touch screen monitors for windows 7 free sample
If you are a Windows 7 user who owns a touch screen monitor and wants to use software for touch screen using Windows 7 multi-touch screen features, I"m sure you are looking Windows 7 software for touch screen.
There are also software for Windows 7 multi-touch properties. Here in this guide I wish to share a list of Windows 7 touch screen software you can find in the software market and download free or buy from software vendors.
Corel is now building touch screen software for Windows 7. In this multi touch screen software list, Windows 7 users will find multi-touch applications from Corel corporeation.
If you think to buy a Windows 7 touch screen PC computer, note that some of the Fujitsu touch screen PCs and NEC touchscreen PCs have built-in Paint it! touch screen software.
Microsoft teams have build Microsoft Touch Pack for Windows 7, a pack of multi touch screen software and touch enabled applications to demonstrate the multi-touch features of Windows 7.
The Touch Pack for Windows 7 includes Microsoft Surface Globe, Microsoft Surface Collage, Microsoft Surface Lagoon, Microsoft Blackboard, Microsoft Rebound and Microsoft Garden Pond touch screen software.
Do not forget, Microsoft Touch Pack for Windows 7 is one of the best free touch screen software you can find containing multi-touch games and multi-touch applications.
With PhotoShow Touch, Windows 7 users can drag and drop, scroll, tap, swipe and pinch photos using their fingertips by a touch on the touchscreen monitor easily.
PhotoShow Touch adds music, captions, styles and animated stickers among hundreds of samples to photos and enables Win7 users personalize their photos.
The Snowflake Suite plug-in for Windows 7 behaves as an additional layer on top of Windows 7 OS in order to enrich multi-touch capabilities of your computer.
Sorry it took so long to spot this question. The reason you"re not seeing anyone talking about the touchscreen function is that it has nothing to do with the screen and everything to do with primary vs secondary. however the touch screen is simply a mouse built into the monitor. The only reason it lines up with your finger is that it"s calibrated to do so. If you run multiple monitors with touch screen or run one with and one without neither screen can calibrate properly for both screen areas and the touch mouse is locked to the primary monitors logical location.
The only way I"ve seen this done effectively is to use the touchscreen as primary. We did have a customer try dual touch screens but he had major driver issues and eventually ended up disabling the feature on his secondary monitor.
The owner of this website (www.beetronics.com) has banned the autonomous system number (ASN) your IP address is in (20473) from accessing this website.
This article may have been automatically translated. If you have any feedback regarding its quality, please let us know using the form at the bottom of this page.
Note: As of April 11, 2017, Windows Vista customers are no longer receiving new security updates, non-security hotfixes, free or paid assisted support options, or online technical content updates from Microsoft. This article will no longer be updated and remains for information only. Please visit the Microsoft site for the full end of support statement.
The Dell SX2210T Monitor has touch screen capabilities. Drivers must be installed and the USB cable from the monitor connected to the computer for the touch screen feature to work. The touch screen can work either with your fingers or a stylus. The touch screen has many customizable options similar to other touch input devices. The option screens can be accessed from the Control Panel by clicking Pen and Touch.
Note: Not all versions of Windows Vista and Windows 7 will have all of the Pen and Touch options mentioned. This is a limitation of the version of the operating system. See the table below near the bottom of this document outlining the available touch features.
Flicks allow a user to perform common actions by simply using a short stroke of the stylus or finger. The Flicks tab allows a user to define the flick action and the sensitivity in which a flick action is registered (Figure 2).
The Touchtab has options that enable your finger to be used as an input device and enable gestures. In the Touch actions section are options to adjust double-tap speed and the press and hold time for right-click. Under the Touch pointer section is the option Show the pointer when I"m interacting with items on the screenwhich will cause a transparent image of a mouse to appear when tapping (Figure 4).
Panning can be enabled on this tab. The inertia setting allows for the screen to continue to scroll like spinning a wheel once your finger has been lifted. The inertia resistance setting can be increased which will cause the scrolling to stop sooner or decreased which keeps it moving longer (Figure 5).
Allows you to increase or decrease the magnification of the screen content/object. Moving two fingers apart on the touchpad zooms in on the active window.
Install the Touch Monitor drivers. Ensure they are installed by checking device manager under Human Interface Devices for NextWindow 1950 Touch Screen.
Run the diagnostics by opening the Control Panel and clicking Touch Screen Tools. Click the Run Diagnostic Test button and one of the following messages will be displayed:
Please clean the screen and if problem persists, call customer support. (If there is a blockage in the optical path, the diagnostic routine will display a red "X" on the Display Configuration window).
Clean the screen with a dry lint free cloth or use an approved monitor cleaner. Excessive smudging could cause the touch screen to not properly register inputs.
Run the calibration by opening the Control Panel and clicking Touch Screen Tools. Click the Calibrate button. The text on the Calibrate button will change to say Calibration in Progress and the mouse pointer will go to the upper left corner. Press and hold your finger still on the pointer until the pointer moves to the upper right corner. Repeat this process for all four corners. The button text will change back to say Calibrate and the mouse will then return to the center of the screen. The calibration will then be complete.
Most of us now have an iPad or some sort of tablet or smartphone. What’s more, we’ve probably tried using it to make beats, play synths, do some field recording or control our studio computer with a swipe-swipe of our fingers. We’re completely at home with the multi-touch screen environment. But we’re also perhaps grumbling at the size of the tablet/phone screen, quickly running out of processing power, and troubled by how best to integrate our devices into a larger studio setup. Conversely, on our desktop and laptop studio machines, we have far more power available and access to all the software tools we could wish for, but often find ourselves reduced to controlling them with a mouse.
Multi-touch technology has been very slow to make any headway in the world of desktop computing — and not just in terms of music production software, by any means. Surprisingly, perhaps, given the ubiquity of the iPad, Apple’s OS X doesn’t support multi-touch, but on ‘the other side’ it’s been available in some form or other since Windows 7. And, with high-quality, 24-inch, 10-point touchscreens now available for a very modest outlay(around £300), Windows 8 maturing through version 8.1, and Windows 9 on the horizon, some developers are now making significant progress. So, maybe it’s finally time to figure out how and where multi-touch-capable software could enhance your own recording studio, whether that be a modest home-studio setup, or something on a grander scale.
2. Infra-red technology creates an optical grid across the screen, and registers a ‘touch’ when the beams are interrupted. This is particularly suited to larger screens. It’s what the Microsoft PixelSense was based on, and it can be found today in Slate Digital’s Raven MTX. It benefits from great accuracy, not having to use annoyingly reflective glass, and the ability to register touches from any object (not just a finger). However, infra-red screens are vulnerable to accidental ‘touches’ from elbows, clothing, insects and so on.
3. Capacitive technology has risen to the top, primarily through Apple’s use of it. It’s durable, reliable and accurate, with a good resolution (although not as good as a stylus), and the price has come down significantly due to the sheer number of phones and tablets using it. Capacitive screens work by creating a minute electrical field, from which a capacitive object (such as a finger) draws current, creating a voltage drop at that point of the screen. Most phones and tablets now use a variant called Projected Capacitive Touch technology, which essentially doubles up the grid for improved accuracy and tracking, and also supports passive styli and gloved fingers.
Microsoft introduced native multi-touch support with Windows 7, but two things prevented it from really catching on: the expense of the hardware, and the fact that a desktop OS designed primarily for a mouse isn’t particularly suited to being operated with your fingers. In an act of pure genius (or madness, depending on your point of view) Microsoft then designed Windows 8 to offer the user two distinct interfaces: the Windows Start (formerly known as Metro) Modern UI, which is intended for touching, and the standard desktop, for use with a mouse and keyboard. It meant that the OS could be used on multiple devices: tablets, phones, laptops and desktops could all use it, and be touchable and futuristic in every environment.
Unfortunately, for the vast majority of people without a multi-touch interface, it was confusing and a bit frustrating, in that half of the OS couldn’t comfortably be used when all you had was a mouse! Microsoft addressed many of these concerns with Windows 8.1, and further updates have brought the Modern and the Desktop interfaces much closer together. The experience for the mouse and keyboard user is now much more that of an enhanced or augmented desktop, rather than a touch interface that was out of their reach. At the same time, sales of all-in-one computers with multi-touch screens, hybrid laptops and Surface tablets have grown at a healthy rate, and so multi-touch technology is fast becoming ‘normal’ on regular Windows computers.
So what does the Windows App Store offer in terms of music apps? Unfortunately, not a lot! There’s a sprinkling of piano and guitar-strumming apps, and some sample-triggering and remixing ones. There’s a DJ remix app from Magix called Music Maker Jam, which is pretty cool in a DJ-remixing-preset-samples kind of way, and a recording studio from Glouco, which does at least allow you to record via a mic input and sequence the included instruments. However, Windows Store apps have been hampered by two things when it comes to music production. Firstly there’s been no support for MIDI, and secondly, the output latency is only as good as the system’s soundcard using standard Windows drivers.
With Windows 8.1, Microsoft revealed a MIDI API (application programming interface), which for the first time allows a Modern app to use a MIDI input. Image Line took advantage of this in FL Studio Groove, a decent music-making app that lets you sequence, mix and mess about with a range of sample-based instruments and drums — and, of course, you can play the sounds from a MIDI keyboard.
The MIDI API is the first release from a new team of creative people at Microsoft who are working to improve the MIDI and audio aspects of Windows apps for the forthcoming Windows 9. This is unprecedented, and potentially very exciting for music makers. Pete Brown, from Microsoft’s DX Engineering Engagement and Evangelism department, said that they’re doing some serious engineering here, including on audio latency for Universal apps (apps that run across all Windows devices). They’re working with a lot of partners, big names and small, from both hardware and software worlds, to help prioritise, prototype apps, and so on. He tells me they’ve made sure that the approach is aligned with industry needs and requirements, and not just what Microsoft think needs to be done. It’s been the most open development process he’s ever seen, and he says the response from partners so far has been extremely positive.
The first thing to realise is that multi-touch is not mouse emulation, as it was on the old single-touch screens you might find in supermarkets or information kiosks. So, although it sometimes appears as though you’re just mousing about with your finger, it doesn’t always work as expected. In an application not designed specifically for touch control (we’ll call these ‘non-touch’), like Cubase, a single finger can access all the menu items, controls and parameters, just as with a mouse. But with some of the plug-ins you’ll find you can’t play the virtual keyboard unless you pull your finger across the notes, giving you a sort of Stylophone effect.
This problem is more obviously demonstrable in Adobe’s Photoshop: you can select all the tools and menu items, but you can’t actually use your finger on an image — nothing happens! Oddly, if you start with your finger off to the side of the page and then drag it onto the image, it then allows you to draw — but only as long as your finger stays in contact. Adobe say that this is because there’s no touch standard, and they already have their own APIs for use with Wacom’s pen and touchscreen products. They’re also waiting for Apple to join in the game, which, I fear, is something we’re going to hear from a lot of developers of software aimed at creative professionals.
Slight oddities aside, most DAW software actually works very well with single touches on a multi-touch screen. In my tests, Cubase, Pro Tools, Reason, Studio One, Reaper and Tracktion all happily let me poke around to my heart’s content. Ableton Live proved good for launching clips, entering notes and moving regions, but I ran into trouble when attempting to move parameters: once I’d grabbed a control with my finger, the knob or slider would zoom to the maximum or minimum value with the slightest finger movement without letting me easily set any value in between. Fortunately, there’s a fix for this. You have to create an ‘options.txt’ file in the Preferences folder, which lurks in the back end of the dusty reaches of your file system, and add the line “-AbsoluteMouseMode” (more precise information can be found at: www.ableton.com/en/articles/optionstxt-file-live). This allows all of the parameters to be moved much more smoothly. Unfortunately, I’m not aware of similar fixes for Bitwig Studio, which has exactly the same problem, or Digital Performer 8 (DP8), where some plug-ins exhibit this behaviour.
In the project/session/arrange window, with the exception of Live, Bitwig and DP8, all DAWs responded to the pinch/zoom two-finger gesture to either expand the track height or extend the timeline. Reaper even managed to do both directions at the same time. So, although these programs are not multi-touch compatible, there’s not that much you can’t do with your fingertip, assuming you can get your fudgy finger on the sometimes tiny knobs. You could load Reason, Live or Pro Tools onto a Windows 8 Pro tablet, such as the Microsoft Surface, and get on with making music without having to add a mouse to the equation.
Similarly, with stand-alone plug-ins and instruments, such as Native Instruments Reaktor or the Arturia Mini V, you’ve got single-finger control over all the parameters — the only problem is the Stylophone effect on virtual keyboards described earlier. Even though, especially on a tablet, you can go without a mouse, there’s no reason why you would want to do so on a desktop machine. The beauty with multi-touch on the desktop is that you can use everything. So, you can use your mouse and keyboard as normal, but perhaps when tweaking plug-ins or working closely in the arrange page you can simply reach out and touch it. Which is fantastic!
1. Console View: all the faders and pan, sends and other controls are all gloriously multi-touchable. You can use all your fingers, all at once, and mix to your heart’s content.
3. Matrix View: You can drag/drop samples, launch clips and move things about with as many fingers as you like, making this performance side of Sonar a real joy to use — Ableton and Bitwig really should take note.
Sonar struggles, though, when it comes to consistency. In the Pro Channel expansion to the Console view, the knobs in the EQ, Tube and Compressor respond to a rotary, sort of half-circle finger movement, but the knobs in the rest of the Pro Channel respond with an up and down movement. You can’t control more than one knob at a time, although you can move other controls on the console. Using the visual display on the EQ suffers from the same ‘zooming about’ problem found in Live, though they’ve dealt with this issue in X3 by sliding out a lovely large EQ window, which works perfectly with multi-touch.
Meanwhile, back in the arrange window, nothing is actually touchable! You can’t move any regions, cut them up or edit them in any way. Nor can you add notes to the piano roll or change any automation. All the arrange page things that can be done in the other non-touch-enabled DAWs can’t be done in the touch-enabled Sonar, which is a bit strange. The changes they made to the Pro Channel EQ shows exactly what’s required for touch to work effectively: you need big knobs. The Console view is hampered in places by the size and throw of the faders and some of the small controls. Cakewalk have built multi-touch into their existing GUI, and although it works well in some areas it also demonstrates why this might not be the best way to approach it — although, of course, you still have your mouse and keyboard.
StageLight has the standard arrange window, with tracks and a timeline, piano roll and automation, but it also has drum pads, a step sequencer, and a virtual keyboard that you can lock to various preset tunings, making it very easy to play all the right notes. In version 2 they’ve added in some nice-sounding synths and instruments, all with touchy parameters, and they’ve introduced an Ableton Live-style loop arranger, with large clip boxes to poke with your fingers. One very neat feature is that, through its support of VST plug-ins, it includes a multi-touch GUI version of the standard VST parameters window. It’s very simple, with each parameter displaying just a slider and a value, but it hints at what’s possible. Matthew Presley, Product Manager at Open Labs, mentioned how the right-click element of touch — where you hold until a menu appears — is something they’ve found frustrating. In refreshing the interface for version 2, they decided to get around that by creating a ‘Charms’-style toolbar at the side with all the editing tools, including a ‘Duplicate’ button, which takes the finger pain out of copying and pasting. Their core concern was to make it easy, so that people can just get on and make music.
At $10, there’s nothing really to touch StageLight. The pricing model is similar to that used in so many iOS and Android apps, and it’s something we’ll probably see much more of — the standard software is very cheap, or free, and then, through an in-app store, you can purchase additional features as you get more serious. It’s a refreshing change from the ‘Lite’ versions of DAWs we’re so familiar with, where you always wish you could afford the ‘real’ version just to get a little more functionality. StageLight is increasingly being pre-installed on many Dell, Lenovo and Acer Windows 8 tablets, and as kids these days are unlikely ever to possess an actual desktop computer, this might very well be where they start making music.
Ben Loftis, Product Manager for Mixbus, had this to say: “In a touch interface, you must accommodate calibration errors, parallax, and the splay of your finger. You don’t have any haptic feedback. So, if you want an analogue-console experience on a touchscreen, you will need a touchscreen that is larger than the analogue counterpart. But exactly how much larger depends on the hardware and the user. Currently, Mixbus v2 chooses between three sizes, based on your monitor resolution. But v3 will give the user an infinitely variable-scale slider, so we can accommodate more combinations of screen size and resolution. Also, our plug-ins (like the XT series) are arbitrarily scalable: if you stretch the plug-in window, all the knobs get bigger. We think this will be important for touch users, because many existing plug-ins use tiny buttons.”
The grandaddy of digital audio on the PC, Software Audio Workshop (SAW) Studio added multi-touch control to their Software Audio Console (SAC) live-sound mixer application as long ago as 2010. It was tied into the revolutionary 3M multi-touch screens, as favoured by Perceptive Pixel. Unfortunately, it hasn’t got any further than that, and still only supports multi-touch on these rather expensive screens. The layout of SAC lends itself brilliantly to multi-touch and currently works very well with a single touch — but it would be good to see this opened up to more current and cheaper technology.
It turned out that the development libraries they were using (JUCE C++) contained multi-touch features, which were primarily intended for the iPad, and that these had simply translated into the GUI of the VST version. This is true of all their plug-ins. They’ve now released a larger-GUI option for LuSH-101, for people with fat fingers.
I had a similar experience more recently with Arturia’s Spark 2 soft synth. Arturia told me that they hadn’t planned to make it multi-touch, but that they’re very happy that it is. So, the programming languages already exist to allow developers to include multi-touch functionality without specialist add-ons or tools — which means plug-in manufacturers may start to produce their multi-touch GUIs even while the DAW makers drag their feet.
The alternative to direct touch control of the DAW or plug-ins is to use touchscreen technology as a controller. The Jazz Mutant Lemur, the first commercially available multi-touch controller, has now evolved to become an iPad app, and there are now dozens of iPad apps for controlling DAW software via MIDI or OSC — in fact, there are even a few for Android. They’re selling well and there’s obviously a desire and use for it. The lack of haptic feedback (an actual physical knob or fader) doesn’t appear to be a barrier to most users, despite the recent Kickstarter campaign to manufacture knobs that you can stick onto the surface of the iPad (http://sosm.ag/ipad-knob-kickstarter).
James Ivey, Pro Tools Expert hardware editor (www.pro-tools-expert.com), who owns a Slate Pro Raven MTi controller, put to me the case for touchscreen controls over physical faders: “I was using a Euphonix [now Avid] Artist Control and Mix. So I had 12 faders to play with. With the MTi I have unlimited faders — what’s not to like? I really don’t buy into the “Oh, it’s not a real fader or pot” thing. I’m so much faster on the Raven. It’s big, it’s clear and if I don’t like something about the workflow or arrangement I can change it.”
Perhaps more of a barrier, then, is the physical size of the iPad, and the connectivity when away from the cosy security of your home network. With a touchscreen attached directly to your desktop you have none of the connectivity problems, because the screen is right there: attached via HDMI or DVI, it’s part of your system via a virtual MIDI driver. Although Windows tablets may suffer from the same size issues, hybrids, all-in-ones and dumb touchscreens don’t give you a proper console-sized surface to play with either. Probably the most important point is that the controller can ‘be’ anything — knobs, faders, pads, XY controls, you are not stuck to a hardwired configuration.
SmithsonMartin’s Emulator Elite is an awesome crystal-clear, projected capacitive, 10-point touchscreen that folds out into a beautiful sheet of glass. This is then rear-projected upon to create what looks like the ultimate futuristic DJ performance tool. At $15,500, the ‘Elite’ part of its name is apt. However, a rather more reasonable $99 buys you the screen’s core controller software for use with the desktop. CEO Alan Smithson is a DJ and fully admits that 90 percent of their focus is on the DJ market, but the capabilities of Emulator Pro extend far beyond controlling Traktor and offering performance tools.
Emulator comes with its own internal virtual MIDI driver, which makes it a complete doddle to hook up with Cubase, Pro Tools and other such applications. There are a number of different pages, so you could have mixer control on one and plug-in control on another, or you can add controls to container windows which fold down to a button and reopen when you need them.
Emulator Pro runs only in full-screen mode, but that doesn’t mean it has to obscure the DAW: a feature that’s particularly useful for single-screen setups is the ability to ‘cut holes’ out of its GUI, so that the software running beneath is visible through it. That may be useful to reveal meters, a preview screen, or the arrange page, for instance — the possible configurations are endless.
If Emulator Pro is found lacking anywhere, it’s in the depth of the MIDI side of things. Channel and controller numbers is as far as it goes, so it can’t send SysEx commands or emulate a Mackie HUI, for example. However, Shane Felton (of www.alien-touch.com) has been working on an implementation to get 24 channels of Mackie HUI Control into Pro Tools running on his Apple Mac. The result looks not unlike the Slate Raven MTi, and includes many of the same shortcut buttons and controls. He uses Bome’s MIDI Translator to provide the HUI emulation and three virtual MIDI drivers (one for each group of eight faders) that are setup in Pro Tools. The template files are available to download from his web site, though he stresses that it’s a work in progress and would value contributions.
Back in the world of Usine Hollyhock, we find all sorts of depth, function and parameter possibilities. It has its own ‘add-on’ community, where you can find all sorts of control scripts and patches. Whether it’s SysEx, OSC or even Mackie Control emulation, you’ll find that someone has created a patch for it. Usine Hollyhock has the potential to be an extremely versatile controller. It doesn’t quite have the design simplicity or focus of Emulator (since it’s much more than a controller) but it has far more depth and breadth in terms of control potential.
CopperLan is a networking protocol that connects compatible music software and hardware together. Each device can reveal its parameters by name and be controlled by any other device automatically. It’s a bit like MIDI control, but at a much higher resolution and without all that manual mapping and learning you have to do. There are wrappers for non-compatible plug-ins, but for these you have to manually configure the controls. A CopperLan-compatible touchscreen controller could potentially map itself automatically to whatever CopperLan-compatible plug-in is selected. It can also work internally, without the need for a network, which makes it such an interesting solution for a virtual controller running on the same machine as your DAW.
The DTouch mixer is essentially the fader section of Pro Tools’ mixer window, with cut-outs around the meters so that they shine through. Once the alignment is set up, the design is flawless and you wouldn’t know you were using anything other than the Pro Tools mixer. The toolbar provides all the usual transport controls as well as buttons to activate groups, open selected plug-ins and such like. In the edit window, although no multi-touch controls are overlaid, you get an expanded toolbar full of useful tools and functions. There’s also a load of buttons to which you can assign your own macros. The toolbar allows you to zoom around and perform edits without having to return to the mouse, which is what makes the workflow so effective.
At the time of writing, DTouch was available for Windows 7 only, Pro Tools only, and at a mandatory resolution of 1920 x 1080. However, Devil Technologies were kind enough to let me try their Windows 8.1 beta version, which should be available by the time you read this. One side-effect of the alignment and tight integration is that it’s not very flexible — there’s no ability to edit the controls or create knobs and faders for other things as there is with Emulator and Hollyhock. Instead, its beauty lies in the seamlessness with which it functions alongside the DAW.
I asked the company about the possibility of releasing a generic HUI-based controller, but they tell me they would much prefer to do something that’s designed for the specific DAW — and, encouragingly, they have a Cubase/Nuendo version in the works already. They are also testing out ways to support two screens; currently you have to have everything on the single touchscreen monitor, or the alignments start to shift. Having said that, I tried it over two screens and found that it can work very well, especially on the mix window. But I guess you are back to selecting clips in the edit window with the mouse if you’ve moved it to a non-touch screen.
One unique feature is the ability to incorporate an external, hardware HUI-compatible controller alongside DTouch, to give you the best of all worlds. At 200 Euros it’s more expensive than other, more flexible options I’ve discussed, but it’s a no-nonsense dedicated solution that’s supremely good at what it does.
Let’s get back to the DAW manufacturers. We know the multi-touch support is patchy at the moment, but what of their plans for the future? There appears to be a rough split between software manufacturers for whom multi-touch is already a priority and those who, though interested, are focusing their effort on other priorities. It seems that most of the major firms are the least interested in implementing multi-touch, though. Of course, I should qualify this with the caveat that most major developers don’t like dropping hints about anything that’s unreleased — but Steinberg, for example, went as far as to state that they don’t currently see a market for multi-touch outside the iPad.
The industry’s strong focus on the iPad in recent years is certainly one reason why we see so little development on desktop multi-touch. The other major one is the absence of any lead from Apple into multi-touch support in OS X. It’s not hard to foresee the increasing convergence of iOS and OS X in the medium term, but a fully multi-touch-capable version of OS X still seems a long way off. This, to me, seems a little short-sighted.
James Woodburn, CEO at Tracktion, had this to say: “Eighteen months ago, we were asked regularly when Tracktion would be available for iPad — in fact, we already have a ported version of Tracktion that runs on iOS, but we chose not to release it, as it makes no sense to us to run a full DAW on a limited platform. We would rather design a solution that utilises the iPad’s key benefits and does not expose the limitations The demand, at least from our user base, for iPad support has really dropped off a cliff in the past 12 months — and demand for PC touch is on the rise, albeit quite slowly.”
The key to unlocking multi-touch on the desktop is in the design and implementation of the interface. DAW software has to perform lots of different and precise tasks, some of which lend themselves to touch, but many more of which would be hampered by the fatness and inaccuracy of human fingers. It just makes no sense to build touch into something that would be better accomplished with a mouse. There are also issues with hands and fingers masking the very controls you’re trying to fiddle with.
Bremmers Audio Design have been working with touchscreens for many years, and in MultitrackStudio they’ve developed a neat pop-up window that materialises a couple of inches above what you have your finger on. That means you can see both the control and the value. Even with this function, trying to edit the score window accurately with a fingertip is truly an exercise in futility — but it’s no problem because that’s why we have a mouse. With the iPad it must all be about touch, but with the desktop you can use each and every tool at your disposal, be it single-touch, multi-touch, gestures, stylus, mouse, keyboard, trackpad, Leap Motion, Kinect, hardware controllers, or something else entirely. The desktop should remain an awesomely creative and versatile place.
Spending time over the last few weeks rummaging around in the world of touch-enabled software, I’ve created a bit of a personal wishlist of features. Let’s hope some developers are reading!
In a DAW, I don’t want to be restricted to touch any more than I want to be restricted to a mouse. I want touch controls to become available when I need them — like the way the Pro Channel EQ slides out in Sonar (their mixer needs to do something similar). I want to be able to pinch/zoom into a region and then draw in automation with my finger; but I don’t necessarily want to have to use my fingers to copy and paste, trim audio or move notes.
Scrolling has to be easy, perhaps gesture-based, so that I don’t have to fudge around trying to finger empty space between controls to move the GUI. But I don’t want to have to return to the mouse just to move the screen a bit. Reason has that neat side-panel that shows a zoomed-out version of the rack — which is perfect for finger scrolling. Once you start adding hardware into the equation, a touch-screen makes for a much less jarring experience than putting your hand back on a mouse. In using Arturia’s Spark 2 with the Spark LE controller, it’s so great to be able to simply tap on the screen to change a parameter, preset or sample — it’s a far more fluid experience than moving from a creative hardware place to that mouse zone. It’s also done my RSI no end of good!
In terms of virtual control, which is perhaps more useful, because you could use it with various different bits of music software, a multi-touch screen holds enormous potential. It could be anything, could control anything. Imagine your 24-inch multi-touch screen set before you like a mixer, but placed physically beneath your main (non-touch) screen, and whenever the focus changes to the mixer, or plug-in or instrument, the touchscreen evolves to display the appropriate controls. Something a bit like Novation’s Automap that automatically pulls out the parameters and lays them out in front of you. CopperLan seems to hint at this sort of power, but requires everyone to be compliant for it to reach its potential. That’s the sort of integration we really need.
Hopefully, as more manufacturers realise the advantages of the desktop platform for multi-touch interfaces, they will use it to enhance our music-making environments. With Windows 9 just around the corner, Universal apps and Microsoft’s new-found interest in music production, there’s very little competition and great buckets of processing power and potential for the software developer, which can only mean great things for we touchy-feely users.
The history of touch technology can be traced back to the touch-sensitive capacitance sensors of early synthesizer pioneers such as Hugh LeCaine and Bob Moog — and it’s interesting to note that Apple’s iPad shares the basis of its touch technology with the humble Theremin! The concept of the multi-touch screen was first realised in the 1980s by Bell Labs, but probably made its way into the public consciousness through sci-fi films and series such as Dillinger’s desk in Disney’s Tron (1982) and Star Trek — The Next Generation (1987-94).
In computing terms, multi-touch refers to the ability of a surface to recognise the presence of more than one point of contact. This is distinct from single-touch interfaces, which essentially emulate the mouse input, and moves us through the world of pinch/zoom and gesture control, with which everyone’s familiar, to the possibility of individual touches creating individual actions and responses simultaneously.
The technology that we know today evolved out of a few sources: Fingerworks, a gesture-recognition company who pioneered a number of touchscreen products and were bought by Apple in 2005; Jeff Han’s Perceptive Pixel (bought by Microsoft in 2012) who, back in 2006, were demoing vast multi-touch walls and dazzling us with the concept of pinching photos and swiping maps; and the original Microsoft Surface, now called PixelSense, which started development in 2001 and was an interactive table that combined multi-touch capability with real-world object interaction.
It’s remarkable that in 2005 JazzMutant developed their own multi-touch technology to release the Lemur multi-touch OSC controller commercially. In 2007, with the release of the first iPhone and, a few months later, Microsoft’s Surface (PixelSense) 1.0, we had both ends of the multi-touch spectrum spectacularly catered for. But it would take a few more years for that middle space of tablets, hybrid laptops and multi-touch monitors to really find their technology and pricing sweet spots.
This is a very good question and a hard one to answer! The manufacturers are all over the place, with dual-touch being marketed, confusingly for the end user, as multi-touch, and screens designed for Windows 7 being pushed for Windows 8.
Windows 7, which is still favoured by so many music-makers, supports two-point multi-touch out of the box — so gestures and pinch/zoom all work fine. More points are supported with an additional download, but there was not really any part of the Windows 7 OS that made use of it — so your mileage using multi-touch on that platform will be almost entirely down to the software you’re running.
If designed for Windows 8.x, multi-touch monitors must have at least five simultaneous points. Many all-in-one machines meet only this minimum requirement, whereas hybrids and tablets tend to have 10. Quite honestly, having played with multi-touch for music making for a while now, I’ve rarely found myself using more than two touch-points at once, although sometimes I’ve used up to eight when messing about on a mixer to see what I could do. That said, a 10-point screen is more likely to be a projected-capacitance type, and so of a higher quality than screens offering fewer points.
I chose the Acer T232HL for my own use, because at the time it was the only thing available from the new generation of multi-touch screens. Dell kept promising one, but it kept getting delayed and finally came out about six months after I got the Acer. The Acer remains well regarded, particularly for its ability to lay almost flat, so it was a good choice in that respect.
The biggest gripe I have with multi-touch at the moment is with touch latency — my Acer T232HL adds about 10-15ms (an estimate, having tried playing drum pads, and so on). My understanding is that performance in this regard is rather better on tablets such as the Microsoft Surface, but I’ve not had a chance to test that, so whether, for example, it’s better enough for you to play drums without the latency proving a distraction,z I can’t yet say. Unfortunately, the published specs won’t help: the ‘Response’ time listed in a monitor’s specification usually refers to the change from black to white, and there’s no documentation on the touch response.
We music makers are a greedy bunch. While the rest of the world is moving to ever smaller and more portable devices, we seem to be accruing acres of screen real estate as we hook up multiple monitors to our DAW computers. One question that arose during the course of my research, therefore, was whether you could use two multi-touch screens concurrently with a single computer. This could give you greater access to more parameters spread across a larger desktop or, perhaps more interestingly, allow two people to work on different screens on the same project — for instance, one controlling the mixer, while another controls virtual instruments or effects.
It was a similar story with music applications: when trying to stretch Sonar X3’s mixing console across both screens, for example, the same issue of which screen held the focus arose. The process for grabbing focus is very obvious when you have the screens at different resolutions, because the console jumps in size as you move it between the screens, and that jump is the screen grabbing focus.
However, if I split different tasks out to different screens, the results were more intuitive. Moving Sonar’s mixing console to one screen and its project window to the other allowed me to use them simultaneously with no trouble at all. So, you could have one person mixing levels while the other triggers loops in the Matrix, all in one project on one computer. The bottom line is that multiple multi-touch monitor setups are feasible for music-making, and different applications can be controlled at the same time from some different screens, but I suspect a little trial and error will be required to arrive at the best workflow for each application.
I need to replace my PC setup at home, so your article on buying a new family PC was really great for me. Currently the PC is only used by the children for accessing the web, running Minecraft, iTunes, playing The Sims etc. I would really like to try using a touchscreen monitor to get the best out of Windows 8. I am aware of the argument about gorilla arms, but after using an iPad, I find myself prodding all computer screens with an (unrealistic) expectation that something should happen.
There’s a lot of choice for people who want all-in-ones, but I would much rather get a tower to leave the option of potentially upgrading the PC. I am a developer and might use the family PC for development in the future.Dave
You can add a touch-sensitive screen to any PC – or even an old laptop – by buying a touch-sensitive monitor. There must be a market for them, because most leading monitor suppliers offer them. This includes Acer, AOC, Asus, Dell, HP, Iiyama, LG, Samsung and ViewSonic. The less well-known HannsG also has competitive offerings.
However, touch sensitivity requires extra technology, which is an extra cost, especially for large screens. Touch-sensitive monitors are therefore more expensive than traditional designs, which must restrict the size of the market.
As you have found, there are lots of all-in-one PCs with touch screens, but they are basically laptop designs with separate keyboards. Slimline designs impose thermal constraints on the processor, which will typically operate at a TDP between 15W and 35W, or less. The processor will be throttled when it gets too hot, and the PC may shut down. By contrast, spacious desktop towers can use processors that run at 45W to 90W or more, so you get more performance for less money.
Towers provide space for adding more memory, ports, faster graphics cards, extra hard drives, optical drives (DVD or Blu-ray) and so on. They are also much easier to repair, so they should last longer. The main drawback is that they take up more space than laptops or all-in-one designs. This may be critical if you want to mount the screen on a wall, which is common with touch-screen PCs used for public information access.
You must consider the flexibility of the design. While the “gorilla arm” argument is simplistic to the point of stupidity – teachers have been using blackboards for centuries – there are important considerations to do with screen distance and angle.
The better all-in-ones provide flexibility to handle different programs and different uses. Often the screen leans back, and in some cases, can be used in a horizontal position. This makes it practical to play electronic versions of family board games, navigate around maps, play a virtual piano, and so on.
Desktop monitors are usually designed to be used with the screen in a vertical position, and relatively high up. This puts the screen a long way from your hands, so you are less likely to use it for touch operations. This contrasts with using a laptop, where the screen may be as handy as the keyboard.
If you decide to go for a touch-screen monitor, choose one that is easy to tilt backwards and possible to use in a horizontal position. Obviously, you should be able to return it to an upright position for word processing and so on.
Alternatively, you can buy any touch screen you like, if you mount it on a monitor arm that enables the screen to be moved around. This may actually be a better option, but it will probably cost more.
Touch-screen monitors are a bit more complicated than traditional designs, because they are active rather than passive devices. Traditional screens just have to show a picture, whereas touch-screen monitors have to feed information back to the PC. They often do this via a separate USB cable that runs next to the VGA/DVI/HDMI/etc video cable.
Monitors also vary according to the number of touch-sensitive points. This can range from five to 40, but 10 is usual for Windows 8. Further, different monitors may use optical, resistive or capacitative touch technology. Capacitative touch provides the same experience as using a tablet, which is what you want.
Some monitors support a new standard: MHL (Mobile High-definition Link). This enables you to connect a compatible smartphone or tablet to the monitor to show videos with high-resolution sound (up to 7.1 channels, including TrueHD and DTS-HD). The mobile device gets charged while it’s attached.
Other considerations are the usual ones: screen size and resolution, brightness, type of technology (LED, IPS etc), number of ports, whether it includes loudspeakers, and so on. Since you’re a developer, you’ll probably want to knock out a quick spreadsheet to compare all the options.
Note that touch-screen monitors designed for Windows 7 – probably with two touch-points – are less than ideal for Windows 8, where the bezel has to be flush with the display for edge-swipes. However, I don’t expect there are many Windows 7 touch monitors still on the market.
I have very little experience of different touch-screen monitors, and haven’t tested any, so you will need to do your own research. I can point to some of the products that are available, but unfortunately it may be hard or impossible to see them before you buy one.
PC World, for example, only seems to offer three touch-screen monitors. These are all Acer models with Full HD resolution (1920 x 1080 pixels) and screen sizes of 21.5in (£179.99), 23in (£249.99) and 27in (£379.99). These have MHL support, USB 3.0 and tilt stands that adjust from 80 to 30 degrees, so you could do worse. The 23in IPS-screen Acer T232HLA looks like the best option.
From my Amazon searches, the ViewSonic TD2220 looks like an economical option at about £180. It’s a 22in Full HD display. However, the 23in HannsG HT231HPB is slightly cheaper (£157.95), and Amazon reviewers give it 4.6 out of 5 stars.
Other touch-screen monitors that might be worth a look include the 23.6in AOC Style i2472P (£262.98), the 21.5in Dell S2240T H6V56 (£207.38) and the 23in Dell S2340T (£339.95). There’s also a ViewSonic TD2340 for £199.99, apparently reduced from £439.99, and a 24in Samsung S24C770TS for £449.99.
If you have a modern Windows 8 laptop, then you can probably use Windows 8’s touch gestures on its built-in touchpad. In the same vein, you could just buy a touchpad for your desktop PC and use it with a cheaper non-touch screen. Logitech’s rechargeable Touchpad T650 is an expensive option at £114, though the wireless T650 looks a better buy at £39.99.
If you travel with a laptop and iPad, you need this app. I needed a second screen, but Duet gives me even more. Full gesture support, customizable shortcuts, Touch Bar, tons of resolution options, and very little battery power. How is this all in one app?
I just love this app. Especially when I am travelling for work an working from the company branches. Then I use my iPad as second monitor for Outlook, Lync and other chat while I use the laptop big screen for remote desktop to my workstation at the main office. :)
Yes, Hope Industrial touch screens use resistive technology, which is pressure-sensitive and can be used with any type of stylus, as long as it is not sharp or rough (which could damage the touch screen surface). Please contact our sales department for more information.
After initial setup, the touch screen should not require periodic re-calibration. Installing new drivers could erase a previous calibration and at time re-calibration is done by preference since some users prefer a different calibration style (e.g.: pointer centered on the finger-tip vs. centered on the finger).
Drivers are available for a wide variety of current and legacy operating systems, including all major Windows versions, Microsoft Embedded operating systems, Apple Mac, and many others. A full list and downloadable drivers are available in our support section.
Yes. Our touch screen drivers allow multiple displays to be connected to a single PC whether in mirroring mode (multiple screens showing the same desktop) or extended desktop (a single desktop stretched across multiple displays).
Our Windows drivers allow each touch screen to be calibrated independently whether you are using USB or Serial for connection to the PC. Once configured, the cursor will follow your finger to any connected touch screen. For configuration help or more information, please contact our support group for assistance.
Support for Linux-based operating systems is available through both native drivers, and by using driver-less methods that rely on the HID device compatibility of our touch screens. A full review of the available methods is available on our blog.
By clicking on the link, you will be leaving the official Royal Philips Healthcare ("Philips") website. Any links to third-party websites that may appear on this site are provided only for your convenience and in no way represent any affiliation or endorsement of the information provided on those linked websites. Philips makes no representations or warranties of any kind with regard to any third-party websites or the information contained therein. I understand
Touch panel technologies are a key theme in current digital devices, including smartphones, slate devices like the iPad, the screens on the backs of digital cameras, the Nintendo DS, and Windows 7 devices. The term touch panel encompasses various technologies for sensing the touch of a finger or stylus. In this session, we"ll look at basic touch panel sensing methods and introduce the characteristics and optimal applications of each.
Note: Below is the translation from the Japanese of the ITmedia article "How Can a Screen Sense Touch? A Basic Understanding of Touch Panels"published September 27, 2010. Copyright 2011 ITmedia Inc. All Rights Reserved.
A touch panel is a piece of equipment that lets users interact with a computer by touching the screen directly. Incorporating features into the monitor like sensors that detect touch actions makes it possible to issue instructions to a computer by having it sense the position of a finger or stylus. Essentially, it becomes a device fusing the two functions of display and input.
It"s perhaps not something we think of often, but touch panels have integrated themselves into every aspect of our lives. People who enjoy using digital devices like smartphones interact with touch panels all the time in everyday life—but so do others, at devices like bank ATMs, ticket vending machines in railway stations, electronic kiosks inside convenience stores, digital photo printers at mass merchandisers, library information terminals, photocopiers, and car navigation systems.
A major factor driving the spread of touch panels is the benefits they offer in the way of intuitive operation. Since they can be used for input through direct contact with icons and buttons, they"re easy to understand and easily used, even by people unaccustomed to using computers. Touch panels also contribute to miniaturization and simplification of devices by combining display and input into a single piece of equipment. Since touch panel buttons are software, not hardware, their interfaces are easily changed through software.
While a touch panel requires a wide range of characteristics, including display visibility above all, along with precision in position sensing, rapid response to input, durability, and installation costs, their characteristics differ greatly depending on the methods used to sense touch input. Some typical touch-panel sensing methods are discussed below.
As of 2010, resistive film represented the most widely used sensing method in the touch panel market. Touch panels based on this method are called pressure-sensitive or analog-resistive film touch panels. In addition to standalone LCD monitors, this technology is used in a wide range of small to mid-sized devices, including smartphones, mobile phones, PDAs, car navigation systems, and the Nintendo DS.
With this method, the position on screen contacted by a finger, stylus, or other object is detected using changes in pressure. The monitor features a simple internal structure: a glass screen and a film screen separated by a narrow gap, each with a transparent electrode film (electrode layer) attached. Pressing the surface of the screen presses the electrodes in the film and the glass to come into contact, resulting in the flow of electrical current. The point of contact is identified by detecting this change in voltage.
The advantages of this system include the low-cost manufacture, thanks to its simple structure. The system also uses less electricity than other methods, and the resulting configurations are strongly resistant to dust and water since the surface is covered in film. Since input involves pressure applied to the film, it can be used for input not just with bare fingers, but even when wearing gloves or using a stylus. These screens can also be used to input handwritten text.
Drawbacks include lower light transmittance (reduced display quality) due to the film and two electrode layers; relatively lower durability and shock resistance; and reduced precision of detection with larger screen sizes. (Precision can be maintained in other ways—for example, splitting the screen into multiple areas for detection.)
Capacitive touch panels represent the second most widely used sensing method after resistive film touch panels. Corresponding to the terms used for the above analog resistive touch panels, these also are called analog capacitive touch panels. Aside from standalone LCD monitors, these are often used in the same devices with resistive film touch panels, such as smartphones and mobile phones.
With this method, the point at which the touch occurs is identified using sensors to sense minor changes in electrical current generated by contact with a finger or changes in electrostatic capacity (load). Since the sensors react to the static electrical capacity of the human body when a finger approaches the screen, they also can be operated in a manner similar to moving a pointer within an area touched on screen.
Two types of touch panels use this method: surface capacitive touch panels and projective capacitive touch panels. The internal structures differ between the two types.
Surface capacitive touch panels are often used in relatively large panels. Inside these panels, a transparent electrode film (electrode layer) is placed atop a glass substrate, covered by a protective cover. Electric voltage is applied to electrodes positioned in the four corners of the glass substrate, generating a uniform low-voltage electrical field across the entire panel. The coordinates of the position at which the finger touches the screen are identified by measuring the resulting changes in electrostatic capacity at the four corners of the panel.
While this type of capacitive touch panel has a simpler structure than a projected capacitive touch panel and for this reason offers lower cost, it is structurally difficult to detect contact at two or more points at the same time (multi-touch).
Projected capacitive touch panels are often used for smaller screen sizes than surface capacitive touch panels. They"ve attracted significant attention in mobile devices. The iPhone, iPod Touch, and iPad use this method to achieve high-precision multi-touch functionality and high response speed.
The internal structure of these touch panels consists of a substrate incorporating an IC chip for processing computations, over which is a layer of numerous transparent electrodes is positioned in specific patterns. The surface is covered with an insulating glass or plastic cover. When a finger approaches the surface, electrostatic capacity among multiple electrodes changes simultaneously, and the position were contact occurs can be identified precisely by measuring the ratios between these electrical currents.
A unique characteristic of a projected capacitive touch panel is the fact that the large number of electrodes enables accurate detection of contact at multiple points (multi-touch). However, the projected capacitive touch panels featuring indium-tin-oxide (ITO) found in smartphones and similar devices are poorly suited for use in large screens, since increased screen size results in increased resistance (i.e., slower transmission of electrical current), increasing the amount of error and noise in detecting the points touched.
Larger touch panels use center-wire projected capacitive touch panels in which very thin electrical wires are laid out in a grid as a transparent electrode layer. While lower resistance makes center-wire projected capacitive touch panels highly sensitive, they are less suited to mass production than ITO etching.
Above, we"ve summarized the differences between the two types of capacitive touch panels. The overall characteristics of such panels include the fact that unlike resistive film touch panels, they do not respond to touch by clothing or standard styli. They feature strong resistance to dust and water drops and high durability and scratch resistance. In addition, their light transmittance is higher, as compared to resistive film touch panels.
On the other hand, these touch panels require either a finger or a special stylus. They cannot be operated while wearing gloves, and they are susceptible to the effects of nearby metal structures.
Surface acoustic wave (SAW) touch panels were developed mainly to address the drawbacks of low light transmittance in resistive film touch panels—that is, to achieve bright touch panels with high levels of visibility. These are also called surface wave or acoustic wave touch panels. Aside from standalone LCD monitors, these are widely used in public spaces, in devices like point-of-sale terminals, ATMs, and electronic kiosks.
These panels detect the screen position where contact occurs with a finger or other object using the attenuation in ultrasound elastic waves on the surface. The internal structure of these panels is designed so that multiple piezoelectric transducers arranged in the corners of a glass substrate transmit ultrasound surface elastic waves as vibrations in the panel surface, which are received by transducers installed opposite the transmitting ones. When the screen is touched, ultrasound waves are absorbed and attenuated by the finger or other object. The location is identified by detecting these changes. Naturally, the user does not feel these vibrations when touching the screen. These panels offer high ease of use.
The strengths of this type of touch panel include high light transmittance and superior visibility, since the structure requires no film or transparent electrodes on the screen. Additionally, the surface glass provides better durability and scratch resistance than a capacitive touch panel. Another advantage is that even if the surface does somehow become scratched, the panel remains sensitive to touch. (On a capacitive touch panel, surface scratches can sometimes interrupt signals.) Structurally, this type of panel ensures high stability and long service life, free of changes over time or deviations in position.
All in all, however, these touch panels offer relatively few drawbacks. Recent developments such as improvements in manufacturing technology are also improving their cost-performance.
The category of optical touch panels includes multiple sensing methods. The number of products employing infrared optical imaging touch panels based on infrared image sensors to sense position through triangulation has grown in recent years, chiefly among larger panels.
A touch panel in this category features one infrared LED each at the left and right ends of the top of the panel, along with an image sensor (camera). Retroreflective tape that reflects incident light along the axis of incidence is affixed along the remaining left, right, and bottom sides. When a finger or other object touches the screen, the image sensor captures the shadows formed when the infrared light is blocked. The coordinates of the location of contact are derived by triangulation.
While this type differs somewhat from the above touch panels, let"s touch on the subject of electromagnetic induction touch panels. This method is used in devices like LCD graphics tablets, tablet PCs, and purikura photo sticker booths.
This input method for graphics tablets, which originally did not feature monitors, achieves high-precision touch panels by combining a sensor with the LCD panel. When the user touches the screen with a special-purpose stylus that generates a magnetic field, sensors on the panel receive the electromagnetic energy and use it to sense the position of the pen.
Since a special-purpose stylus is used for input, input using a finger or a general-purpose stylus is not possible, and the method has limited applications. Still, this has both goo