Project Megapixels - Picture Construction Subgroup Area

[sputnik]

we missed each other on a couple of points- it's the cost of doing this in a forum which is fine since we all get to see the discussion. on this first point I was thinking about making do with less. you guys have determined the size of the pixels and i'm taking that assumption and making it a little coarser by adding 8 inches between the 3x3 blocks. this might stretch the available pixels.

Yeah, I see what you're saying. 3'x3' or 1m x 1m is essentially the same as far as I'm concerned.

the way I understand this (and the concept I'm trying to describe) the 3x3 pixel unit would mean nothing to the satellite and the picture. It is just a way to organze the work and to modularize the picture construction. As I said I am concerned that any one section break up and then the rest of that module becomes a mess. if we modularaize it into 3x3 units then it isn't so bad if one has a problem.

The 3x3 or 1m x 1m pixel size is extremely important to the picture. In the best of worlds each pixel would be an individual element so that none were tied to any other and if one tore loose it wouldn't affect others. I think it would be very easy to lay down individual pixels too, just 4 spikes, one in each corner. However, that means that there are 4 times as many spikes as pixels. Plus every spike has to be pulled out of the ground (as well as pounded in (see Dork's post at the end of page 1). If someone commits to 100 pixels, I'd like to have it arrive as a row of 100 pixels, already joined together. Then, just unroll it on site, attach to the grid, tack to the ground in a few places, and it's done. Cleanup is relatively quick, cut the grid ropes, remove the spikes and burn or otherwise discard the bundle.

never been to the playa. do the tops of tents get as dirty as the bottom? I'm thinking that the shaking from the wind will help. maybe the dust is just too sticky. if so then this is just an open issue that we need to resolve. the rolling out at the last minute sounds impossible as you indicated.

The dust is like talc. It isn't sand at all. It does tend to stick to things. My tent is covered in the stuff, despite having taken it out and having it rained on in November.

when you are modeling this on top of your van I'd recommend not being so close to the roof. The best place would be to put it in front of the van or a few feet above and as far forward on it as possible.

some places of the van you will be in dead air. in others you will be catching an unrealistic updraft.

Good points. That's why I want to put a camera up there as well so I can see what's happening.

[Bob]

Re: hog rings: they hold well, and could be tossed in the bonfires, but have very sharp points hazardous to bare or sandalled feet. I'd avoid them unless you can bury the points in the fabric on the underside.

If you're thinking about the kind of orange twine DPW uses to hold up the trash fence, you might try phoning the org office and asking for the staff who do DPW's purchasing to find out where to get it cheap in large rolls. I think it's polypropylene, same material as common orange & black truck rope. We've used fence twine for repairs on tying grommeted tarps off to cables when we run out of zipties, using a sort of blanket stitch through the grommets.

Re: dying, make sure if it rains that unfixed dye won't bleed through onto the playa.

[S1m3R]

Here is another concept for putting this thing together.

The assumption is that the picture will get full of dust no matter what we do so the attachment scheme should not take that into account. keeping it "clean" becomes a separate issue. So now the requirements are keeping it attached to the playa and keeping the pixels in the order that we want so that the picture emerges.

Try this on - people bring fabric cut in squares. no hems - no seams - no sewing. at the playa we have a device called a fabric drill. The core group takes the fabric and pile it up with the edges in rough alignment. all four corners are drilled so that we now have holes.

We bring enough ground staples such that we can staple the four corners of each pixel. each pixel is now overlaped with corners having one layer of fabric and inside corners having as many as 4. note that this means there are many duplicate corners which means fewer staples needed. I haven't worked the math but I think 2000 pixels is going to be something less than 3000 unique corners. I'll talk to the finite element guys tomorrow/today and find out what the equation is.

a ground staple is a piece of rebar that has been candy-caned. the staple is pounded down into the playa all the way so that the top of the "cane" is also in the ground. the people putting in the staples will need to push aside the fabric between the pixels so that the top of the cane doesn't have to pierce any fabric.

Cool aspects of this. no perimeter fence. no ropes needed to build structure - just something to lay out the grid. no sewing or glueing. volunteers only need scissors and they don't have to put any of the pixels together. we do all assembly on the playa - less coordination. it is a compartmentalized construction so if one pixel has a problem it is just that pixel. if one staple comes out of the ground it is only at most 4 pixels flapping around. people can walk on the picture if they are willing to walk around the staples and carry a broom. everyone who gets near the thing should be handed a broom so they can help sweep it clean. instead of the spectators being a problem they become the solution.

The problems obvious to me are getting the fabric drill and access to electricity and a table to set the thing up on and keeping the drill clean so that it goes back to the owner as good as we got it. making several thousand ground staples seems a little troublesome. pounding in several thousand ground staples sounds like work.

questions I have. how big is the drill bit? is rebar necessary or could we use a smaller metal diameter. The bit looks small in this picture. they use these things to mark fabic for subsequent assembly that does not involve the holes.

http://www.reliablecorporation.com/prod ... /xd%5Fcd3/

My ex-wife used to have a fabric drill. I don't know if she still has it nor do I know if she would let us use it. In other words, I might be able to take care of getting that done if this is the plan we land on.

[S1m3R]

3x3 in my previous post was 3x3 PIXELS. clearly not obvious unless you lived inside my brain.

I hear you on the assembly time and hassle on the playa if all the pixels are separate. It will be a judgement call by managment (sputnik) considering the pros and cons. Maybe a mix of the ideas - For the really committed people they could bring the big sections sewn together and we could just plop them in place with slits made for enough staples to keep it on the playa.

note that the velocity of the wind at ground level (in a theoretical flat plate analogy) is zero. Since the ground is not flat this doesn't work perfectly but it does mean that the wind is much less of a problem on the surface than it is just a few inches or feet off the ground.

[Gatecrasher]

Hey all. It's me, the satellite guy.

Check out my post in the other forum.

http://eplaya.burningman.org/viewtopic. ... &start=390

A couple things on logistics re: satellite.

There is no gaurantee that either satellite will take a good collection. Plan for both of them to minimize risk, which means, depending on the flight schedule, you may have as much as 2 days between satellites (or they could both be over the same area on the same day). The satellites are overhead about every 3rd day, so be prepared for a Thursday collection if it comes out that way.

Also. DG his higher resolution by flying lower, IKONOS has less blur by flying higher (.65m vs 1m). They both use identical instrumentation and made different decisions about how to get the best pictures of the ground. I can likely get the IKONOS shot taken at no cost without any hassle. Side-by-side, DG may or may not have better detail, IKONOS will have better color and it will be more difficult to see the color-shift vs. the B&W, YMMV and it will be interesting to see which comes out better, but they both will look pretty good.

Note. Don't think about selling the images. Neither company shoots any exclusive shots, (meaning anyone can buy it out of the satellite archive) and they will be made available, and it's not really in the BM spirit. I don't think that's what you were intending by the previous comments, but if the people at DG need money, I'm sure you can get donations.

Speaking of which. I guarantee both companies will take pictures of BRC this year. I'll do it for Ikonos, and those copy-cats at DG will probably do it as well. Both companies will eat up the chance to take pictures of something cool, rather than the regular stuff they do now. Trust me. :-)


Mike

[S1m3R]

I talked to the Finite Element guys. They confirmed that the number of grids is close to the number of pixels if the pixels are sharing corners like an FE model does. There is no clean equation for a complex shape. it depends on the number and length of free edges to interior corners. Not real important in my mind. There are few enough pixels that someone can just raw count the corners to get an idea of the number of staples needed.

I was also thinking that sand would probably be a worse case scenario for experimenting with attachment schemes. I was recently working with some 1/4 inch aluminum rod that was purchased at a home depot. I suspect it would do the trick in terms of ground staples. It would be easy to cut and bend too.

Thanks for the information Gatecrasher. I have to say I'd be sad at this point if those satillite shots are taken and this isn't part of the mix.

[sputnik]

Scott,

I hope we get a chance to talk in person at some point, becuase I am totally not following what you just said.

[chroma]

Maybe I'm crazy but how about magnets?

Here's what I'm thinking: we use magnets to attach each pixel to its neighbors and to rebar anchors. ThinkGeek sells what they call 'Curiously Strong Magnets'. They're really just small rare earth magnets which are available for bulk purchase from several places online (e.g. here, or here). Since the fabric is pretty thin I think the magnets might bind tightly enough to hold well in the wind.

We could lay long segments of rebar down on the playa, spaced say 10 pixels apart, either in a grid or just in one direction (e.g. every 10 columns only). Then we use magnets to attach the fabric to the rebar. The rebar could be anchored with candy canes or we could even just leave it if we think its heavy enough to stay put on the ground. Each of the pixels between rebar sections is attached to its neighbor with a pair of magnets at each corner of the pixel. If we think more anchoring is needed we can run steel wire between the anchored rebar sections and attach pixels to the wire using the magnets.

Quick sketch:


Pros:

• volunteers do not have to sew sleeves or add grommets to their pixels.
• setting up the grid might not require pounding any rebar, simply laying it out flat on the ground.
• quick set up: simply grab a bunch of magnets, find the location of your pixel and snap it in.

Cons:

• while apparently a 1/4" magnet can produce some 5lbs of force, I don't know if it will be enough. Testing would be absolutely necessary.
• cost: in bulk a 1/4" cylinder magnet runs about $0.23. If we have 2000 pixels and we need about 2 per pixel (one on top, one on the bottom), the cost is 4000 x 0.23 = $920.
• magnets == moop? If the magnets don't hold, picking up 4000 tiny magnets off the playa is gonna suck.

I was partially inspired by LED Throwies. If we're feeling REALLY adventurous we could even make half the magnets into throwies and then Megapixels would be visible at night!

Anyway, just a crazy idea, but thought I'd throw it out there...


benj

[Dork]

Interesting thought. Dealing with those rare earth magnets is a pain, though, because of how strong the attraction is and how brittle they are. Even if they were strong enough to hold the fabric together and to the rebar, the difficulty in assembling and disassembling the whole mess might make it impractical.

Moop would be less of an issue. Just run a piece of metal over the surface and any loose magnets will get picked up.

[sputnik]

I'd suggest you give that idea a try. I'd be curious to see how it works. I like the idea of using LEDs to light this up. Maybe we should make the perimeter lighting out of LEDs and skip needing a gennie.

[ThePikey]

People will probably wander off with any small trinket-like light source.

"Dude look what I found!" "Oh, sweet, where'd ya get it?" "Oh, just over there, there's a whole bunch..."

(But maybe I'm too pessimistic)

[chroma]

People will probably wander off with any small trinket-like light source.

"Dude look what I found!" "Oh, sweet, where'd ya get it?" "Oh, just over there, there's a whole bunch..."

(But maybe I'm too pessimistic)

Fair enough. I guess we just need to make sure that the LEDs look like an integral part of the project as opposed to loose trinkets. Either that, or gift enough throwies so that the demand is curbed.


I've ordered 120 magnets for a small 7x7 pixel test (minipixels?). I'll also experiment with some LEDs. I'll post results & photos once I get everything assembled.


benj

[sputnik]

I've ordered 120 magnets for a small 7x7 pixel test (minipixels?). I'll also experiment with some LEDs. I'll post results & photos once I get everything assembled.
benj

Cool. You have a leaf blower, right?

[CampCounsellorBRAD]

I was thinking that a quicker, far stronger way, was to attach the pixels to two ropes (hemp twine - thus burnable). One on each side (say left and right sides). The fabric would be folded over on those sides and stapled with office staplers... perhaps 5-10 staples along each 1metre length.
then the rope would be attached to rebar with zip-ties. (easy and strong).

The cost of doing this is about 3 cents. Doing magnets at .23 each will get prohibitive very quickly. Also magnet power laterally, (sliding them along) is far weaker than pulling out orthogonally to surface... another thing to check).

Also, the blinky idea rocks.... but the time to manufacture 100s is very high. Probably 15 per hour of labour is max... so could be fun, but very time consuming. Besides, we would want people to contribute more fabric pixels than spend money on blinkys (well until we have 5000 pixels)

For the gennie idea... what about using LED xmas lights. Cost a lot more, but won't need much power... maybe we could run on a couple of rechargeable batteries with inverters... if we minimize lites, (just enought to prevent people from killing themselves on rebar stakes) we might get away with no gennie... and recharge the batteries during the day.

Cheers,
Brad.

[thisisthatwhichis]

One thing to consider:

If all of the pixels end up attaching together... IE. with rope and rebar along grid lines, we have built a VERY large sail laying on the playa floor. When the wind starts picking up some pixels, the others attached to them will magnify the affect of the wind, and put major stress on individual stakes. If stakes start to pull out, you then have a major problem.

The idea of securing individual pixels, or small sections attached together, makes sense, so if a section goes, it's a little easier to repair.

[CampCounsellorBRAD]

Lets see... if they start popping out... and someone grabs the end... the headlines would read: "Burner goes flies a kite... by the wrong end... end up with aerial picture of entire event..."

Yep you make a good point. Maybe making the sections of pixels limited by a small number would prevent this point. However, the strenght of 1' pieces of rebar in the ground to lateral movement is very very high.

If we were to limit the "sections" of pixels to 5-20 long, the power of these to overwhelm 6-8 rebar would be minimal (looking at the number of stakes to hold down a shade structure. In my burner experience, I've seen shade structures collapse, ropes break, but never seen rebar pulled out.

Regardless of how many sections we stake down... (either in pieces of 5-20 or longer) we need to figure out how many metres to space rebar. Each rebar will be a lot of work. (to put in, and to take out). We need to figure out the absolute minimum. If we use too few, a section could come loose (however if sections are limited with Safety cut off pieces - this won't be horrible)

My Guess is that we need rebar every 4 or 5 metres. A 20 metre piece would need in total 8 rebars (4 each side). But since a rebar would usually (except at ends) hold a rope on both sides it would only need 4 rebar. however, that is still ~500 rebar pieces!

Assuming a person could drive one every 4 minutes and remove about the same... that is 8 minutes each or 66 person hours... doable but a huge commitement of work. (Especially to remove when all the FUN after the burn has ended).

My thinking is that we very carefully consider how much rebar we will need to secure the image to the playa. (I don't know the answer, but just calculating this starts me to worry abit about this issue).

Too little, sections blow loose in a windstorm.

Too many, and the logistics of securing (and removing) become non-trivial. (People WILL forget and leave... leaving Dan with the LNT scut work to do. Which could take days....

Anyone else have suggestions?

(Another idea I have, but not sure how silly, is to run perpendicular rope to secure sections to playa. Sorta like weaving. This rope will not hold pixels per se, but just keep the rope holding edges down. Prevent sailing.)

[sputnik]

My thinking is that we very carefully consider how much rebar we will need to secure the image to the playa. (I don't know the answer, but just calculating this starts me to worry abit about this issue).

Whenever I get out the calculator it scares me.

[ThePikey]

I think a sizable enough fraction of people will want to help lay 'their pixels' down on the playa, or assist in setting up in some fashion. If only for a little bit. The playa provides! Have faith.

Having enough people to guide them as to where stuff needs to go and fabric staplers on hand to keep them busy and interested and not wandering off, that will be the set-up challenge.

(I assume you're planning on laying out the grid of cables, rebar, and xmas lights *before* anyone shows, right?)

Getting people to help *clean up* on the other hand... that one may be a little more problematic. But tearing down shouldn't be as hard as setting up. Yanking out rebar is easy-peasy if you have enough leverage, just pop 'em out with a crowbar. (That's how I did it last year, and I had a fractured wrist at the time. No prob.) Stashing the rebar into buckets/boxes/whatever will be a little more time consuming, don't want the moop after all. Hopefully schlepping fabric over to the burn platform won't be too hard. (Especially if it's last day and you can use a vehicle to do the toting.)

[thisisthatwhichis]

Wasn't your Moop day starting Sunday? If so, I think it could be done... There are enough folks staying into Monday. And Moop will be a lot quicker than setup.......

[sputnik]

I think a sizable enough fraction of people will want to help lay 'their pixels' down on the playa, or assist in setting up in some fashion. If only for a little bit. The playa provides! Have faith.

Having enough people to guide them as to where stuff needs to go and fabric staplers on hand to keep them busy and interested and not wandering off, that will be the set-up challenge.

Yes, I am hopeful in that regard

(I assume you're planning on laying out the grid of cables, rebar, and xmas lights *before* anyone shows, right?)

Somewhat. I'll be arriving on Sunday. My camp setup is pretty simple, and should be done in an hour or less. I expect it may take a while to find the site and then we'll have to start prepping. It's possible that someone will arrive earlier than I will and will get things started.

Getting people to help *clean up* on the other hand... that one may be a little more problematic. But tearing down shouldn't be as hard as setting up. Yanking out rebar is easy-peasy if you have enough leverage, just pop 'em out with a crowbar. (That's how I did it last year, and I had a fractured wrist at the time. No prob.) Stashing the rebar into buckets/boxes/whatever will be a little more time consuming, don't want the moop after all. Hopefully schlepping fabric over to the burn platform won't be too hard. (Especially if it's last day and you can use a vehicle to do the toting.)

I am hoping that some people will come back to collect their stuff and we'll get some help there. But, yeah, I expect cleanup to be rather quick (except for MOOP scans).

[sputnik]

Wasn't your Moop day starting Sunday? If so, I think it could be done... There are enough folks staying into Monday. And Moop will be a lot quicker than setup.......

It really depends on the satellite passes and how much is done by the first one. I've heard the first will be on Thursday and that there may be a competitor on friday. The next pass would be on Sunday. I'm planning for a thursday/friday image and then teardown beginning on Friday afternoon if possible so I can enjoy the weekend.

[CampCounsellorBRAD]

If we organize the grid before we reach the playa, the set up should be self organizing. The trick is pixel color. Thought would be have a device that could unwind/wind thin twine - automatically measure off 1meter lenghts. Then we could spray paint them according to the graphic. Then when we get there. the thin twine would be the 'guy' wire. Take your pixels to the next available spot with your color painted on it. Cover it up, staple it in place. DONE!.

The trick would be to rig up a device to measure out a meter and paint them quickly... Anybody have a better idea how to make the placement of pixels simple than ANYONE walking by can help?