<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Gábor Szántó on Medium]]></title>
        <description><![CDATA[Stories by Gábor Szántó on Medium]]></description>
        <link>https://medium.com/@djplayerapp?source=rss-74581f6367f6------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*GMAp97-hjq6hCa4i.png</url>
            <title>Stories by Gábor Szántó on Medium</title>
            <link>https://medium.com/@djplayerapp?source=rss-74581f6367f6------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 03:43:13 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@djplayerapp/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Behind The Touchscreen Latency Benchmark By Agawi]]></title>
            <link>https://medium.com/@djplayerapp/behind-the-touchscreen-latency-benchmark-by-agawi-d8feca9b8a49?source=rss-74581f6367f6------2</link>
            <guid isPermaLink="false">https://medium.com/p/d8feca9b8a49</guid>
            <dc:creator><![CDATA[Gábor Szántó]]></dc:creator>
            <pubDate>Thu, 10 Oct 2013 19:18:42 GMT</pubDate>
            <atom:updated>2013-10-10T19:18:42.096Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/700/0*PlNPQRJUMidQzFVy.png" /><figcaption>Agawi Touchmarks</figcaption></figure><h4>It’s not the touchscreen latency you may think.</h4><p>Smartphone latency is one very important field for me. My flagship app, <a href="http://djplayerapp.com/">DJ Player</a> is targeted for pro DJs, and live music/DJ applications require the best realtime response.</p><p>I become very excited when cloud streaming company Agawi released the first quantitative and objective benchmark of app response times, called <a href="http://appglimpse.com/touchmarks.php">TouchMarks</a>. But their findings can be misleading, so i’d like to clear a few points here.</p><h3>It’s Not That Touchscreen Latency</h3><p>The TouchMarks number is an “overall feeling” mark, not the latency of any hardware component, <em>especially not the latency of the touch panel</em>. TouchMark is the amount of many smaller component’s latency combined.</p><p>Every touch panel has an internal microcontroller detecting the touches, and it’s connected via USB. The firmware is optimised for the USB polling rate, which is 1 ms. Now you can see, <em>detecting pure touches can be really fast</em>. I have some touch panels (i wrote OSX drivers for), and they have no more than 3 ms latency.</p><h3>Then Why The Best Result Is 45 ms?</h3><p>Because the operating system’s scheduling and the GUI internals are the biggest factors.</p><ol><li>The operating system parses the touch panel’s input, and tells your app’s main thread that some touch event happened. But it runs whenever it runs, because it’s not a high priority thing: it receives touch events 4-16 ms later, usually. If your app is not well written, the latency can be even higher.</li><li>Your app reacts to the touch input somehow. But when your app instructs a change in the user interface, <em>it doesn’t happen immediately</em>. Graphical instructions are batched together by the graphics driver, and the screen is changing per frames.</li></ol><p>60 fps refresh rate means 16.7 ms time for one frame. And graphics drivers do double, triple or whatever buffering, so the instruction has no effect on the next, but the third frame from now…</p><p>Let’s calculate the best (lowest latency) theoretical scenario:</p><p>- 2 ms for the touch panel + USB</p><p>- 5 ms for iOS touch input processing and delivering to the app</p><p>- 34 ms due triple buffering in the graphics driver and 60 fps</p><p>The result is 41 ms. Look for Agawi’s best measurement: 45 ms. <em>Got it?</em></p><h3>Agawi’s Guesses Starts Wrong</h3><blockquote>“…our best guess at Agawi is that Apple’s touchscreen hardware is better optimized or more sensitively calibrated for capturing and processing touch.”</blockquote><p>Now you see it’s not the case.</p><blockquote>“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.”</blockquote><p>Yes and no. As Agawi’s code is using OpenGL on both iOS and Android, the huge latency difference comes from 2 factors, all software based:</p><ol><li>A typical Android app receives touch events later, due thread scheduling. Agawi should create an Android NDK (C++-based native app) to test.</li><li>The graphics driver on Android devices is less optimised than Apple’s.</li></ol><h3>So What Did They Measure?</h3><p>A little bit of thread scheduling and a lot of graphics driver latency. And just a tiny tiny bit about the touch panel.</p><h3>Congratulations…</h3><p>… for Agawi, they did a great marketing job with the first touchscreen latency benchmark, superb idea! I hope it opens some thinking how touchscreen latency can be improved.</p><p><a href="http://www.youtube.com/watch?v=vOvQCPLkPt4">Microsoft is also researching in this field</a>, and they found 1 ms would be the best. There is a long road ahead to that…</p><p>It doesn’t matter for most mobile applications, but it would be cool to see an option in iOS or Android to have custom frame rates. With 240 fps and double buffering for example, we could enter a new region of the “touchscreen feel” for special apps.</p><h3>Shameless self-advertisement</h3><p>Until then, you can test the lowest possible mobile app latency with my <a href="http://djplayer.com/">DJ Player app</a>.</p><p>Plus, check our fresh startup with unique desktop-grade processing for mobile devices, called <a href="http://superpowered.com/">Superpowered</a>!</p><p><em>We really care about smoothness on mobile.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d8feca9b8a49" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How Big Digital Music Stores Missed the Opportunity in “Other” Countries]]></title>
            <link>https://medium.com/@djplayerapp/how-big-digital-music-stores-missed-the-opportunity-in-other-countries-a2ea0dee56b2?source=rss-74581f6367f6------2</link>
            <guid isPermaLink="false">https://medium.com/p/a2ea0dee56b2</guid>
            <dc:creator><![CDATA[Gábor Szántó]]></dc:creator>
            <pubDate>Fri, 04 Oct 2013 07:46:29 GMT</pubDate>
            <atom:updated>2013-10-04T11:03:52.256Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/699/0*OkUBF3Vbb0kqBTuI.png" /></figure><h4>The genie is out of the bottle</h4><p>I recently read on <a href="http://djworx.com/unsound-teaser-double-edged-sword-digital-age/">DJWORX</a> about the upcoming Unsound movie, and there is the ordinary conversation how digital and downloads changed the music industry, why record labels suck, who deserves the money, etc.</p><p>I’d like to add another point:<strong> the biggest music stores with the best catalogue were waiting too long to enter “smaller”, less “sexy” countries.</strong></p><p>Take my location, Central and Eastern Europe for example, known for high piracy rate. And let’s go back to the last decade, the 2000s.</p><p>My region was no exception in welcoming the iPod and mp3 revolution at that time. But getting those gadgets and feeding them with content wasn’t easy, because the big tech companies didn’t care about us. People bought the shiny iPods for Christmas and birthday gifts just like in the western world, but left without any music store with a decent catalogue. No iTunes Store at all.</p><p>Imagine a country, where the iPods are having 80% market share. Almost all purchased via “gray import”, because there is no Apple store. And there is no way for people to legally purchase music onto those devices.</p><p>So what did people do?</p><p>Pirating mp3 files. And no, pirating music was not that easy for the average user. It was very hard for the average user. But they learnt it slowly, because they had the time: fast forward to 2008, 7 years after the iPod’s release, we still didn’t have any good option to buy music.</p><p>Some tech scene guys and me were creating the “We Want Apple Hungary” campaign and pitched Cupertino, to get the attention of our country and region, because they still treated us like shit (bad support, small supplies, no services, etc.).</p><p>I was lucky to present on MIDEM 2009 too (the biggest worldwide music biz conference), and talk with important music business people, including iTunes Music Store managers. We had a great discussion, but <strong>all of them gave me blank answers.</strong> I guess most of them couldn’t pick some CEE countries on the map at all. I warned them that their ignorance will have long-term consequences.</p><p>Finally, Apple created Apple Hungary and things changed slowly. We get the iTunes Music Store in 2011.</p><p><strong>But they were too late.</strong> 10 years elapsed since the mp3 revolution, and even the lamest users did learn how to pirate music. It become part of our culture.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a2ea0dee56b2" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>