Government provides civilians with a signal that's up to 4 meter accurate, however most GPS units will get you at least 7.8 meter (95%). You can not get higher accuracy with ONLY GPS. You can if you augment the signal with others, barometric pressure, distance sensors, compas ...yeah, you can certainly get more accurate reading, but not like what's being shown in the video.
I beg to differ CB. I'm willing to bet it's simply hibernating, or in some cryogenic state with a timer. Dead squirrels don't cause problems, they're already dead. However, one that's alive and moving around, that's a different story ...
What everyone seem to overlook, are the IR floodlights up above. While those quads can certainly fly on their own, formation flying like this is controlled by external devices, using the IR flood lights which detect where they are in relation to each other and a computer then adjusts each one. Show me this technology in open air, without any external influence, and you'll have my attention.
Disclaimer: I have built several quads in my time, the last one was done just a few months ago. The community I am on is at Aeroquad.com if anyone wants to start building quads. Quads can fly themselves, using way points or dead reckoning, however till the government decides to open up the GPS signal for more accurate calculations, you won't be able to do what you're seeing in that video, hence the IR flood lights.
I'm sorry, but that video, and many others like it, don't impress me one bit. It's all meant for indoor shows.
To answer your question, "Which microcontrollers are out there for a LED project?", there is no answer. Any microcontroller can be used, it's how you use it that matters.
Different ways of approaching this: a) The Mega has 14 PWM channels. b) Use a serial TLC with gray scale data - the 5941 is one such TLC. The 5940 can drive higher currents if you need it to.
10 of those 5941s will get you 160 PWM pins. You said you have 50+ odd LEDs ... I can only assume they are RGB LEDs since 160 PWM pins is way more than 50+ LEDs.
Another possibility, c), use individual ICs such as the WS2801. One per RGB LED. It has 3 PWM channels, R, G, and B. While they are expensive here in the US, you can get them really cheap overseas, even WITH shipping they are cheaper than getting them in the US at say SparkFun. For that matter, you can get premade and precut RGB LEDs already mated to WS2801s, with wires and all. Just hook up data and clock, provide power and you're good to go. So technically, all you need is a controller that can provide you TWO available pins, nothing else. You don't need PWM pins, you don't need multiple PWM pins, just two pins, digital or analog doesn't really matter, but digital is preferred.
With both the TLC as well as the WS, you just need a data and clock signal to control all the LEDs (if they are all in one long string. If you have multiple strings, then you need multiple data and clocks.) Both of them also have libraries written for them.
You say you don't want to do this with an Arduino, I believe what you meant to say was that you don't want to do this on your Uno board. That ATTiny85 you're using, is an Arduino clone the moment you put an Arduino bootloader on it. That falls under 'doing it with an Arduino' ...
You can built your own clone, or whatever board you want. It's all open source, schematics, design, and all. I built my own clones for all my projects, they are all Arduino clones, running stock Arduino bootloaders. The last time I built a LED project was with two sets of 160 RGB LEDs, each paired with their own WS2801. I did it all in house, didn't buy the premade ones. They do all kinds of fun stuff:
I received a test 8x8 RGB matrix, common anode. For common cathode, Maxim has the MAX7219 which is a serial 8-Digit driver which would work beautifully. 8 segments, 8 digits = 64 LEDs. Use three of them, one per color. Voila.
However, for common anode, they don't seem to have a serial driver. They have 8-digit drivers, but not serial. They have something called 'microprocessor compatible' (ICM7218) but it's not the same. So does anyone have a suggestion of an IC that's comparable to their MAX7219, but for common anode?
I suppose I can always tell the manufacturer to send me a different f-ing matrix.
Agree, much better approach. @KirAsh4, have you used 4xAA? What do you use to get it down to 5V?
I used MIC5205s to drop the voltage. The problem I had was that part of the project needed to be powered at a higher voltage than the uC, so I split Vin, one leg going to half the circuit, and the other leg to the MIC5205 and on to the uC.
Ultimately, during one of my more lucid moments, I realized that I could've added one small component to the one part of the circuit that needed the higher voltage, and actually change its characteristics to where it could've run at the same voltage as the uC. Then I could've used 3x AAs, no regulator and everything would've been just fine. Unfortunately, by then it was too late, the PCBs were already done, and the project was 2 days from completion and use.
Oh yeah, I recycle my batteries every 2 years, on the dot. Out with the old. I tend to take them to electronic swaps and someone will buy them for cheap. For the cost of replacing the battery packs themselves, plus shipping, it's cheaper to just buy a new unit at the store.
I got bit once when I had a power failure, on a 3 year old unit, and my main file server went down instantly. I had lost a pretty important (and pricey) project with it. Since then I said screw it, 2 years replacement it is from now on.
Electronics are electronics, batteries are batteries. One of them will fail, no matter how "uninterruptible" they claim them to be. It's a matter of when they will fail.
Things are back to normal now, so I don't think it is anything in-house. We actually have 3-phase power, and I could see from the meter that phase 3 had gone (at least the LED was off) then phase 2, then phase 3 again. So in our case it's possible for some parts of the house to work OK and others not.
I was going to say, based on your very first post, I immediately suspected one of your phases having gone down. My house has three phases feeding power, and thanks to squirrels, crows, and the occasional snow laded branch, we've had a phase go down. Usually we hear a loud pop when the transformer blows. With all of my computers on batteries, running off of two main 20Amp lines in the house, I flip both breakers and let the batteries take over and eventually shut things off. I systematically kill the other stuff as well. I'll leave circuits with only lightbulbs on them since I can still turn them on, they're just dimmer. And if they blow, oh well. Not nearly as expensive as replacing the fridge, or heater, or computers for that matter.
I'm still not understanding the 10K resistors on the TX and RX lines, whether they need to be pullups or pulldowns. The idea is that I don't get any transient voltage flowing TO the FTDI chip when there is no USB cable plugged in and the rest of the circuit is powered by the external 5V supply.
When I plug a USB cable in, the FTDI needs to be able to send data to the Atmel of course.
Are all your connections correct? Do you have a crystal in the circuit? Are your wires making good contact? Is your reset wires connected properly? How do you have things connected? No one can help you unless you give us more details.