in under 50 ms, but if they see something unfold before their eyes in under 50 ms the brain stores and then post-processes this information.

Let me ask you this: Can you hear a gunshot? Can you hear 10 gunshots in a second? What about 20 per second? These events would be 50 ms apart and if what you're claiming is true a person would not be able to tell the difference between 10 shots per second and 20 per second, since it would be "well, well below the time that the brain is capable of editing out of your awareness."

I bet once you begin to think about this example you'll realize that different parts of the brain are capable of handling the stimuli they detect at different rates. Your auditory and visual processing centers can handle and process it very quickly. You "think" much more slowly than the brain can perceive stimuli in specially developed areas of the brain.


RE: Humans Aren't Computers
By MichalT on 9/23/2013 2:12:24 PM , Rating: 3
We're talking about differences between 1/10 of a second and 1/20 of a second between the current flag ships models on UI input/output. You're really sitting here on a gaming site and telling me that people aren't going to be able to notice the difference between something that responds effectively at 18 frames per second versus 9fps? There are gamers that won't accept anything less than 60fps.

Its not until the device makers gets it to around 24-30 fps that regular people will stop noticing any improvements, and likely will have to be around 60+ fps before people like gamers stop noticing improvements.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 2:17:45 PM , Rating: 2
You're actually comparing two different concepts. FPS is a measure of throughput, the amount you can render in a period of time. Latency is the delay between seeing the visual change. You'd be surprised how much delay there already is in rendering a typical frame in a PC game. Games running with vsync enabled will already queue up frames if running at over 60Hz. This can be up to 3 frames in advance by default in Windows (16ms * 3) + input latency + latency required to send the data to the display device and have the pixel change colors, which can be another 50+ms on a lot of displays. a change in 10-20ms is less noticeable than you'd think.


RE: Humans Aren't Computers
By MichalT on 9/23/2013 2:22:38 PM , Rating: 2
No, they are the same thing. They are both a measure of latency; its how long it takes after you move your finger on the device and it is able to respond. I just put tried to put it in a context that people here would understand.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 2:31:23 PM , Rating: 2
No, you are wrong. Comparing input delay with FPS is two entirely different things.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 2:32:29 PM , Rating: 2
I guess my problem with your comparison was trying to compare to graphics. From a technical perspective, yes you are sampling the input device a certain number of times, but the OS will be capable of smoothing out the motion, and thus the end user experience is always in the form of graphical latency, but possibly at a high framerate.


RE: Humans Aren't Computers
By invidious on 9/23/2013 2:48:27 PM , Rating: 2
In principle you are right, but like half of the other comments on here you are confusing the refresh rates of input devices and output devices. In general I'm seeing a lot of bro science going on here that just isn't founded in reality. Slow refresh rates on android devices is a bad thing but not as bad as some are making it out to be.

The 60 frames per second standard that you are referring is for visual output animations on a computer monitor, which is capped by the monitor, graphics card, and application software. Whereas input device refresh rate is determined by your mouse/keyboard/controller/touchscreen and is capped by the application's logic refresh rate.

The 60fps that you see on a PC are aninations that are triggered by the game logic. The game logic might only be refreshing its logic outputs at 30hz, but the animations display at a higher frame rate to be more appealing to hte human eye.

High quality PC gaming devices have refresh rates in the thousands of hertz, but in reality the game only sees a small sample of those at a much much slower rate. There are definately advantages of having the fast refresh rates on input devices, but it doesn't correlate one-to-one with overall responsiveness.


RE: Humans Aren't Computers
By TSS on 9/23/2013 7:49:55 PM , Rating: 1
Can i just say, that as a gamer who traines his reaction speed to just 75 ms where normal humans average 200ms, i find it highly distressing people discuss a 50ms-100ms delay on a smartphone to this degree.

You people do realise that when i started gaming on the commodore 64, my mother would make dinner at the same time i loaded up the game, just so we could play it after dinner was done? I'm 26 years old now, that's not that old.

Meanwhile today i carry a PC in my pocket that's fast enough to run GTA 3.

So please. People shouldn't care this much about a 50ms difference. Especially gamers, because any self respecting gamer will dismiss "real" gaming on a smartphone anyway because of the damn toutch screen interface (regardless of speed). If there was a controller/keyboard/mouse attached with that kind of lag, then there'd be something to complain.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 8:16:10 PM , Rating: 1
I'm telling ya man, fanbois! Making a huge deal of nothing. I've tried to, you know, chill them out with the facts but alas...


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 8:31:05 PM , Rating: 2
That's hilarious. Facts? You haven't cited a single thing, yet everyone posting about you being wrong has either given personal testament or some citation to an actual source. You have not posted a single fact yet, only opinion, and then you proceed to call everyone who disagrees stupid. The inability to see or comprehend something is not a fact disproving it ;)

Have you even watched the video I linked you in my first reply? The difference between even 10ms and 1ms latency is night and day, and 50ms is even more substantial. You'd have to be seriously braindead to not see it. Seriously, I want you to literally say it. "I watched the video all the way through and saw no difference." And when you do, I will know for fact you are full of it.

Also how does disagreeing with you on this make someone a fanboy or hater? I'm not much of an Apple fan, but I'm making the claim very loud and clear. Their touch latency is clearly superior. I have nothing against Android either, I think it's a fine OS. I just think that you are completely and utterly wrong, and you are too stubborn to admit it. Maybe that's why you don't have any replies to the posts that actually give real information.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 8:26:54 PM , Rating: 2
Response speed has nothing to do with the ability to notice latency between an input and output device.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 8:31:43 PM , Rating: 2
Sorry, "Reaction speed" - Need an edit button :(


RE: Humans Aren't Computers
By kingmotley on 9/23/2013 2:18:38 PM , Rating: 1
115ms (Android) is about 8.6 times per second, while 55ms (iPhone 5) is about 18.1 times per second. Can you tell the difference when your game is playing at 8.6FPS vs 18.1FPS?


RE: Humans Aren't Computers
By danbob999 on 9/23/2013 2:42:25 PM , Rating: 2
You are confusing latency and throughput.


RE: Humans Aren't Computers
By Tony Swash on 9/23/13, Rating: 0
RE: Humans Aren't Computers
By superstition on 9/23/2013 2:23:50 PM , Rating: 1
Tony, everything Apple does is smoke and mirrors. It's all illusion to fool the iSheep. Steve Jobs was a terrible person which is why Apple is such a terrible company. Nothing Apple does is ever innovative. The company just takes money from suckers.

Samsung is much better, and so is Microsoft. Neither company would ever copy anyone. They truly innovate. The people who buy their products know a great value when they see it. They aren't suckered in by outdated small screens and cheesy colors.

Apple is going to crash and burn because of its arrogance, and not even the iSheep are going to keep it afloat, especially since Mac and iPhone aren't cool anymore.


RE: Humans Aren't Computers
By Rukkian on 9/23/2013 2:35:17 PM , Rating: 3
There is much more to the story, however. You say he misunderstands, but there is more to it that just registering touches. Since I have gotten my new phone, I have had several diehard apple fans comment how fast and fluid it truely is. While touch responsiveness is nice, if it isn't able to complete the action quickly, that is (imo) where people will see the lag.

My old phone (galaxy nexus) definately showed some lag, which I would expect out of a 2yo device, but more than just being old, it definately was not as smooth.

My new phone (LG G2) is so buttery smooth, and responsive it is like night and day. While it was not tested, I would bet it is probably close to the responsiveness of the Moto X, (which was tested), and according to that one test failed compared to IOS devices, however my phone seems (imo and to several apple diehards) so smooth, and even faster for pretty much everything than my wife's new Iphone 5c.

While this test is probably a good test, there is much more too measuring smoothness than just a 40ms on a touch response.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 2:47:32 PM , Rating: 3
Exactly! We're talking about how fast the UI itself can process an input.

When people talk about "UI lag" and smoothness they are talking about how fast a program takes to launch, a page to scroll, a menu to load etc etc.

That is NOT input delay.


RE: Humans Aren't Computers
By elleehswon on 9/23/2013 3:03:29 PM , Rating: 2
as an owner of an LG G2, i can also confirm the "buttery" smoothness of this phone. of note, i was looking at the 5s, 5c, and the blackberry whatever it was, and the moto x. all seemed pretty comparable though the android devices definitely had more going on as far as ui "features" ie, more calculations, more to draw, etc,etc. that being said, if the usability is satisfactory, does this metric matter much, if at all?


RE: Humans Aren't Computers
By troysavary on 9/23/2013 4:34:50 PM , Rating: 1
People are so used to downrating you that even when you are right, your comment ends up red. Funny.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 4:40:12 PM , Rating: 1
Often times I feel he actually is right, but then drowns it in so much bias that it just sounds wrong. This is one of his few posts that actually sounds down to Earth.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 5:26:38 PM , Rating: 1
quote:
Often times I feel he actually is right


Well that certainly explains a lot...


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 5:30:06 PM , Rating: 2
haha yes really funny. God forbid at some point I think he makes a valid point about how Apple did one or two things better than the competition. It does happen from time to time.


RE: Humans Aren't Computers
By troysavary on 9/23/13, Rating: 0
RE: Humans Aren't Computers
By Tony Swash on 9/24/13, Rating: 0
RE: Humans Aren't Computers
By Cartman Jones on 9/24/2013 10:58:03 AM , Rating: 2
Now I have a terrible mental image. Thanks. Luckily, I don't know what you look like so my mental image is a bit vague. Thanks for the laugh though.


RE: Humans Aren't Computers
By mugiebahar on 9/23/2013 3:00:15 PM , Rating: 2
Seriously dude, I have both the GS3 and iPhone 4S (same generation relatively) huge difference. And it was discussed a long time ago. It is the fact there is an extra step or layer between hardware and software. But I digress, as there is a huge (I mean huge) difference in response time. I'm almost ADD in some regards and can't stand slow responses. That's why my GS3 is sitting on a shelf. Its a beautiful device but heaven help me its laggy compared to the iPhone. Do I wish they correct the problem, hell yes as I've wasted $450 dollars on a phone that bugs the crap out of me. If that's fixed I will use it but I don't see it being fixed as its a software level issue and a priority level issue. So unless you have both and used them (not just in store either) I mean really use them, you would know that there is a huge difference. Speaking out of ignorance is really annoying, I can stand opinions but not ignorance. That's just spewing crap on the Internet, and there is way too much of that nowadays. And as was also brought out, this is tech cite we actually know our crap so please don't think for a second ignorant comments will not be trampled on.


RE: Humans Aren't Computers
By Solandri on 9/23/2013 3:58:51 PM , Rating: 3
It's probably a lingering aftereffect from when these phones were all single core. Apple gave absolute priority to the UI for CPU time, letting apps suffer (apps couldn't run in the background). Android took a more holistic approach and ran everything equally. You could run background tasks, but the UI could stutter and lag if something intensive was running. When it was clear the UI needed more priority, they did their Butter project. Likewise Apple realized they had to allow stuff to run in the background, and worked to add multitasking to iOS.

Both approaches are benefiting from the rapid increases in processor speed. iOS is now able to run background tasks somewhat competently. Android's UI response speed is improving. In another 5 years, none of this is going to matter - they'll both be fast enough in UI response to be indistinguishable by a human, and both will run background tasks competently.

Same thing happened with Mac vs Windows in the 1980s. Apple designed dedicated hardware to deal with the UI, Microsoft did it in software. Consequently the Mac's UI was a lot smoother at the same resolution and color depth (specifically, when you moved the mouse, the pointer was redrawn much more frequently on the Mac). But CPU speed improved so quickly than in about 5-10 years the two were indistinguishable.

I wouldn't say either approach is "better", they're just different. Apple's approach gives you access to some functionality and smoothness earlier in time. Good for people like you, but not worth it in the long run for people who don't care about millisecond improvements. Most of the improvements they designed are obsolete now (and the money they spent doing it wasted) because it's simply not worth paying for dedicated hardware to handle it when the CPU can do it plenty fast enough. The only exception is the hardware mouse - a dedicated video card hardware function which holds a bitmap of the pointer and bitblts it on top of the image of the screen. And even that is obsoleted by 3D overlays if your OS can use a 3D video card (which conceptually does the same thing but for general 3D objects, not just the mouse pointer - e.g. it's now used to draw overlapping windows in hardware).


RE: Humans Aren't Computers
By Rukkian on 9/23/2013 4:13:38 PM , Rating: 2
I think that what you are describing is more about the GS3, than Android in particular. Samsung loves to put a bunch of crap (touchwiz) that adds a lot of lag in pretty much every phone I have used that they made. The only way to get around it is to somehow get rid of touchwiz, and turn off all of their extraneous bloat.

I definitely see a difference in that, however it is due to backend stuff as opposed to what this metric is measuring.


RE: Humans Aren't Computers
By xti on 9/23/2013 4:43:49 PM , Rating: 2
i have task650's rom on my gs3....its not even close to as fast as iPhone5's OS smooth feel.

everyone here who is comparing MS and specs...go freaking pick up both for 30 min each and then get back to us.

we shouldnt be ok with androids responsive feel..its 20freakingThirteen. we should demand better.


RE: Humans Aren't Computers
By Rukkian on 9/23/2013 5:06:03 PM , Rating: 2
That is the thing, though, it really is dependent on the device. I tried pretty much every rom I could on my old galaxy nexus, and all still showed some lag. I am not sure if it had to do with the type of panel they use or what, but every single Samsung I have ever played with had some lag, especially compared to most iphones.

In contrast, my new phone (LG G2) seems quicker than pretty much any other device I have ever played with (which is a lot). Even my wifes Iphone 5c seems slower.

There is more to it than what this metric is measuring.

At this point, I am not planning on putting a custom rom on this phone (even though I hate some of the customizations they did to it) because it is just so smooth.


RE: Humans Aren't Computers
By TakinYourPoints on 9/23/2013 6:12:54 PM , Rating: 1
This is also worth noting:

quote:
Another possibility is that while the Android and WP8 code are running on runtimes (Dalvik and CLR respectively), the iPhone code is written in closer-to-the-metal Objective-C, which may reduce some latency. In future TouchMarks, we’ll compare C/C++-based Android apps to Java based apps to determine if this is the case.


The fact that nearly everything in Android is running through an interpreter while everything in iOS runs natively is one reason why even the same application runs better in iOS than its Android port.


RE: Humans Aren't Computers
By juserbogus on 9/23/13, Rating: 0
RE: Humans Aren't Computers
By troysavary on 9/23/13, Rating: 0
RE: Humans Aren't Computers
By 91TTZ on 9/23/13, Rating: 0
RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 4:50:45 PM , Rating: 2
More apples to oranges comparisons. Your first has nothing to do with the topic, and everything to do with visual acuity.

Your second is based on a false premise of a myth that others perpetuate. People who claim your eyes cannot "see" beyond 60FPS are just ignorant.

Seriously, I'm surrounded by idiots. What does any of this have to do with the article?

The average human response time is between 150 and 300 milliseconds. Scientific studies have proved most humans don't even perceive a change in time of 100 milliseconds. These are indisputable facts.

But you guys want to sit here and, universally, claim they can perceive an OS touch response time delay of ~40ms?

I'm surrounded by idiots. End of story!


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 4:55:57 PM , Rating: 5
Clearly you are the stubborn idiot, because there are dozens of people surrounding you claiming your are wrong based on personal experience, and I event provided you with a link showing Microsoft researchers show a DEMO of the difference in input latency and their results output, and you see a noticeable difference between them, all the way down to 1ms. Unless maybe you cannot see it, then I feel sorry because it just means you are slightly impaired, and I understand most of your posts now.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 5:11:02 PM , Rating: 1
Thousands of people said Einstein was wrong too :)

Enjoy being willfully ignorant. You just don't get this, and nothing I could possibly say would change that. Keep comparing apples to oranges buddy!


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 5:13:22 PM , Rating: 3
Absolutely astounding.. Really is...


RE: Humans Aren't Computers
By troysavary on 9/23/2013 7:33:36 PM , Rating: 3
You aren't seriously comparing yourself to Einstein?


RE: Humans Aren't Computers
By 91TTZ on 9/23/2013 5:39:13 PM , Rating: 2
quote:
Seriously, I'm surrounded by idiots. What does any of this have to do with the article? The average human response time is between 150 and 300 milliseconds. Scientific studies have proved most humans don't even perceive a change in time of 100 milliseconds. These are indisputable facts.


It's funny that you think that I'm an idiot when I can clearly see the mistake that you're making.

You're confusing response time with perception time. Response time is the measurement of the accumulated delay between your brain perceiving a change in stimulus, deciding an action to take, and then sending the signal to (usually) your hands. On the other hand perception time is person's perception of the chronology of a stimulus, and that stimulus is stored and post-processed. Your brain calculates and comprehends the sequence after it's already happened. The brain can perceive stimuli much faster than it can act on them. As a result, response time will be much longer than perception time.

Since this discussion is about the perception of the delay on different user interfaces on smartphones we're talking about perception time. The reason you see other people bringing up examples of monitor refresh rates or lag in games is because these conditions test a person's perception of time. You're failing to see that because you're confusing time perception with response time.

quote:
Your first has nothing to do with the topic, and everything to do with visual acuity.


You're dead wrong. Visual acuity refers to the eye's resolution and clarity. It has nothing to do with detection of time. My example of refresh rate obviously refers to time, not clarity of the picture. You claim that most humans don't even perceive a change in time of 100 milliseconds. This is wrong. If the brain was unable to comprehend changes in stimulus that occur at intervals less than 100 milliseconds we wouldn't be able to perceive the difference in frequency of vibrations, frequency of sounds, refresh rates of monitors, or firing rate of a machine gun. If most humans didn't perceive a change in time of any less than 100 milliseconds that means that videos would look completely fluid at 10 frames per second, and they wouldn't be able to tell the difference between 10 fps (100 milliseconds), 20 fps (50 milliseconds), 50 fps (20 milliseconds) or 60 milliseconds (16.7 milliseconds).

The reason for the disparity is because not all areas of the brain operate at the same speed. Areas responsible for visual and auditory processing are able to process stimulus much faster than you're able to think or move your muscles.

To sum it up, you think that you're surrounded by idiots because you're the one making the mistake. I guess if you thought that 5x5=26 you'd think the world is full of idiots, too.


RE: Humans Aren't Computers
By TakinYourPoints on 9/23/2013 6:18:58 PM , Rating: 2
You honestly believe that humans cannot perceive moment to moment differences of roughly a quarter of a second, or even a tenth of a second?

Seriously?

Maybe you can't.

Were you one of those guys that was perfectly fine running their old desktop CRT at 60hz? That gave me a headache, 100hz+ or bust.


RE: Humans Aren't Computers
By Reclaimer77 on 9/23/2013 8:22:41 PM , Rating: 2
quote:
You honestly believe that humans cannot perceive moment to moment differences of roughly a quarter of a second, or even a tenth of a second?


Of course I'm not saying that. I'm talking about like ~30 milliseconds and people are pretending that's some HUGE amount of time.

In a phone in the wild, you might notice that difference, you might not. But it's hardly an issue. But go ahead and proclaim it's more supreme dominance by Apple.


RE: Humans Aren't Computers
By inighthawki on 9/23/2013 8:49:41 PM , Rating: 2
quote:
In a phone in the wild, you might notice that difference, you might not. But it's hardly an issue. But go ahead and proclaim it's more supreme dominance by Apple.

So you've gone from admitting that it's IMPOSSIBLE to "maybe you can, maybe not" and assuming anyone who might is immediately proclaiming that they are Apple fanboys claiming "supreme dominance"? Now you just sound defensive. How about finally supplying some citations to all those facts you through at us? Or are you no longer posting because you can't?


RE: Humans Aren't Computers
By vanadiel on 9/23/2013 7:20:33 PM , Rating: 2
Go to this site, click as fast on start/pause as you possibly can (I managed 100 msec), and tell me how much of the animation on the middle right side of your screen moved in that 100 msec time frame or if you even saw it move at all.

http://tft.vanity.dk/inputlag.html


RE: Humans Aren't Computers
By xti on 9/23/13, Rating: 0
RE: Humans Aren't Computers
By Rukkian on 9/23/2013 5:24:03 PM , Rating: 2
That is the GS3. Even the GS4 is somewhat laggy.


A sore spot has been touched
By Tony Swash on 9/24/13, Rating: -1
"There is a single light of science, and to brighten it anywhere is to brighten it everywhere." -- Isaac Asimov














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki