55 Parallax Propeller's, Parallells Processing of Permanent Perturbations.

[Clock Loop]

432 core super computing? But who cares!
BigBoy55jpg.pdf



4884 - Archive [Date 2010.12.16 Time 04.33].zip

Articulating Boiler - Archive [Date 2010.12.16 Time 04.34].zip

BigBoy55.pdf<---skeeeeematik


BigBoy55litjpg.pdf


I like having 1,620 I/O at my fingertips.

To study the nature of randomness in a digital processor network using the same clock source, 54 prop chips were all connected to a master prop.

The master prop holds the eeprom, and accepts prop plug input for memory/program download.
The master prop controls the reset and clock lines for all 54 props.
The master prop also sends all combinations of a word out broadcast to all 54 props that are connected bus network style.

When a slave prop sees its number broadcast on the bus line, it replys with the same number.
Then the slave prop runs code depending on the number broadcast.

Its possible to broadcast messages to all props, or single props, after the initial enumeration is done.

Repeats are possible in any system that has reduced choices and increased speeds.
If repeats are a worry, then a long sized variable should be used in the random ID generation.

Enumeration of the Randomly generated ID's


Every step up in variable size reduces the speed that the whole system can communicate at due to more data being sent over the same speed pipe.

The communicaiton speeds of the bus network can be increased with lower value resistors, but the power draw will increase, if the props are connected to the bus with lower values, one needs to adjust all other bus resistors in relation.

Having?

You take 55 props,
some programming, (attached)
a bag of 470 ohm resistors(500) (not attached)
a 3.3v regulator (2A)
a huge breadboard
wire
57 leds


A prop scope is shown in the picture, which works well on DIAGNOSTICS.
(there are sharks in the water here, no swimming without a propscope on your boat.)


432 cores, 1,620 I/O, 64.8 A max peak draw - 10.8A continuous, @3.3v.
150kbps - communications using the same rx/tx lines for programming all 54 props.
Word location enumeration of 54 randomly generated ID's



****************END TRANSMISSION**************

TERMS OF USE: MIT License & Licensed under the TAPR Open Hardware License (www.tapr.org/OHL)

"Permission is hereby granted, free of charge, to any pers...........................
..............................OMITTED FOR FORUM............................................. ..
.................. OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. "

The dsp/fpga king is dead, long live the prop.

[Humanoido]

Clock Loop, congratulations on the construction of this great perturbation machine. I know how much time you've put into it to make it a reality. This is the current known record for the largest working multiple propeller-based machine.

[electrosys]

This is a really great experiment, or as Humanoido said, the largest working multiple propeller-based machine. Anyone who works electronic know that is not an easy task to do, you are indeed put a lot of works to this experiment.

[Clock Loop]

In the theory of quantum electrodynamics (QED), in which the electron-photon interaction is treated perturbatively, the calculation of the electron's magnetic moment has been found to agree with experiment to eleven decimal places. In QED and other quantum field theories, special calculation techniques known as Feynman diagrams are used to systematically sum the power series terms.


Uses:

-Particle and Planet simulations.

-Spread spectrum data broadcast.
Multi-frequency networks, having perturbations on a base set of frequency ranges, your broadcast frequencies becomes your unique ID(or channel)
Multipin output allows huge DB broadcast levels, make sure you know what your doing if your doing this.

-PWM outputs of many AMPERE at very high frequencies. (motor control, transformer, broadcast, etc)

[tdlivings]

Impressive
I have a side question, you posted animated gif's of the propscope and serial terminal what software tool did you use
to capture and create them. It is a slick way to show propscope captures instead of trying to record it using a video camera.
Thank's
Tom

[Tanstaafl]

All Hail Richard Phillips Feynman ! QED Rules !

[Duane Degn]

This is too cool.

I don't understand this part:

64.8 A max peak draw - 10.8A continuous

Is this a theoretical max?
10.8A works out to be 196mA per Prop. This seems like a lot of current. I'm I missing something?

I keep wanting to try something like this. I keep trying to think of a problem a bunch of Props working together could solve. Make that a problem I care about. Most of my ideas so far deal with image processing (both input and output).

Thanks for sharing.

Duane

Edit: I see now. It's the current this many Props are capable of sourcing (and I suppose sinking).

[Chambersburg]

Greetings.
Nice job on this project.

Will preface this by saying that I have not yet moved up to the Propeller, so I don't know any spin.

What method are you using to truly generate the random numbers?

To me, the only way of getting a set of well-distributed random numbers is to use some external input.

Thanks.
W

[Clock Loop]

Using the Real Random object in the obex created by Chip Gracey

This object generates real random numbers by stimulating and tracking CTR PLL jitter. It requires one cog and at least 20MHz.

http://obex.parallax.com/objects/62/

Background and Detail:                                                                                         │
│                                                                                                                │
│ A real random number is impossible to generate within a closed digital system. This is because there are no    │
│ reliably-random states within such a system at power-up, and after power-up, it behaves deterministically.     │
│ Random values can only be 'earned' by measuring something outside of the digital system.                       │                                                                                     
│                                                                                                                │
│ In your programming, you might have used 'var?' to generate a pseudo-random sequence, but found the same       │
│ pattern playing every time you ran your program. You might have then used 'cnt' to 'randomly' seed the 'var'.  │
│ As long as you kept downloading to RAM, you saw consistently 'random' results. At some point, you probably     │
│ downloaded to EEPROM to set your project free. But what happened nearly every time you powered it up? You were │
│ probably dismayed to discover the same sequence playing each time! The problem was that 'cnt' was always       │
│ powering-up with the same initial value and you were then sampling it at a constant offset. This can make you  │
│ wonder, "Where's the end to this madness? And will I ever find true randomness?".                              │                                                              
│                                                                                                                │
│ In order to have real random numbers, either some external random signal must be input, or some analog system  │
│ must be used to generate random noise which can be measured. We're in luck here, because it turns out that the │
│ Propeller does have sufficiently-analog subsystems which can be exploited for this purpose -- each cog's CTR   │
│ PLLs. These can be exercised internally to good effect, without any I/O activity.                              │                                                                   
│                                                                                                                │
│ This object sets up a cog's CTRA PLL to run at the main clock's frequency. It then uses a pseudo-random        │
│ sequencer to modulate the PLL's target phase. The PLL responds by speeding up and slowing down in a an endless │
│ effort to lock. This results in very unpredictable frequency jitter which is fed back into the sequencer to    │
│ keep the bit salad tossing. The final output is a truly-random 32-bit unbiased value that is fully updated     │
│ every ~100us, with new bits rotated in every ~3us. This value can be sampled by your application whenever a    │ 
│ random number is needed.

[Humanoido]

Generating these highly randomized numbers is a great method for randomizing the unique indexing of all Propeller chips. From the same program, what if you wanted to numerically number the chips sequentially in position from 1 to 55?

[Clock Loop]

I wrote a lengthy reply and the stoopid browser hotkeys erased it all because it decided to go back a page. I really need to turn that off... (what a dumb idea) GRRR.

It basically said,

You need to run my enumeration sequence initially to get each prop unique.
(i see no way to do this without the initial random ID enumeration, due to the broadcast nature of this design)

Then you can re-number the ID's of each prop using the initial enumerations unique ID's.

IF you need any props to retain their ID's after power off, the prop would need pins as ID's or external inputs(eeproms, etc)

If you need individual props to be specific ID's (due to unique pin function) then it would be wise to install a eeprom or pin ID on that one/many prop, and have all other props acquire the ID from enumeration. (any prop with ID 0 would acquire enumeration ID.)

[hinv]

Or, if you are not adverse to using 2 pins, you could just have 1 pin on the master output a high to some input on the slave, and simultaneously send an identifier byte of "1" on the serial bus, once the slave has it's id, it could hold a pin high connected to the next one (sub slave) in the loop and send an identifier packet saying 2 on the bus and so on.
It would use 2 pins from each prop to do the enumeration, but it would get you the permanent numbering you are looking for without having to manage 55 eeproms.

BTW, ClockLoop. This is REAL fantastic, but your video doesn't really show of any synchronizing. How about putting an RGB LED on each prop and playing a little animation across all 55?

[Clock Loop]

BTW, ClockLoop. This is REAL fantastic, but your video doesn't really show of any synchronizing. How about putting an RGB LED on each prop and playing a little animation across all 55?

Ok. It will take me a bit to hook it up.

[Clock Loop]

After moving one too many times, more than 2/3 of these props died somehow, so I will not be doing anything more with this project unless I find spare cash. (what a great loss of prop chips tho, $300)

[mindrobots]

So sad. I felt a disturbance in the Propeller Force!

[Cluso99]

Are you sure they are dead???

[Phil Pilgrim (PhiPi)]

Poke them with a stick. If they twitch, there may still be hope!

-Phil

[Humanoido]

After moving one too many times, more than 2/3 of these props died somehow, so I will not be doing anything more with this project unless I find spare cash. (what a great loss of prop chips tho, $300)

Why would moving kill 2/3rds of the Propeller chips? Probably a loose connection happened on the solderless breadboard or a wire(s) popped off somewhere. I've moved the Big Brain internationally between countries and the chips keep on ticking with those breadboards. I hope you can have another look at this because this is a really great project!

With the big Brain, to keep track of all those wires, I take photos each time before moving. Then when reassembling after the move, the photo is the only reference needed especially if some errant wire was bumped off.

The photo at the link shows how many wires need to be tracked for a Propeller Brain with 100+ chips.
http://1.bp.blogspot.com/-_cQ2cAJvkG...600/wiring.jpg

[Clock Loop]

During transport the props were pulled from the boards, but ended up in attached to nothing (not even static pads) in a container of electronics that contained a battery, perhaps some kinda leg to leg conduction in a pile of props and other things. Always store chips in their own containers if you don't use foam on the legs, kids.

[Humanoido]

During transport the props were pulled from the boards, but ended up in attached to nothing (not even static pads) in a container of electronics that contained a battery, perhaps some kinda leg to leg conduction in a pile of props and other things. Always store chips in their own containers if you don't use foam on the legs, kids.

Oh no! Sorry to hear about the loss...those 9-volt batteries that have close +/- terminals on the end can do damage to components in the same bag. Tape over the ends seems to work for me. How do you test each propeller chip to know its bad? Do they work partially with some pins or are they completely dead and won't run code?

Perhaps you can still get some use from each chip based on partial functioning. I also wonder why the props were pulled from the boards... I find the boards to be an excellent way to keep the lot protected and for transport, wrapping the boards in a layer of anti-static material and following with a packing of soft cotton towels, then stacked with more boards seems to work.

Some of these boards have metal backing and weigh more when shipping, however the high density ones have light weight transparent backing made from recycled clipboards. You can see the arrangement here where the boards (one board to the next) are waiting for disconnection (wires between the large boards) and packing and shipping (green backing is visible).

The Earthquake Protection with the Coddler link describes the Coddler for solderless breadboards. As an experiment some Coddlers were shipped but not with the breadboards. The polymer cracked and was replaced. Here is what it looks like reassembled.