We’re throwing money at our monitor and nothing’s happening!
Sometimes we get hacks sent into our tip line that are outrageously awesome, but apart from a YouTube video we’ve got nothing else to write about. So begins the story of the flying Back to the Future DeLorean quadrocopter. Sadly, the story ends with the video as well. (If you’ve got any info, send it in!)
Fine, we’ll throw in another cool car
Mercedes covered a car with LEDs and made the James Bond’s invisible car from Die Another Day. The Mercedes video cost tens of thousands of dollars to produce, so of course there’s camera trickery; we’re just wondering how much credit Adobe After Effects gets for this build.
Microsoft touchscreen demo might be impossible
Yes, Microsoft does care about user experience. Just take a look at this video from their applied sciences group. They did user testing with touchscreens that updated every 1 millisecond, compared to the ~100ms our phones and tablets usually update. Of course the result was a better UX, but now we’re wondering how they built a touch screen that updates every millisecond? That’s a refresh rate of 1 kHz, and we’ve got no clue how they bodged that one together. We’re probably dealing with a Microsoft Surface projector/IR camera thing here, but that doesn’t answer any questions.
Edit: [Philip Rowney] sent in a tip that it could be this TI touch screen controller that can sample above 1 kHz. The only problem is this chip uses a resistive touch screen, instead of a multitouch-enabled capacitive screen. At least that solves one problem.
And now for something that can measure 1 kHz
[Paleotechnologist] posted an excellent guide to the care and feeding of an oscilloscope. Most of our readers probably already know the ins and outs of their awesome Techtronix and HP units, but that doesn’t mean the younglings won’t have to learn sooner or later.
Good idea, except the part about saving it for spring
In a moment of serendipity, [Valentin] figured out how to use touchscreens with wool gloves. The answer: rub thermal grease into the tip of the index finger. It works, and doesn’t look to be too much of a mess. We’ll remember this for next winter.
The last one didn’t have a picture, so here’s this
[Darrell] used a little bit of LaTeX and Ruby to make colored labels for his resistor collection. We’re struck with the idea of using test tubes to organize resistors. It’s cool and makes everything look all sciencey and stuff.
Touchscreen, I’m going to guess it’s doing it with back projection of the image, the screen is side lit with infrared and there’s something a like an overclocked Wiimote camera also back there there tracking the illuminated finger. IIRC a standard Wiimote sensor will do the capture and tracking in about 4ms, so it’s not far off. It would explain why the screen looks frosty and there’s a hacked up cardboard box around it.
Oh, and it’s Tektronix and Agilent 😉 (Hasn’t been HP for many years now.)
It’s probably not the update frequency they want to adjust for the touchscreen, but the movement-prediction of the pointing device (aka. finger).
With the mouse you also had some delay but didn’t notice it, since the mousepointer was always painted at the same position on the moving frame. So delay wasn’t noticable. With the finger you can see the redraw is behind.
So all MS has to do is to predict where the finger will be at 4 – 10 ms in the future and start painting there, since it takes about that much to repaint, flip the buffers and update the pixels.
Since everything with mass takes a lot of force to change direction quickly, you can quite accurately predict the position in the near future.
Only problem is that this will work fine with higher speeds, but it will be less accurate at slower speed. But fortunately at slower speeds the distance traveled is much less so the delay is less noticable.
So when you adjust the delay according to the speed, you can get a constant delay in mm. (or inches 😉 ) which will be perceived as a constant delay by the user and errors in the predicted position will be far less noticable.
That’s about what I expect MS will be doing when integrating it in their OS.
It might also be possible to do this with a touchscreen driver, but I expect that driver lacks context so it might not work as good as it could be.
That Delorean is sweet, but I don’t think it will get up to 88mph.
Well, it’s a scale Delorean so that means it should only need to get up to 88 scale mph.
I’m a fan of the test tubes. Next time I properly set up a workbench I’m going that route.
whats so special about sampling touch ADCs at 1KHz?
Combined with the fact that the microsoft team had to draw new pixels at 1kHz, I can’t figure this out.
Sure, it’s possible to read the touch data at 1 ms/sample, but the video shows the screen updating at around 1 kHz, or having a 1 ms refresh rate. They could have used the same tech that goes into 120 Hz LCD panels to do this, but 1kHz seems impossibly fast (as in, there’s been no need to engineer something that fast in the last 100 years) for any display technology.
What about oscilloscopes that have multiple tens or hundreds of thousands of waveform updates per second?
The refresh rate is quite high, but I don’t understand the skepticism and/or sheer amazement…
Ah, now i get it, Hackaday summary is WRONG – they arent talking about sampling at 1KHz, but about processing delay
If you google phone screen lag you wont find a SINGLE iphone video 🙂 but over a hundred Android vids showing this
http://www.youtube.com/watch?v=2cBMpDLgX2E
this delay is absent on Apple devices – this is why Android phones need 4 cores and still feel slow.
they seem to project the picture from above, a DLP projector for example uses a fast-moving color wheel and is easily able to update at that speed. Plus they’re Microsoft Research, they have the budget to not use off-the-shelf parts, maybe they built that touch pad (it’s not transparent, could be a custom PCB, it doesn’t need to be resistive), maybe they’re even driving the DLP chip directly…
Mmm… thermal grease… Out of the five tubes on my desk, only one brand was willing to provide a msds through calls, emails, or their website.
Might be safe… but might also cause that freaky black and white person in real life condition…
Please tell me where I can send my money to get a flying Delorean quadcopter.
Send one dollar to Happy Dude, 742 Evergreen Terrace, Springfield. Don’t delay. Eternal happiness is just a dollar away.
‘Happy Dude’
YES! You win a lifetime supply of free internets!
I thought it was obvious that a higher sampling rate of touchscreen input leads to a better input. Does Microsoft even need to allocate a research team for this? Well.. they should also dedicated another team to acknowledge that it will also consume more power.. [/sarcasm]
I think the point was to see how fast is fast enough.
I would guess that if you sample your touchscreen about twice fast as you update your display you would get good enough response. Now what FPS you would need for it to run correctly…
At the end of the Delorian video he has details of the stuff used.
On his channel
http://www.youtube.com/user/native118
He has other vehicles he’s built.
umm, how hard was that?
🙂
That Delorean was cool but judging from the way it was wobbling in flight I think Doc had been drinking, but since it was from Russia that’s not unusual.
http://forum.rcdesign.ru/f123/thread250976.html?highlight=
Touchscreen: Let’s limit this to easily obtained components, and the skills/budget of at least some here.
Start with the camera from a Playstation Eye which has a refresh rate of 120hz, or 8.33ms.
Couple that with a heavy-duty FPGA for threshold detection and filtering. Probably very little latency there, let’s say we end up with a 10ms refresh rate. That’s pretty good. But 10ms isn’t 1ms.
However, a human finger can’t significantly change velocity or direction in 10ms. So you can predict with high accuracy where the finger will be 10ms in the future, or longer; and perform that prediction with a much higher frequency than any real data available from the camera.
By using such trickery, you can effectively eliminate all visible lag from the camera, simplified display hardware (probably also on a dedicated FPGA), and the projector. Well enough to produce results indistinguishable from what MS showed, even if that’s not how they actually did it.
It’s pretty easy to do two finger touch with a standard resistive touchscreen and that’s enough for most things (one+two finger tap, two finger scrolling etc). The problem is that it isn’t possible to separate two finger gestures from more touch points and it isn’t possible to get the center of touches.
My g1 iPad has much better response running doodle-buddy than all but their fastest example. I think it’s probably runnin at 15ms to match screen update, which is pretty good!