Show HN: Check out the 23" Android 'tablet' I made for $600 | Hacker News Reader

Show HN: Check out the 23" Android 'tablet' I made for $600

martin.drashkov.com

160 pointsmartythemaniak55 comments

I want to post a longer and more detailed post in a few days, but at a high level, this is a TI Pandaboard (dual core 1GHz, 1GB RAM) running Gingerbread attached to a Quanta touchpanel (the Acer T230H in this case). The monitor costs around $330, the pandaboard around $175 and you'll need a few other things like a power adapter etc.

The setup is a bit laggy in parts for different reasons. Google earth shows a lot of IO errors, while the launcher just wasn't made to push that many pixels around - taking the resolution down gets rid of it. On the other hand, once the GPU takes over in Fruit Ninja, everything works fine and is actually quite a lot of fun.

Like I mentioned in the post, the size of this thing makes simultaneous use by 2 people practical - Fruit Ninja is a lot of fun with 2 people and games like Flight Control will work very well too.

Since multi-touch is disabled in Android (in the US) due to patent issues, would this harm the ability to have simultaneous use?

It's not disabled. At least not on any Android device I've tried within the last 2 years. Pinch-to-zoom has worked since 2008 IIRC.

US multi-touch has been enabled for a couple of years now. I don't know the specifics, I imagine more cronyistic backroom patent deals that typify our broken patent system.

About 1.5 years ago did a college project with one of these things. Is the Quanta touchpanel an optical one? Are there lenses in each of the top corners? If so, we had a similar screen. We had to get it to work with Linux, which back then was a tough due to kernel not quite having driver support (believe it's included since kernel 2.8.32, partially in .31) It's not true multitouch because somewhere halfway the infrared cameras can get confused, so you can't really use more than two points, though sometimes you can use 4. Benefits are however that you don't have to actually touch the screen but can hover slightly over it. For the project we deployed a location tracking system on it, using a custom Open Street Maps OpenGL app with markers, manipulated using the multitouch.

Where this would be most excellent would be for photo or video editing - particularly once they quad cores come out.

How about for textbooks or schoolbooks? This is at a size that I could actually see myself using it, if it were something I could lay flat on a desk. It would even be practical for taking notes, I think.

Maybe it's just that I'm damaged too much by resistive touchscreen terminals (rural startup flashbacks…), but I wonder whether this really will be the future. Someone at the BUILD conference mentioned that we'll be looking back in wonder that there was a time when monitors didn't have touch input. On the other hand, didn't we try stuff like that quite a few years ago? If not touchscreen, then lightpen input… "Gorilla arm syndrome" comes to mind.

But just like the recent wave of 3D movies, I'm not sure whether it's a case of "the technology caught up" or "those who don't know history…". Zeus, I sound positively luddite here. Don't get me wrong, I really like my ipad, but the applications, the form factor -- the whole use case is quite different. Well, let's see… Wonder if I'm repeating the words of command line geeks after they first saw a Macintosh in '84…

Didn't want to digress from the post that much. It is a cool project, martythemaniak. With cheap enough screen and support hardware, we'll probably see a lot more terminals in the wild. A bit of sheet metal and plastic, a portrait '23 inch screen and some custom Android-based firmware and you could make some great info and tourism terminals. And if no one's working with them, you'd have a reasonably large area to show ads.

Or VESA mount it, and you'd have a great info panel similar to the Panic Status Board[1], with some direct interaction possibilities.

[1]: http://www.panic.com/blog/2010/03/the-panic-status-board/

i felt the same thing but my concern was elsewhere.

my concern is with the "touching" itself. it didn't seem to me that we have achieved what were described in great sci-fi books and movies. it should be this way: a virtual pad/keyboard/etc should popup in the air so you never have to "touch" the actual screen :)

it's the input device that should be improved, not the screen. let's dream about it :)

kinect kinda got it.

We had those laser-based mobile "virtual keyboards", but for something really interesting we'd have to improve the projection technolgy, i.e. full-fledged "holograms", which would apply for both the screen and the input device (if the quality is more "Leia"-like, then the physical screen will stay a while longer). I don't think we're there yet.

But I fully expect to see Wiimote/Kinect-like 3D gesture technology to be incorporated into devices, probably soon after the devices themselves have a (pseudo-)3D display. Tablets already have front cameras, shouldn't be a big jump hardware-wise.

I would imagine large touch screen becoming more ingrained in our every day life, but not necessarily our personal computing devices. Our desks, replacements for cork-boards, clip boards, etc.

Or maybe; something like a drafting table combined with voice input and kinect-like tracking would be pretty powerful (and cheap, within a couple years).

I doubt that either touch or a kinect-style will become primary interfaces. However, I think as the tech develops further and the price drops it will become ubiquitous as a feature.

I can't see how a vertical monitor is helpful. Some of the touch screen PCs that HP and Lenovo have launched tilt back so you can use your hands without gorilla arm syndrome.

I imagine the bigger hurdle isn't the tech, but selling people on horizontal monitors that they're touching all day and the lack of a keyboard (or keyboard as non primary input).

I feel like we're stuck in this middle land of tablets without kickstands and PC monitors that can't turn into horizontal touch screens. I imagine there's a compromise here that'll work. Maybe all-in-one desktops that tilt with hide-away keyboards and tablets with built-in stands and perhaps even little pens for when you're using an app not optimized for tablets or if you need to draw or mark up a document.

Someone will come along in the next 2 years and do haptic feedback right and hardware keyboards will go the way of the floppy drive. There's been a number of patents recently on this front from basic haptics to displays with controllable surface texture.

No.

On even a netbook keyboard I can type at pretty much full dictation speed if I have to, with high accuracy. I can feel if I've made a mistake, I can feel if I've hit two keys rather than one or if my hands are drifting off centre.

I am utterly, utterly unconvinced that that level of performance is going to be available from a haptic touchscreen any time, particularly at a cost comparable with keyboards I can buy for less than the cost of lunch. The patents may well be interesting and have all sorts of beneficial applications, but wholesale replacement of computer keyboards by them isn't one.

So with haptics you should be able to:

(1) type at pretty much full dictation speed if you have to (some people can already do this)

(2) Feel if you've made a mistake.

(3) Feel if you hit 2 soft keys rather then one if your hands drifted.

I don't see what's so unbelievable then; even today the physical keyboard has been removed from the most successful new devices. That's without haptics. Long term I suspect the keypress paradigm itself will fall to gestures which you don't even need a screen or keyboard, a camera can track your finger movements.

This particular panel is optical touch - inside the bezel it has IR projectors at the top and sensors on the bottom. It is not as good as a capacitative panel, but far, far better than resistive ones.

One reason I like the ASUS Transformer is that it shows, IMO, that touch and keyboard/mouse can be complimentary. As prices continue to fall and touch becomes ubiquitous, I think we'll see more and more of these.

Ergonimics can also go a long way - a large monitor that's sitting at a ~25 degree angle will likely eliminate the bulk of the "gorilla arm" problem.

As much fun as giant touch screens look (no thanks to minority report) they're not very usable for hours of work. Try it out for yourself, sit at your desk raise one or both hands towards your monitor as if it was a touch screen and see how long you can hold them there. If your hands are above your heart they'll get tired pretty quickly and eventually feel a bit numb.

Enquiring minds would like to know the hardware / android-image you used for this project. Tablets, even large ones, are nothing new now... but I'd really like to know how to build my own.

I'm guessing, with no evidence whatsoever, that this is an off-the-shelf desktop all-in-one running android x86.

Something like http://www.ebuyer.com/281328-acer-aspire-z3101-aio-desktop-p...

Touchscreens were popular before Apple in business applications. That might not be hipster enough, but Apple certainly didn't invent or popularize them. Further, at this form factor, you have that Microsoft table as well as a great many all-in-one touch screen computers from HP and Lenovo et al. Can we please not credit Apple with inventing another thing they didn't invent?

I would say their relationship to touch interfaces is similar to their history with mouse-based GUI: they didn't invent it from scratch, but they innovated upon the existing ideas in ways that helped the technology reach mass market.

Only if you've got a short memory.

Think back to 1999-2000ish. Palm Pilots were The Next Big Thing, Microsoft were pushing Windows CE handhelds of which every OEM had to have their own. Psion had their wonderful but sadly disregarded Series 5.

There was a large ecosystem of handheld touchscreen devices that was frankly pretty close to mass market many, many years before the iPhone, and from which I'd suggest iOS and Android borrow a number of interface concepts.

Admittedly, what qualifies as "mass market" is highly subjective. But I think it's fair to say Apple's innovations greatly accelerated adoption in both areas, opening up the technology to large numbers of people who would have never used it otherwise.

When you say popular, you mean failure. It is the small form factor that has made the difference

http://en.wikipedia.org/wiki/Touchscreen#Gorilla_arm

I say it's the connectivity that made the difference. My first PDA was a dell axim x5 and it was not much bigger than a modern iphone or android. But it was as fun to use as a computer that is not connected to the internet.

There's a project here in my town by a couple of my fellow engineering students working on a related idea they're calling awesometouch. They've installed a few in our city already as far as I know.

I don't believe they're running Android, and the machines are pretty special purpose, but still similar in concept as far as huge touchscreens go: http://awesometouch.org/

Yeah, awesomeTouch runs some custom Java on top of Linux, although both their hardware and software have most likely evolved since last I saw.. They just did betaSpring, I believe.

Also, hail from Lexington!

LexKY in the house! Our team just got back from Betaspring, we have come a long way since our early days hacking with the NUIGroup / CCV tools: http://nuicode.com/projects/tbeta

We've narrowed our focus to indoor map applications. We're still deploying them on giant touchscreens with AwesomeTouch, but have also started a project called BuildingLayer to aggregate & crowdsource maps for the insides of buildings (because people always get http://lostinabuilding.com). And just ask John King from CNN, the only thing better than games on a giant touchscreen are maps!

Thanks for the mention @jadedoto. You're correct, we're not running Android on our screens, but there is an Android port for MT4j, the Java multitouch framework we use: http://www.mt4j.org/blog/index.php?entry=entry110404-105831 If this Android x86 port comes further along (and there's some way to do remote administration for Android), maybe we can start developing for Android first, and porting that over to the large screens.

On the hardware side, one of the unfortunate limitations with most large touchscreens in that price range is that they only accept 2 touchpoints. For most applications, that's sufficient, but Martin mentioned that one of his goals is to enable multiple users to simultaneously interact. While the opportunity for that is a bit limited on his Acer screen, the larger screen size alone does make it easier for two users to view the screen simultaneously, and it's easier for them to alternate in usage without passing the screen back and forth.

This doesn't exist.

In three years Apple will invent it, brand it perhaps as iStation, and sell it for $10,000 at which point it can officially be said to exist.

It will only run software from the App Store.

I take it you're going for irony here, but, well, your hyperbole is right: as a shipping product, or anything other than a very cool hardware proof of concept, this absolutely does not exist.

When Apple invents some new UI conventions suitable for such a form factor, spends years testing and tweaking them, engineers all the hardware to fit neatly into a single unit with optimal battery life, builds a cost-effective manufacturing process and ships it along with an SDK full of frameworks custom-suited for such a large touch surface, then it will exist.

And yes, it will only run software from the App Store.

seriously, how does this comment (and mine too, I guess) add to the conversation?

we have a guy who's writing about some cool stuff he's done with Android, and you have to chime in with some Apple related snark? why? Apple wasn't even mentioned. the amount of negativity in these discussions is already high enough

It's a great idea as a gimmick, but on the long run gravity will soon make our arms go down on top of our mouse and keyboards. Maybe if it is inclined then it would make more sense since it is a more natural position. The point is that it isn't very ergonomic and we all know that programming on a 90 degrees touchscreen isn't a super idea! Still, it all depends on the UI and the use! Great experiment!

So, I'm really keen to do something like this but with a dining table and a projector set above it.

I had dinner at Inamo in London last week - their tables are essentially this kind of set up.

The interface is projected onto the table and from there you order your food, play games, watch live streams from the kitchen, etc.

Here is a video (apologies for the music, but this is the best video I could find:)

http://www.youtube.com/watch?v=el71OBLyfDs

The grumpy old man in me wonders why, exactly, people are supposed to be sharing a meal and convivial fellowship together need to have entertainment. I could understand the benefit of perhaps being able to order (drinks, particularly) via such an interface, but playing games and watching videos? Really?

dude, it was quite cool honestly even though I could immediately reverse engineer what they were using software and hardware-wise into v simple units (flash for the UX, a normal projector, touch pads, standard x86 board strapped to the projector).

Well, my probably even older great aunts seem to appreciate table games, more specifically card games. Board games are also older than any of us. I don't see how is this fundamentally different.

I'm not grumpy at the idea of playing games with friends and family--I just don't understand why meals must include other entertainment.

But, to each their own.

In other words, like a Microsoft Surface table, but set up for dining?

What's the advantage of using a projector over a touchscreen?

The first thought that comes to mind is spills. Spill stuff on a table, no problem, you just wipe it up like any other table. Spill something on a touchscreen and you have to worry about it gumming up the works.

That's why you'd project the UI on the table. Spills? no problem. But, you'll have a projected image on top of everything that's on the table (food, utensils, your arm...)

Wouldn't be an issue if you projected it from the underside, even though that doesn't seem to be the plan here.

They put the plane white plates in a specific place and so they project images of your order onto the plates - that was pretty cool. See the video - it's in there.

I remember trying a prototype from Microsoft about 5 years ago. It looked like a giant touch screen coffee table (http://www.microsoft.com/surface/en/us/default.aspx). The price at the time was 50k euros if my memories serves. I remember thinking I would wait a few years before getting one:) Looks like this time is coming soon!

That is Surface 1. Version 2 looks a lot better. See the CES 2011 demo here:

http://www.youtube.com/watch?v=0NesSYWODmM

There's also this Finnish start-up specialized in huge multi-user touchscreens, see this for example: http://www.youtube.com/watch?v=L1kVopprSRo

More here: http://multitouch.fi/case-studies/

I'd like to see some pictures of the device itself from different angles, a touch screen monitor is not a big deal in this day and age.

Very nice. Reminded me of a smaller touchscreen project I saw online - the iPhone-inspired, touchscreen, WinXP-based kitchen computer:

http://www.studio-lights.com/blog/iphone-inspired-kitchen-to...

If you decide to put this into production and need help/advice on how to bring the costs down, manufacturing, or logistics let me know. Really you just need to standardize the hardware, find one manufacturing that uses nearly all those parts and you'll be good to go.

Great job! I think 13" would be THE size to experiment on since it would be almost the same as an A4 sheet of paper.

Build a few into my desk already!

That's going to be awkward on the toilet.

Maybe have the monitor on some sort of swivel mount in the bathroom? I think that'd be awesome... but maybe awkward to type on a touch keyboard that big. :)

not necessarily..

Reply on news.ycombinator.com