"Universal Jigsaw Puzzle" Hits Stores In Japan

Riktov writes "I came across this at a Tokyo toy store last week, and it's one of the coolest things I've seen in a long time. Jigazo Puzzle is a jigsaw puzzle, but you can make anything with it. It has just 300 pieces which are all just varying shades of a single color, though a few have gradations across the piece; i.e., each piece is a generic pixel. Out of the box, you can make Mona Lisa, JFK, etc, arranging it according to symbols printed on the reverse side. But here's the amazing thing: take a photo (for example, of yourself) with a cell-phone, e-mail it to the company, and they will send you back a pattern that will recreate that photo. This article is in Japanese, but as they say, a few pictures are worth a million words. And 300 pixels are worth an infinite number of pictures."

Comments

puzzle?

[Anonymous Coward]

Not much of a puzzle if you can assemble the pieces in any orientation and layout you wish.

Re:puzzle?

[Toonol]

It's a puzzle when you have a specific image you're recreating. If you aren't, it is then just a toy... or perhaps an artistic medium.

Re:puzzle?

[sexconker]

Puzzles require thinking and solving.
This is a cardboard version of pixelblocks.

http://www.pixelblocks.com/

Re:puzzle?

[Red Flayer]

I had something similar as a kid, but there were two colors of varying saturation. Grey (from black to white), and gold (from almost white to intensely saturated). The pieces were different shapes, though, so there was a puzzle aspect to it, though all the pieces were some number of squares in different configurations (like tetris pieces, but more shapes in different sizes of 1 to 10 or so squares).

It came with a few patterns to copy from (tiger, city landscape, I can't recall the others, since I never did them).

Unfortunately, there was no www at the time, so no website to submit pictures to for patterns. One of my brothers did make some nice pictures based on photographs.

The memory is a bit hazy, but I know the company that made it was asian (I remember there were pictograms and poorly translate English on the box).

I know, I know -- cool story bro.

Re:puzzle?

[Kagura]

I saw this at Tokyu Hands a couple days ago. Now I know what it was. The picture is only just barely similar to whatever photo you send them when you look up close. You have to view it from far away to have it appear to have the detail of the photo.

Re:puzzle?

[Anonymous Coward]

Sounds much like oil painting. Up close it looks like crap, but stand a few feet back and it looks great.

Re:puzzle?

[tyroney]

Not quite. Many of the "pixels" are a solid shade, but quite a few of them are various (still monochrome) gradients at variety of angles. By rotating the gradients and putting them next to certain shades or other gradient tiles, all sorts of interesting (if blurry) features can be (re)created.

Re:puzzle?

[shentino]

I'd swear that it almost looks like a tool to teach someone about basic JPEG encoding.

Re:puzzle?

[Anonymous Coward]

You're not smart because you can point out that an innovation changed the point of the original creation. Derrrrr. Yeah, it's not a puzzel anymore, that's kind of the point. I don't think anyone was mislead by that except you.

Re:puzzle?

[Nazlfrag]

That would be interesting, but these don't look much like DCT blocks used in JPEGs. here's an example. It works in a similar way though, presenting enough of the right cues to allow our eyes and brains to fill in the gaps.

Re:puzzle?

[Anonymous Coward]

I printed out my application code and pasted it on the puzzle pieces. Then I reassembled it in four different ways. Makes just as much sense each way, my boss said. He will die for that. Soon.

Re:puzzle?

[fbjon]

Not only that, but that example just shows how the individual frequencies look. In a nutshell, the actual blocks in the picture are composed of sums of several of those little ones, so in order to make a puzzle like that you'd have to stack transparent pieces. Which would be pretty cool now that I think of it.

Re:puzzle?

[Nathrael]

Sounds a lot like glorified real-life Paint to me.

Re:puzzle?

[Hognoxious]

Or a high res version of a tangram.

Looks familiar

[aBaldrich]

The puzzle version of ascii art?

Re:Looks familiar

[davidavdav]

A slightly coloured version?

Re:Looks familiar

[Anonymous Coward]

Racist!

Re:Looks familiar

[Tynin]

No, it is the puzzle version of Post-It art.

Re:Looks familiar

[ILuvRamen]

Well I don't know about that. I've used some converters and it was a little rough. 300 pixels is better resolution than some old cell phone screens lol. I think you can do better than ASCII art level with this thing.

Re:Looks familiar

[Anonymous Coward]

I hope you saw "Worldwide Patent P." printed on the box... so no more of that infringing Post-It (tm) please!

 

Re:Looks familiar

[Yvan256]

It's 300 pixels total. I don't think I've ever seen a cellphone with an LCD display with a resolution lower than 17x18 pixels.

Re:Looks familiar

[allcaps]

only slightly racist

infinite?

[token_username]

More like 300! I'd say.

Re:infinite?

[Anonymous Coward]

Assuming a fixed aspect ratio.

Re:infinite?

[Anonymous Coward]

Erm, with 300 pieces there are definitely more than 300 combinations of pieces. Considering you can put it into any shape. That's like saying you're 1024 pixel monitor can only show 1024 different pictures.

Re:infinite?

[token_username]

! means factorial dude.

Re:infinite?

[Anonymous Coward]

300! means 300 factorial. Ie, 300*299*298*...*1

Re:infinite?

[Anonymous Cowpat]

there was a exclamation mark. 300 factorial, 300 x 299 x 298 x 297x ... x 2 x 1.

Incidentally, if anyone wants to calculate that, you'll need to use a float, and probably a double-, or quadruple-precision (YMMV) one at that.

Re:infinite?

[Anonymous Coward]

3 E 614 is pretty close to infinite.

Re:infinite?

[Anonymous Coward]

You forgot to include rotations. Its 1200!

Re:infinite?

[Zocalo]

Not necessarily. There are 300! sequences of the pieces, but you also need to allow for a few more variables:

• Rotation of pieces that are non-uniform in colour
• Multiple arrangements - 15x20, 10x30, images etc.
• Not using all the pieces - a 17x17 image for instance
• Combining multiple sets of the same colour
• Combining multiple sets of different colours

OK, the last two are technically cheating, and all but the first option would possibly require custom code since all the example images appear to be 15x20 portraits, but a suitable algorithm probably wouldn't be that hard to figure out. I saw this on Firehose last night and worked out a few likely routines this morning, so I'd expect some custom FL/OSS code (and cheap Chinese manufactured knock-offs) to be available in fairly short order. After that the race will be on to create the largest most impressive image before the fad inevitably passes.

Re:infinite?

[nneonneo]

Assuming all pieces are used, and that none of the pieces are symmetric or identical (that is, all pieces are different, and each rotation is different), then the actual number of possible images comes out to:

9*(4^300)*(300!)

where 9 is for the number of possible rectangles (1x300 up to 15x20), 4^300 accounts for the rotations of each piece, and 300! accounts for their arrangement.

The result, according to Python, works out to around 1.143*10^796, which is large, but not infinite.

Re:infinite?

[Anonymous Coward]

just use lisp

Re:infinite?

[1729]

there was a exclamation mark. 300 factorial, 300 x 299 x 298 x 297x ... x 2 x 1.

Incidentally, if anyone wants to calculate that, you'll need to use a float, and probably a double-, or quadruple-precision (YMMV) one at that.

If you're computing an integer-valued function, the result should be an integer. In Python, which uses arbitrary precision integers by default, it's as simple as:

>>> import math
>>> math.factorial(300)
306057512... [truncated to get past the lameness filter]

Re:infinite?

[Anonymous Coward]

Over 9000!

Re:infinite?

[sexconker]

Assuming it's a standard 4-sided, 1 connection per side puzzle piece, and discounting gradient pieces (rotations), and assuming each piece is unique (since I don't know the number per colors)...

Piece 1 can go in 1 place.
Piece 2 can go in 4 places.
Piece 3 can go in 6 places.
Piece 4 can go in 8 places.
Piece 5 can go in 10 or 9 places.

Etc.

Maximum ridiculousness results in 600 place choices for the last piece. This gives a loose upper bound of 600*598*596...*4*1 / 4 (cardinal rotations) total layouts. ( = 300! * 2^297 layouts.) This is a loose upper bound because of the fact that when placing a piece you will often connect to more than 1 existing piece. Simply substitute 2^297 with x^299 / 4, where x is your average edge growth factor.

There are only 9 rectangular layouts.
300x1
150x2
100x3
75x4
60x5
50x6
30x10
25x12
20x15

We're looking at 9 * 300! sane builds.

We're looking at 300! * 300! * X^299 / 4 for free-layout builds, where X is the average edge-growth factor (and is between 1 and 2).

! indeed.

Re:infinite?

[NiteMair]

Don't forget the whole thing can be turned over also.

Re:infinite?

[jameskojiro]

But 41,000,000,000 is the largest number in Maths.

Some have speculated there may be a larger number: 41,000,000,001?

Re:infinite?

[sexconker]

Turning it over results in the same thing, as does rotating it.
You CAN'T flip the individual pieces though (they won't connect if you do).

I'm also assuming you use 300 pieces - no more, no less.

Re:infinite?

[Anonymous Coward]

What are birds? We just don't know.

Re:infinite?

[Anonymous Coward]

where 9 is for the number of possible rectangles

This also assumes that you're building a rectangle.

Re:infinite?

[vikstar]

where 9 is for the number of possible rectangles

This also assumes that you're building a rectangle.

and that all your pieces are attached, if you allow arbitrary separation then the number of combinations is infinite.

Re:infinite?

[itslifejimbutnotaswe]

Who says you have to use all the pieces? You'd also need include all the other sizes that one could produce with few pieces (and then the selection of that subset of pieces...

Re:infinite?

[Anonymous Coward]

Depending on your view on the infinitude and quantization of the universe...

Re:infinite?

[maxume]

At worst that just doubles it (because every possible arrangement of a subset is going to be part of one of the arrangements of the full set).

Re:infinite?

[Etcetera]

there was a exclamation mark. 300 factorial, 300 x 299 x 298 x 297x ... x 2 x 1.

Incidentally, if anyone wants to calculate that, you'll need to use a float, and probably a double-, or quadruple-precision (YMMV) one at that.

C1 = 300!

R1 = 3.060575122185 x 10 ^ 614

The joys of hypercalc :)

Re:infinite?

[v1]

the majority of the pieces are solid monochrome, making their orientation irrelevant

Re:infinite?

[Anonymous Coward]

They made a special word for this kind of situation: transfinite

Re:infinite?

[nneonneo]

If we still assume rectangular layouts, this is still not too hard to do. We simply sum d(n)*(4^n)*(n!)/2 for each n from 1 to 300, where d(n) is the number of divisors of n.

Dividing by 2 eliminates the upside-down variant of each image, since we don't care about the orientation of the whole image (however, we do distinguish the horizontal and vertical orientations).

The result of the full summation, again according to Python, is 1.1432*10^796, which, comparing with the previous result of 1.14299*10^796, is really not too far off (the main reason is that removing just one piece causes the number of permutations to shrink by a large factor, e.g. 300! is 300 times bigger than 299!).

We are still, however, assuming a rectangular layout. If we allow arbitrary edge-connected layouts, the difficulty of the enumeration problem increases substantially (if we remove the edge-connected constraint then the number of layouts is effectively infinite).

Re:infinite?

[camperdave]

Piece 1 can go in 1 place.
Piece 2 can go in 4 places.

I'll buy that, so far.

Piece 3 can go in 6 places.

Here it breaks down. Suppose you place piece one in the center, and piece two on top of piece one. Piece 3 can go:

1. On top of piece 2
2. To the right of piece 2
3. To the right of piece 1
4. Below piece 1
5. To the left of piece 1
6. To the left of piece 2

Now true, that is six places, but for only one of the layouts of pieces one and two. Suppose you place piece one in the center, and piece two to the right of piece 1. Piece 3 can go:

1. On top of piece 1
2. On top of piece 2
3. To the right of piece 2
4. Below piece 2
5. Below piece 1
6. To the left of piece 1

Another six layouts. In fact, there will be six layouts for each of the four layouts of the first two pieces.

Unfortunately, this method of counting then becomes unwieldly. Consider the following layout of four pieces:

Piece one is in the center.
Piece 2 is above piece 1.
Piece 3 is to the right of piece 2.

Now piece 4 can go:

1. On top of piece 2
2. On top of piece 3
3. To the right of piece 3
4. Below piece 3
5. To the right of piece 1

Wait a minute. Those last two are the same configuration. You're going to get eight configurations of those four pieces that are identical. The problem of duplicate layouts increases with each additional piece. Your count may yield a maximum, but it won't yield a true count.

Re:infinite?

[rockNme2349]

if you allow arbitrary separation then the number of combinations is infinite.

+3 Interesting? Really?

It could only have been worse if it was Informative.

Re:infinite?

[MichaelSmith]

More like 300! I'd say.

Thats pretty damn close.

Re:infinite?

[RichardJenkins]

300! combinations at 300x1
300! combinations at 150x2
300! combinations at 100x3
.
.
.

Re:infinite?

[kklein]

But 41,000,000,000 is the largest number in Maths.

No way. 24 is the highest number. Fugeddaboutit.

http://www.youtube.com/watch?v=RkP_OGDCLY0

Re:infinite?

[Tacvek]

306 057 512 216 440 636 035 370 461 297 268 629 388 588 804 173 576 999 416 776 741 259 476 533 176 716 867 465 515 291 422 477 573 349 939 147 888 701 726 368 864 263 907 759 003 154 226 842 927 906 974 559 841 225 476 930 271 954 604 008 012 215 776 252 176 854 255 965 356 903 506 788 725 264 321 896 264 299 365 204 576 448 830 388 909 753 943 489 625 436 053 225 980 776 521 270 822 437 639 449 120 128 678 675 368 305 712 293 681 943 649 956 460 498 166 450 227 716 500 185 176 546 469 340 112 226 034 729 724 066 333 258 583 506 870 150 169 794 168 850 353 752 137 554 910 289 126 407 157 154 830 282 284 937 952 636 580 145 235 233 156 936 482 233 436 799 254 594 095 276 820 608 062 232 812 387 383 880 817 049 600 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000

Re:infinite?

[ChrisCampbell47]

The result, according to Python, works out to around 1.143*10^796, which is large, but not infinite.

37 minutes.

As I read "infinite" in the summary, I thought "OK, let's see how long it takes for one of these yahoos to calculate how many combinations there really are", since it is of course not infinite. The post went up at 6:02pm, and the parent of this post went up at 6:39pm. Congratulations :)

Re:infinite?

[ddegirmenci]

How about rotating the image as a whole? so.. 4 x 300!

Re:infinite?

[Nazlfrag]

Looking at this it seems most if not all of them do have a gradient, just a majority have a subtle one.

Re:infinite?

[derGoldstein]

You people and your "math speak"... Maybe he was just excited to type the number 300? Did you ever consider that? Huh?

Also, he could have meant the bitwise operation. Which means it could have been interpreted as "300... NOT!".
Broaden your horizon, dude.

Re:infinite?

[derGoldstein]

Higher precision, please.

Re:infinite?

[Anonymous Coward]

That is, of course, if he started the calculations right after the article was posted right? In any other case your calculations worth mmm lemme see...

(37 minutes later)

NOTHING!!!!

Re:infinite?

[Anonymous Coward]

if you allow arbitrary separation then the number of combinations is infinite.

Nice try, but there are still upper and lower bounds on the distance, since you can't separate the pieces more than (just slightly less than) the width of the Universe, or overlap particles. Still finite, just a much bigger number.

Re:infinite?

[rune.w]

You could probably get more than that if you rotate 90 degrees on the same spot those pixels that have a color gradient.

Re:infinite?

[Anonymous Coward]

Plus all the permutations that follow from rotating 180 and 270 degrees! ;)

Re:infinite?

[Anonymous Coward]

Well, according to Wikipedia, "contrary to statements sometimes found in the popular press, there is no evidence to suggest that distances in space are quantized in units of the Planck length," so... fine, you win this time.

Re:infinite?

[bennomatic]

Hey, that's the key for BluRay disks! You better take that down before the MPAA comes after you.

Re:infinite?

[Entropy98]

The result, according to Python, works out to around 1.143*10^796, which is large, but not infinite.

This brings up an interesting point. Although infinities exist, there are some things that are very large, but not infinite.

Take a 1 or 2 mega-pixel image (or bigger or smaller). There is a finite number of possibilities. It is almost like there is a finite amount of things to see. What if you had time to see all those?

The same applies to DVDs, or Blu-rays. What if you had time to see every possible DVD?

Would there be anything left to see?

If my computer could create truly random videos I guess with enough time I would get Star Wars on Blu-ray.

Re:infinite?

[Phantasmagoria]

Actually, you should divide by 4 (not 2), to eliminate the 3 extra 90 degree rotations of the same final image.

Re:infinite?

[bennomatic]

So why, exactly, do all final images have to be in a rectangle? I want to make my image in a rough hexagon shape. Or a star or circle.

And why do all of the pixels have to be connected sequentialy? Maybe I have an image I want to make that has a lot of white space, so I build it on a white table that is sparsely populated, but a picture nonetheless.

Hell, maybe I want to stand some of the pieces on their sides, to add a delicate interplay of light and shadow. Or stack some of them in various vertical formations.

I can't prove that it's infinite, but I bet you a shiny new penny that for every finite set of combinations you can describe, I can come up with one more idea.

I know, I'll mount the pieces on little chunks of styrofoam and float them on top of a koi pond in to which I've put 300 scuba diver toys, leave and let the toys re-arrange the pieces randomly, come back and take a picture. I'll bet that combination won't be one of your 1.143*10^796.

Re:infinite?

[Arlet]

It looks like every piece can also be rotated

Re:infinite?

[Trogre]

Indeed, but I would be truly impressed if anyone could regard that number of combinations anything but infinite to all intents and purposes.

Re:infinite?

[ed314159]

They say that a picture is worth a thousand words. However, if we consider that a word is 16 bits and therefore has 2^(16) possible states, 1.143*10^796=2^(16*w) => w = 71.8 Apparently, a picture is worth less than a hundred words. Who knew?

Re:infinite?

[Anonymous Coward]

No. Each piece can be rotated individually, meaning you have to multiply your huge-ass number by 4^300

Re:infinite?

[Hurricane78]

Still shorter than the time you took to generate your message. What did you use? VoiceXML with a dial plate phone to a virtual machine on the oldest phone VMWare mobile supports, running Linux which is running Emacs which is running another virtual machine running a browser with a JavaScript virtual machine interpreting raw C code simulating a whole Apache server with PHP interpreting Ruby with the VoiceXML application?

Congratulations :)

Re:infinite?

[Anonymous Coward]

The GP was his own post...

Re:infinite?

[StripedCow]

41,000,000,000 ought to be enough for anybody!

(but please don't truncate my bank-account)

best regards,
Bill

Re:infinite?

[Anonymous Coward]

You forget the human visual system. The same combination of tiles maybe not always be perceived the same way.

Re:infinite?

[L4t3r4lu5]

*** Warning! Warning! Political satire imminent! Warning! Warning! ***

America wishes its national debt was $41,000,000,000!

Zing!

Re:infinite?

[TheRaven64]

Please never apply for a job writing software in the financial sector.

Re:infinite?

[TheRaven64]

a word is 16 bits

I think you might want to get a newer computer...

Re:infinite?

[RockDoctor]

You people and your "math speak"...

Which would have been more impressive? The defeat of 300 Spartans by the million-strong army of Xerxes, or the defeat of 300! Spartans by a million-strong Persian army? (In the first case the Spartans are outnumbered 3000 to one ; in the latter case, the Persians are outnumbered some 10^600 to one.
That's a lot.

Re:infinite?

[Eraesr]

one, two, lots, many

Re:infinite?

[Steve Max]

If the values you're working with are of the order of one, then yes: 3x10^614 would, for all accounts and purposes, be infinite. But if you're working with values of the order of 10^1000, then 3x10^614 would be pretty much zero.

One of the smartest physicists who has never been awarded a Nobel prize I've ever known used to say that "big", "small", "near", "far", etc are meaningless words if used by themselves. They have only meaning when youput them in context; 3x10^614 is big compared to one, but small compared to 10^1000, just like the Sun is very close to us compared to the nearest quasars, but very far away compared to the nearest cloud.

Re:infinite?

[LordKronos]

At first I thought the number was wrong because of all the zeros at the end, then it quickly dawned on me that with all the 10s and 5s you should expect a bunch of zeros. However, trying to verify the quantity of zeros, I'm having trouble figuring out where they come from (I double checked with perl's Math::BigInt library, and your number is correct...or at least the quantity of zeros is).

There are 74 zeros there. Obviously we can expect 1 zero for each multiple of 10 we multiply by (so that's 30 zeros), plus 100, 200, and 300 each give an additional 0 (thats 3 more zeros). We also get a zero for each 5 that's multiplied by a 2, 4, 6, or 8, though we're limited by the number of 5's so it doesn't matter which number you choose (that gives us 30 more zeros).

So, we've got 30 + 3 + 30 = 63 zeros. Where do the other 11 come from? I'm feeling kind of dumb about it (probably overlooking something very obvious), but I just can't figure it out.

Re:infinite?

[hey!]

My version of the puzzle can generate infinite numbers of pictures. All the pieces are white and it comes with a box of paints.

It's part of my line of "can't lose" Christmas toys, along with the ever popular "No-Miss Target Shooting (Barn not Included)."

Re:infinite?

[LordKronos]

It figures....I thought about this a long time before submitting and couldn't come up with anything. As soon as I submit, it dawned on me that each multiple of 25 actually counts as an extra 5, though the multiples of 100 already exhibit the extra 5. So we get 25, 50, 75, 125, 150, 175, 225, 250, and 275 giving 9 more zeros. 2 left to go. 125 and 250 also count as a third 5, so that's our last two.

That was a fun and pointless exercise.

Re:infinite?

[Anonymous Coward]

Ah, but I would consider two arrangements that differ only by a rotation over 180 degrees to be the same. (90 degrees can't happen because you allowed only 9, not 18 rectangles.) So I'd say the correct answer is 9*2^599*300!.

Now what if you allow arbitrary shapes with holes instead of just rectangles...

Re:infinite?

[Permutation Citizen]

You may read Jorge Luis Borges, "The Library of Babel".

Re:infinite?

[Anonymous Coward]

That's like saying you're 1024 pixel monitor...

I am not!

Re:infinite?

[omarius]

I think about this from time to time. Scarily, it appears that some oddball things are calculable in information theory. But it' can be somewhat obvious, like your point about DVDs. On a DVD, a movie is just a binary number, so, ergo, there are a finite number of possible movies. This goes for music, too. I've also wondered if there's a finite number of, say, playable games of chess. It's an odd but interesting thing to think about.

Re:infinite?

[Anonymous Coward]

Is this correct? You count only half the number of rectangles....1x300 to 15x20, but not 20x15 up to 300x1...those other 9 will be duplicates of the first 9, only if you rotate the entire puzzle. All 1x300s have an associated 300x1, that will look the same, only if you rotate the entire puzzle, etc..., but if you are not counting those, because they are rotationally similiar, then why are you counting the rotationally similiar 1x300s, etc.

Example: take any 1x300, and rotate all pieces 180 degrees, you now that the first 1x300 but upside down (rotationaly simliar) You count those, but leave out the other instances of rotationally simliarity. It seems to me that you either need to count them all(so that the number of possible combinations is true, but only from one perspective (no rotating the puzlle allowed)), or exclude all rotationally similiarities...Ideas?

It seems to me that the answer, only counting rectangles, is either 18*(4^300)*(300!) if you only ever look at the puzzle from one orientation.
or
if you decide that you can rotate the entire puzzle so that you do not count patterns that are rotationally the same, you get (18*(4^300)*(300!))/4 different patterns

I'll try another example:(only using 4 pieces...) M, W, an E are supposed to represent the same thing, only rotated.
You include:
MMMM and WWWW
those are rotational simliar; they are just rotated 180 degress. This represents the 1xN

You do not however include: this represents the Nx1
E
E
E
E
which is rotational simliar to MMMM rotated 90, counter-clockwise

Re:infinite?

[omarius]

Still finite if you factor (npi) in the human JND for color and the minimum size of a blob of paint!

Re:infinite?

[d34dluk3]

It's actually 1/2 of that, as you've failed to account for the 180 rotation of the entire picture. (i.e. a picture which is identical is constructed twice by your algorithm, once right side up and once up side down.

Re:infinite?

[saderax]

.0

Re:infinite?

[JimFive]

On a DVD, a movie is just a binary number, so, ergo, there are a finite number of possible movies

What? The set of binary numbers is finite? The set of binary numbers that will fit on a DVD is finite. I'm not sure that there is a finite number of ways of interpreting that number into video, however.

I've also wondered if there's a finite number of, say, playable games of chess.

Yes, there is a finite number. No, I don't know what it is. Chess rules limit the game in certain situations. A position can only be repeated 3 times before the game is a draw. If there is no piece capture or pawn move for 50 moves then the game is a draw. This means that there is a longest game. If there is a longest game then there must be a finite number of games. I have read that the longest possible legal game is 5949 moves, but I don't know how this was calculated.
--
JimFive

Re:infinite?

[Anonymous Coward]

WHAT??? We gave the banks more money than the largest number!!! Also, they almost gave themselves a bonus as big as the biggest number.

Re:infinite?

[sexconker]

I did in fact state that it was a loose upper limit.
And then I stated that the true could be calculated from the average edge growth factor, which is somewhere between 1 and 2.

No, I'm not going to try and figure it out.
No, I'm not going to try and throw out all duplicates.

I just know it's between 1 and 2.

Re:infinite?

[Anonymous Cowpat]

What I should have mentioned was that that advice was for use in fortran, where no integer which any compiler will support by default, will hold a number of that size. Thus you need to use a float. I once tried using an integer to get 100! and got wierd results for a few minutes before I twigged that the number was WAAAAAY too big even for an integer(kind=8), and switched to a real(kind=8).

Perhaps my mistake was to call it a 'float', rather than a 'real'.

Yes, I know it won't give you accuracy to a single figure, but it will give you a decent idea of the size of the result, and will actually be able to be held in memory without using either (a) enormous variables, or (b) higher-level languages which will use enormous variables without telling you.

Re:infinite?

[Anonymous Coward]

You have to multiply this by 2. While this is all the images that can be made, that image can be displayed in landscape and portrait orientations.

Re:infinite?

[element-o.p.]

Yeah, I forgot the tag at the end ;)

Re:infinite?

[Anonymous Coward]

This assumes that every combination produces what one considers to be a "picture" (aside from just visual noise). No script can calculate this, since it differs for every person.

Re:infinite?

[dlgeek]

You're not sure if there a finite number of interpretations of a finite set? Of course there are. Anyway, in the case of a DVD, you have a finite range of numbers (from 0 to approximately 2^40,372,692,583 or so), and every number is either a valid DVD video (meaning it can be decoded properly) or not (most numbers).

Wikipedia says that the number of legal positions in chess is estimated to be between 10^43 and 10^53 and the number of possible games is approximately 10^123. In contrast, it is beleived that there are only about 10^80 atoms in the universe, so it's pretty much impossible to enumerate all the games of chess.

Re:infinite?

[GigaHurtsMyRobot]

I think you're missing that 100 200 and 300 don't add 1 zero, they add 2 each.

Re:infinite?

[GigaHurtsMyRobot]

No you weren't missing anything, I'm just an idiot.

Sweet

[Narpak]

Now tell me the pattern for creating an image of unspeakable evil; like the Great Cthulhu.
Cthulhu fhtagn! Cthulhu fhtagn! Ia! Ia! Ia! The sleeper awakens!

Re:Sweet

[Red Flayer]

Impossible. Anyone who had seen the image would be a gibbering heap of insanity, and unable to tell you the pattern. At best you could hope to get enough clues to figure out the pattern yourself... but if you assembled it, you'd either off yourself or also turn into a quivering mass of human flesh.

The key here is to get someone else to assemble the image... you'd find a likely mark (some kind of paranormal investigator, for instance) and then mislead him into thinking the image he's assembling will *stop* the summoning of Cthulhu. Drop enough clues in the right places, use decoys to mislead him of your true intentions, let him be an ignorant pawn in your great game. With luck and skill, you can get him to do the dirty work for you. And the irony of him contributing to the Great Awakening by striving against it is quite delicious.

At least, that's the way I'd do it. Your way is too direct, and not worthy of true evil genius.

Re:Sweet

[CptPicard]

You must be new here, it should be obvious. Send the company a picture of goatse, and have your pattern...

Re:Sweet

[Enleth]

You know a CoC game master when you see him.

However, you need some more experience to be able to truly become the one behind all the evil geniuses the investigators try to stop.

In this case, one of the investigators should become a shizophreniac due to past mental strain and suffer from terrible nightmares and symptoms of sleep deprivation even though he goes to sleep each night (or so he thinks), and at the very end it should be revealed (that is, if the other investigators are still alive and doing well enough to reluctantly let them win), Fight Club-style, that the evil genius and leader of the local Cthulhu cult was the investigator's secret second personality acting at night, and not only using himself (that is, his original personality) as an ignorant pawn, but also his friends who join the "investigation".

And this *is* true evil genius.

Re:Sweet

[Anonymous Coward]

Cthulu has already attacked Stack Overflow /. is next!

Re:Sweet

[Anonymous Coward]

i wish i still had mod points!!!

Re:Sweet

[Anonymous Coward]

Some days I think I am a geek....then people like Red Flayer post a message and I realise that I have a long way to go!

Re:Sweet

[Boronx]

All through '43, we had separate puzzlers each working on a single piece, for safety. One time, a puzzler saw the piece of another and had to be hospitalized for three weeks.

Re:Sweet

[MichaelSmith]

All through '43, we had separate puzzlers each working on a single piece, for safety. One time, a puzzler saw the piece of another and had to be hospitalized for three weeks.

What did the Germans come up with? I bet it was so well documented that it couldn't send anybody mad.

Re:Sweet

[SEWilco]

What did the Germans come up with? I bet it was so well documented that it couldn't send anybody mad.

We couldn't find their puzzle among the puzzle pieces of Berlin. We did find the documentation, but unfortunately that's enough to drive men mad also.

Re:Sweet

[jbezorg]

Shoggoths Wanted

Re:Sweet

[Artifakt]

Naah! You'd probably end up with your 'likely mark' being Bob Howard, and since he isn't cleared CASE NIGHTMARE GREEN, the wards in his personnel file would keep him from turning into anything, and since, as an uber-geek, Bob is already pretty much a quivering mass, a paradox would result, a paradox big enough that Cthulhu would be the one who would have to turn into something instead.

Re:Sweet

[Maxo-Texas]

They tried again in the 90s and used python. But the coders all went mad and ran around without any cloths on.

I hear some of those python full monty coders still turn up now and then.

Re:Sweet

[fru1tcake]

Better still, send them a link that points to goatse, but then redirects to a Youtube clip of Cthulhu singing Rick Astley's "Never Gonna Give You Up". That'll show em!

Re:Sweet

[tomtomtom777]

You must be new here, it should be obvious. Send the company a picture of goatse, and have your pattern...

I'm not sure that is possible. From google translate of the second link:

However, this fact "made a slight tint of color patterns," is what, this Jigazopazuru liver. By utilizing this pattern, with a whopping "Self Portrait" I would make the. Moreover, not only their faces can also face other people. Even great men in history, even heterosexual love, even in the face of the pet ... "If life on Earth, even in the face any" I would make! What kind of mechanism and say something, but ... Well anyway, let's say that you actually try.

Re:Sweet

[neurovish]

All through '43, we had separate puzzlers each working on a single piece, for safety. One time, a puzzler saw the piece of another and had to be hospitalized for three weeks.

What did the Germans come up with? I bet it was so well documented that it couldn't send anybody mad.

"My dog doesn't have a nose"

Re:Sweet

[phaggood]

Chilluns. *sigh*

The article's in Japanese; so where do I in the US get one in time for Christmas for my puzzle lovin' kin?

Well... not infinite.

[pwnies]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Fixed that for you.

Re:Well... not infinite.

[VagaDragon]

Actually it's 1200! (assuming every piece can be rotated 4 different ways to form a unique "pixel") or 6.35078909 * 10^3175

Re:Well... not infinite.

[girlintraining]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Most of which will resemble little more than random noise and have no value.

Re:Well... not infinite.

[Anonymous Coward]

Isn't it 300! (number of complete board possibilities) * 4^300 (rotations of all pieces)

1200! implies after you place one a certain way, you have 1199 more possibilities, which is untrue; you have 1196.

1200 * 1196 * ... :)

Re:Well... not infinite.

[nneonneo]

I'm not sure where you are getting 1200! from. It should be 300!*(4^300), because the piece rotation is independent of their ordering, and that works out to 1.27*10^795.

Re:Well... not infinite.

[Arthur Grumbine]

And 300 pixels are worth 3.060575122 * 10^614 pictures

I'd consider that a sufficiently small value of infinite.

Re:Well... not infinite.

[Anonymous Coward]

Not quite. Assuming:

• arrangements use every piece,
• every piece is a different color, and
• no piece has a gradient with rotational symmetry, then

there are 300! ways to select the pieces, times 4^300 rotations, times 18 possible aspect ratios.

Re:Well... not infinite.

[sexconker]

That's only if each piece is a gradient piece, and the gradient is directed across one of the cardinal rotations.

And only if each piece is also unique.

And it's still wrong.

You can place them in multiple layouts.
I did the math above, but in summary:

Assuming it's a standard 4-sided, 1 connection per side puzzle piece, and discounting gradient pieces (rotations), and assuming each piece is unique (since I don't know the number per colors)...

We're looking at 9 * 300! sane builds.

We're looking at 300! * 300! * X^299 / 4 for free-layout builds, where X is the average edge-growth factor (and is between 1 and 2).

Re:Well... not infinite.

[node 3]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Fixed that for you.

Actually, your value is just an approximation (ignoring the rotation issue others have already brought up).

In terms of even an exceptionally long-lived human's lifetime, your value is equivalent to infinite.

Re:Well... not infinite.

[Zalbik]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Fixed that for you.

But many* of those pictures are quite rude, as evidenced by the puzzle I'm just completing...OMG...what are you doing to that poor kitten?!?!?

* many as in a large number as opposed to a large percentage.

Re:Well... not infinite.

[Anonymous Coward]

I wish I had that much cocaine

--the comicjk cocaine troll

Re:Well... not infinite.

[Anonymous Coward]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Most of which will resemble little more than random noise and have no value.

I almost want to make a java version of said puzzle, and let people run random combinations on my website. It would be interesting (much like watching clouds) to observe "signal" being made from "noise"

Re:Well... not infinite.

[new500]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Most of which will resemble little more than random noise and have no value.

Looked at from what angle?

Re:Well... not infinite.

[molecular]

And 300 pixels are worth 3.060575122 * 10^614 pictures

Most of which will resemble little more than random noise and have no value.

Yes, and some will show what the future looks like in 50 years.

The hard part is the selection.

Re:Well... not infinite.

[Anonymous Coward]

And *you* are assuming the same size rectangle in every case. It's actually much, much higher than that.

JPEG

[Quietust]

Seems remarkably similar to how JPEG compression works. Not surprisingly, the resulting pictures look a lot like overcompressed JPEGs.

Re:JPEG

[Anonymous Coward]

Whoa? Those puzzles enable to do a discrete cosine transformation and run length encoding with them?

Re:JPEG

[nneonneo]

JPEG chunks an image into 8x8 blocks. An overcompressed JPEG contains so little information per block that the blocks devolve into simple gradient patterns (try this yourself with a grayscale image: save it with a quality near "0" and you will see the individual blocks clearly). If you think about it a bit, this makes sense: the block is being approximated by a combination of a small number of cosine waves (in the limit, it's a single wave along each image dimension), so the result is a gradient, because most of the coefficients have been thrown out by compression.

In this sense, the puzzle pieces can be thought of as representing these simple block patterns. With a 15x20 rectangle of pieces, by JPEG standards, this is essentially an overcompressed 120x160 image. You'll note that if you take your overcompressed JPEG and scale it down to around 25% (30x40), then, provided the original image shows only a single subject, it should still be reasonably recognizable, because the human visual system patches together the pieces to produce a coherent image, even if it is highly distorted.

Re:JPEG

[zippthorne]

Yeah, I was thinking the same thing. Also, that it would be interesting to actually create a jpeg puzzle (monochrome, of course), with higher-order blocks than just simple gradients. Also, what would be the best distribution (vocabulary if you will) of blocks to fit certain kinds of pictures.

Re:JPEG

[camperdave]

Do you have a link to a site that explains how jpegs work, for those of us not in the know?

Re:JPEG

[nneonneo]

Wikipedia explains it fairly well, though it is still a bit technical. The Huffman coding details are not important to the main idea; what is important is the concept of using the Discrete Cosine Transform (DCT) and a subsequent quantization step to reduce the precision, and therefore size, of the high-frequency components.

At extreme compression levels, only a few of the patterns in this image will end up with non-zero coefficients: specifically, those which are low-frequency (in the top left corner). You'll notice that these also happen to resemble gradients, which is what I meant when I said that the 8x8 blocks devolve into simple gradients.

Re:JPEG

[sootman]

Rats, you beat me to it, but yeah, that's exactly what I thought when I saw their pics. Wikipedia has a good intro to how JPEG compression works. Basically, images are cut into 8 pixel by 8 pixel squares, each of which is a kind of gradient, and when you look at all these blocks together, it looks like a real image--kind of like how you can take a bunch of 1-inch line segments, arrange them end-to-end into a shape, and from a few feet away it looks like smooth curves.

a few pictures are worth a million words

[Bourdain]

a few pictures are worth a million words

Especially when the accompanying text is in Japanese and I can't read it

Re:a few pictures are worth a million words

[Anonymous Coward]

There's an app http://translate.google.com/ for that!

Re:a few pictures are worth a million words

[ELitwin]

Why is this modded funny? The full quote is "This article is in Japanese, but as they say, a few pictures are worth a million words." Am I missing something?

Re:a few pictures are worth a million words

[Anonymous Coward]

Moreover, not only their faces can also face other people. Even great men in history, even heterosexual love, even in the face of the pet ... "If life on Earth, even in the face any" I would make! What kind of mechanism and say something, but ... Well anyway, let's say that you actually try.

I happen to read Japanese fluently, but this was worth it.

Re:a few pictures are worth a million words

[RealGrouchy]

Yes, what you're missing is what you're not missing. Like the GP, I had only read the link text before clicking it.

And since I also thought of the same wisecrack as GP (who got to it first), the transitive property of humour makes it funny(?)

- RG>

Re:a few pictures are worth a million words

[Anonymous Coward]

/. has been trolled. Again. HAND!

Re:a few pictures are worth a million words

[hey!]

...now is an amazing "Still not decided in advance to complete" that is the emergence of a mysterious puzzle Nantomo!

Mysterious puzzle, indeed.

But...

[Anonymous Coward]

can it

$ make linux

?

Not quite.

[girlintraining]

It appears to be monochromatic and it also used nearest-approximation algorithms... Which means that the extra pieces are inserted as "random noise" once the general shapes are mapped out. Clever, but... low resolution.

Re:Not quite.

[Kazoo the Clown]

I suppose if you tweaked the gamma and contrast enough you could make anything work without inserting "random noise"...

Re:Not quite.

[Kazoo the Clown]

It essentially has a fixed histogram. I wonder what you'd get back from them if you sent them an image specifically designed to be hard to fit into that histogram...

Re:Not quite.

[Monkeedude1212]

Simply increase the puzzle size, a small tweak to the algorithm (unless it accomodates) and the resolution is fine once again

Re:Not quite.

[rm999]

If the histogram is wider (higher variance) than your image, than they can "stretch" your picture out by upping the contrast. If it is narrower (less variance), than the noise approach is probably the best solution.

If your image has a totally different shape (e.g. a few white patches on a black background), find a new image :P

Re:Not quite.

[SpinyNorman]

There's different ways you could match the pieces to a given photo.

One way would simply to map from the levels curve (pixel brightness vs # pixels plot) of the "puzzle" pieces to the levels curve of your photo. i.e divide the area under your photo's levels curve into 300 squares and assign them in black-to-white order to pieces from the "puzzle" also sorted in black to white order. This would work best if the levels curve of your photo is not totally lopsided compared to the curve of the pieces.

Another way, along the lines you suggest, would be to put aside sufficient(but as few as possible) pieces from the puzzle so that the levels curve of the remainder roughly matches that of your photo, assign pieces to the photo as for the first method, then insert the put aside pieces either as random noise or just as a border or somesuch. This may indeed look better for photos that don't well match the complete set of provided pixels.

Wow

[esocid]

You can even make a 404 error out of it!

Oh Come ON!!!

[Anonymous Coward]

Oh come on. I thought the people on this site were somewhat intelligent? Lets apply some slightly advanced algebra and come up with something better than " 300 pixels are worth an infinite number of pictures." The maximum number of ways you can arrange 300 things is 300!, or about 3.06 X 10^614. Granted a very large number, but definitely not infinite. Stop running dumbass shit like I'd expect to find on digg, and post things that actually make sense. We might not all be grammar geeks, but we are most likely math/science/computer geeks.

Re:Oh Come ON!!!

[Chris Burke]

The maximum number of ways you can arrange 300 things is 300!, or about 3.06 X 10^614. Granted a very large number, but definitely not infinite.

Okay but what if there were 301 pixels, would that be infinite?

And anyway, since pedantry loves company, I'll point out that 300! is the maximum number of orderings of 300 things, not necessarily the maximum number of arrangements. How many arrangements there are depends on what you consider the "rules" for a free-form puzzle like this. Since the pieces do have interlocking teeth I'm going to say that minimally the pieces have to be interlocked (otherwise the possible arrangements truly would be infinite to the extent the universe is), but beyond that does it have to have a specific geometry like 15x20? Does it even have to be rectangular, or can it more resemble a game of dominoes?

Re:Oh Come ON!!!

[LOLLinux]

Sure that isn't infinite. But our solar system will be long gone before all 3.06 x 10^614 different arrangements could possibly be all worked through.

Re:Oh Come ON!!!

[_xeno_]

And anyway, since pedantry loves company

And on that note:

Plus, some of the pieces are identical, so swapping them doesn't create two unique images. Some of the pieces are gradients, so that they can be rotated in any of four directions to create a unique image. Others, on the other hand, are a solid color, and since the pieces are both vertically and horizontally symmetrical, the rotation doesn't matter. Both of those need to be taken into account to figure out how many permutations there are that create distinct images.

Re:Oh Come ON!!!

[Anonymous Coward]

I am also interested in that. What is the total number of puzzles you could create assuming all pieces are used and no pieces are identical?

So from one of the first posts,
http://entertainment.slashdot.org/comments.pl?sid=1474488&cid=30396502
we have

9*(4^300)*(300!)

where 9 is the number of rectangles. So now replace 9 with the number of arrangements. What that is, I don't know.

Legos?

[Darkness404]

Is this really any different than using Legos to make pixel art?

Re:Legos?

[Monkeedude1212]

Yes and No. If I assume you mean with a monochrome scale of pieces, it is still different, because each piece is not simply 1 colour. Each puzzle piece has a different pattern and shades to it. (Though some varying very very little).

Unless you meant Lego pieces with different patterns and shades on it, then yes it would be the same.

Re:Legos?

[sexconker]

My parents only let me play with Duplo.

Re:Legos?

[Riktov]

What makes this different from Legos, pixelblocks, ASCII art, or even a JPEG image, is that the selection of pixels/pieces is predetermined, limited, and they must all be used to make the image. For all those other forms you're allowed to pick the closest color value for each pixel.

With this puzzle, supposing you did it manually, scanning row by row, and picking the best-fitting piece for each pixel. It'll look great at first, but soon you'll be running out of good matches and having to choose less and less optimal pieces; you can't say "give me a 45% with a sideways gradient" when you've used them all up. So the algorithm has to consider how to distribute all the pieces throughout the image for the optimum match.

Re:Legos?

[Anonymous Coward]

legos pixel art is another dimension entirely.

Anything?

[Itninja]

You keep using that word. I do not think it means what you think it means.

I would seem that the word "anything" to you means "any monochromatic, low-resolution image".

For those without a decent calculator...

[Anonymous Coward]

300! (factorial) ~= 3.06 x 10^614

That's how many combinations there are, if each piece is unique and is used in the same 15x20 grid each time.

To put that in perspective, there are only about 10^80 atoms in the universe. You would need 2042 bits to represent that number in binary.

So yeah. For all intents and purposes, that's limitless.

Re:For those without a decent calculator...

[alrudd1287]

so you're saying its not even enough combinations for good encryption!

It's just a bad compression algorithm

[Locke2005]

but you can make anything with it. That's like saying you can convert any picture to a 15 by 20 pixel JPEG; technically you can, but the usually the result isn't worth looking at. That said, I'm sure a lot of people will send in pr0n to convert into patterns, just to see what it looks like in ultra-low resolution monochrome.

Re:It's just a bad compression algorithm

[dbIII]

Of course you can. As an example I've changed my signature to your portrait.

Re:It's just a bad compression algorithm

[Anonymous Coward]

...just to see what it looks like in ultra-low resolution monochrome.

Ahhh... that takes me back to long nights with my old 14.4k US Robotics modem and alt.binaries.pictures.erotica

Re:It's just a bad compression algorithm

[camperdave]

Of course you can. As an example I've changed my signature to your portrait.
\'.'/

Locke2005 == Macaulay Culkin?

Re:It's just a bad compression algorithm

[Faux_Pseudo]

This is going to sound like I have way too much time on my hands. Which was true till I got into even more twisted uses of my time.
I can testify that there is a market for low res mono-chromatic pr0n. My website is dedicated to it. While ASCII art pr0n doesn't pay that much it has opened the door to paying gigs that have netted me thousands of dollars in commissions. Rule 34 for fun and profit.

Re:It's just a bad compression algorithm

[TheAlkymyst]

but you can make anything with it. That's like saying you can convert any picture to a 15 by 20 pixel JPEG; technically you can, but the usually the result isn't worth looking at. That said, I'm sure a lot of people will send in pr0n to convert into patterns, just to see what it looks like in ultra-low resolution monochrome.

http://xkcd.com/598/ Solves this problem then?

Rasterbator-like

[An anonymous Frank]

I think it would be (more) interesting if you could also order a set where the pieces are whatever size you want, ..., so that if you want to *sorta* recreate the mona lisa, but on a wall surface that's 4 feet wide by 5 feet tall; those (larger) pieces will be much easier to apply/fix to a wall, than a bunch of printed pages of paper.

wasgij puzzle

[gurps_npc]

This is the opposite of a jigsaw puzzle, so I call it a wasgij puzzle. Jigsaw puzzles only fit one way and you use the picture to aid you in fitting the pieces. This wasgij puzzle fit any way you want and you fit them together to form the picture you desire.

Re:wasgij puzzle

[NiteMair]

Where does one find a wasgij for cutting out one such puzzle?

Re:wasgij puzzle

[prockcore]

Except that you are sent a key, which only fit one way to form the photo you want.

Basically, the puzzle isn't the pieces, those are universal. The puzzle is the pattern that you get sent.

Re:wasgij puzzle

[Anonymous Coward]

Japan, evidently?

Re:wasgij puzzle

[Anonymous Coward]

Wasgij is actually a brand name puzzle; the design on the box shows a sequence of events resulting from the actual image enclosed in the box. Usually, if you find some point in the background of the box-art and try to visualise the scene from that perspective looking towards yourself, it becomes easy to ascertain the edge of the puzzle. From there, it's a conventional puzzle without a guide picture. The most famous of these is this one: http://www.jigsawgallery.com/prodpics/FJ10950.jpg
The actual puzzle is a view of the front of that car (and everything behind it), as seen from the cart-driver's perspective.

Re:wasgij puzzle

[Anonymous Coward]

There's already a form of puzzle called a Wasgij. It's where the pic on the box is the reverse view of what the puzzle makes. So the pic on the box may be a crowd of passengers pointing excitedly towards and above you, and the puzzle is a jetliner doing barrel rolls. Or something less realistic, perhaps.

Re:wasgij puzzle

[Anonymous Coward]

The company that makes the "Wasgij" series of jigsaw puzzles might not agree...

http://www.google.com.au/search?q=wasgij+puzzle&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-GB:official&client=firefox-a

do not taunt happy fun puzzle

[SuperBanana]

It essentially has a fixed histogram. I wonder what you'd get back from them if you sent them an image specifically designed to be hard to fit into that histogram...

A squad comprised of a Ninja, a gradeschool girl with magical superpowers, a vampire, and a giant robot. On your doorstep. With a note that politely says, "Do not taunt happy fun puzzle."

Re:do not taunt happy fun puzzle

[Hurricane78]

Sounds like a typical Japanese sexy party to me... ^^

The puzzle is what they send to you . . .

[Tanman]

you know, the pic. with the puzzle pattern.

monochrome pipapics

[Anonymous Coward]

www.pipapic.org

Pixelblocks

[_KiTA_]

I liked these better when they were called PixelBlocks.

I have a FF1 Fighter and a DQ1 Slime on my desk. :)

Re:Pixelblocks

[xenocide2]

The difference is that these blocks have gradients, so there's a computational optimization aspect to it. Given a set of similar but gradient, how do you arrange and orient them to best reproduce an image? The gradient aspect leads to a better quality image than a simple grid of pixels.

Colorblind

[assemblerex]

Not colorblind friendly for sure! One box of blank pieces please!

Re:Colorblind

[Garble Snarky]

Don't colorblind people have trouble distinguishing different hues? If so, then this toy would be BETTER for colorblind people than virtually anything else - all that matters is the relative intensity of the pieces, not the specific hue of any of them.

love the name

[gEvil (beta)]

I love the name Jigazo Puzzle. It sounds like a Japanese guy saying "jigsaw" but with a thick Japanese accent.

Re:love the name

[Anonymous Coward]

I would think it's an English loanword, and hence your remark is pretty much exactly on point.

Re:love the name

[Riktov]

"Jigazo" means "self-portrait" in Japanese. Clever naming.

Obama

[Anonymous Coward]

lol, I was thinking writing something funny about how long it would take before we saw a puzzle of Obama. Then I clicked on the link... dammit.

Lego Mosaic

[LoudMusic]

Seems an awful lot like the Lego mosaics that people make. Lego also did a mosaic product for a while where you could upload an image and they would send you parts and instructions for making the image with 1x1 Lego plates.

I believe there is even software now to make the 'maps' yourself, much like cross-stitch, etc.

120x160

[StreetStealth]

Mod parent up.

It's the gradients on the pieces, and the principles of human vision that JPEG takes advantage of, that give this puzzle its cool effect, creating the appearance of a much higher resolution than the 15x20 "pixels" everyone else is referring to.

You can't make a (easily) recognizable Mona Lisa in 15x20 pixels. You can in 15x20 cosine gradients.

Error diffusion

[StreetStealth]

If your image has a totally different shape (e.g. a few white patches on a black background), find a new image :P

Even then, you'd probably get something basically recognizable -- I'd imagine the error diffusion just puts a lot of noise in a black area that's too big. Heck, it may even run an unsharp mask over the image to exaggerate details when the predicted output noise reaches a certain threshold.

I bet the algorithms for this bear a number of similarities to photomosaic systems as they're both working with a known set of "subpixel units."

Ken Knowlton's mosaics

[dpbsmith]

Given that the pieces tile exactly as if they were squares, I'm not that impressed.

I'm much more impressed by what Ken Knowlton manages to do with seashells.

If it was not japanese...

[mad flyer]

It would be considered totally retarded... But as usual the 'japan' aura hide the fact that it's nothing more than manual rasterisation...

Yes you can use human to do machine (computer) work... is it always smart/usefull/entertaining in a non masochist way ? not often...

Next up on 'news from japan' two new devices who, when combined can perfectly replace your printer:
pen and paper...

300 pixels are worth 64 words

[Michael Woodhams]

which we can encode in 2042 bits, or 64 words (assuming 32 bit word size.)

(Adjust answer as needed to account for rotations, duplicate pixels etc etc.)

If I worked at that company..

[Anonymous Coward]

The instructions to create 'Beautiful Flower' would just end up with a jigsaw that looked remarkably like goatse.

Racing Car? ... goatse

Cat? .. goatse

Goatse? .. tubgirl.

infinite?

[element-o.p.]

Just a minor quibble, but 300 pixels are not worth an infinite number of pictures. They are actually only worth 300! (or 306 057 512 216 440 636 035 370 461 297 268 629 388 588 804 173 576 999 416 776 741 259 476 533 176 716 867 465 515 291 422 477 573 349 939 147 888 701 726 368 864 263 907 759 003 154 226 842 927 906 974 559 841 225 476 930 271 954 604 008 012 215 776 252 176 854 255 965 356 903 506 788 725 264 321 896 264 299 365 204 576 448 830 388 909 753 943 489 625 436 053 225 980 776 521 270 822 437 639 449 120 128 678 675 368 305 712 293 681 943 649 956 460 498 166 450 227 716 500 185 176 546 469 340 112 226 034 729 724 066 333 258 583 506 870 150 169 794 168 850 353 752 137 554 910 289 126 407 157 154 830 282 284 937 952 636 580 145 235 233 156 936 482 233 436 799 254 594 095 276 820 608 062 232 812 387 383 880 817 049 600 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000) pictures.

Is it just me...

[barfy]

Or does the product sound like Mickey Rooney's character in Breakfast at Tiffany's saying "Jigsaw Puzzle".

It's a "jigazo puzzle".

Now all we need...

[Anonymous Coward]

...is an infinite number of monkeys with an infinite number of typewriters.

Where can I get one?

[Anonymous Coward]

Nice Christmas present... Seems to cost about 1890 yen? $22 or so. Ship to US?

I see a couple on Ebay for $70 (pffft, yeah right). Any ideas?

dithering

[Anonymous Coward]

I want one of these in a special CGA-colors edition. :-)

Re:dithering

[Anonymous Coward]

Well it seems they have a purplish-red and a blue set as well as black, so just get three sets for all your ugly cyan, magenta, black and white dreams to come true.

Stop the "300!" BS!

[SlothDead]

Why does everyone write about "Oh, its only 300! images! Oh noes!11"? It's not 300! combination, it's actually a lot more. Keep in mind that each piece can be rotated four ways which gives you

300! * 4^300

combinations.

Re:Stop the "300!" BS!

[Anonymous Coward]

A know you're trying to be funny, but 300! is factorial, not combinations. With combinations, order doesn't matter.

All i know is:

[cyberzephyr]

I would like to buy one! I think it's unique and cool. I probably would buy a couple for friends. Unfortunately, i coul not get babel fish to translate for an address. :-(

Re:All i know is:

[Mad Bad Rabbit]

Google Translate (http://translate.google.com/#ja|en| ) seems to work well enough
(except for the text embedded in images). The price is 1890 Y ; but as far as I can
determine they don't ship outside Japan.

Re:All i know is:

[cyberzephyr]

Damn! :-(

Thanks anyway. I'll see if i have a friend on the side to see if they get me one. :-)

Multiple sets?

[DarkofPeace]

If you combine multiple sets, do you get better resolution or just a larger picture?

Re:Multiple sets?

[JSBiff]

The answer to your question is: "yes"

I know you phrased the question as either/or, but you see, if you have a larger picture, and you look at it from farther away, it seems to me that it would appear to be higher resolution, wouldn't it? If you looked at it from the same distance, however, it would, I suppose, just look larger.

So, "yes".

Lame

[Anonymous Coward]

Wow this is alot lamer than the summary makes it out to be. Ever heard of a mosaic before? My mom makes them all the time.

view it from far away ...

[J_Omega]

... or squint

They're still worth a kiloword!

[Anonymous Coward]

> Most of which will resemble little more than random noise and have no value.

They have value! They're each worth one kiloword. Though a few of the words might not be very cromulent, but you always get a few bad ones when you buy in industrial sized quantities.

Highly offtopic

[perryizgr8]

I just entered fac(300) and before I could release the calc button I got 16 significant digits. And this is using a java app on my 200mhz phone with music playing in the background and slashdot open in opera mini. Can anyone tell me if it really multiplied 300 times or is it some neat trick?

Re:Highly offtopic

[MathiasRav]

Some of it is table lookup, but regardless, 300 fp multiplications shouldn't take forever. You mentioned significant digits, so it's (unsurprisingly) not arbitrary precision.

But it's not a puzzle anymore!

[Hurricane78]

There are no differing nipples and holes anymore, so you can’t fail anymore. Which means that you can put it together in a wrong way without noticing. Oh, wait, there is a list of how to put it together, killing the whole point of putting a puzzle together.

I don’t think that that way it will have any chances, after this little hype is over.

Re:But it's not a puzzle anymore!

[d34dluk3]

There are no differing nipples and holes anymore, so you can't fail anymore.

This is the worst argument for homosexuality that I've ever heard.

here's the amazing thing

[Anonymous Coward]

"here's the amazing thing: take a photo (for example, of yourself) with a cell-phone, e-mail it to the company, and they will send you back a pattern that will recreate that photo."

Sorry, but that is not amazing.

300!?

[Mr_Miagi]

This is madness!

Re:300!?

[MtlDty]

No, this is Sparta

Japan rediscovers the pixel....

[Anonymous Coward]

Sorry to sound rash, but what the hell is supposed to be so amazing about this? Neither the concept of mosiacs or tesselating shapes is new, and putting the two together is hardly a great leap in thinking. This is hardly different to what I expect a lot of kids did with their Lego.

DAmn

[aepervius]

I see you spoiled one of the best scenario I used as game-master to break in blazed wannabee Call Of Cthulhu RPG investigator.

A better idea

[drinkypoo]

Instead, get yourself an arrangement of colored tiles, create a custom palette based on them, then convert the image of your choice to your palette colors. Et voila! Set grid to 1px, Zoom in, and get to tilin'. Then you can glue it all down to a coffee table when you're done and have a work of art that some idiot will pay you money for.

However...

[Anonymous Coward]

it cannot make a picture of my wife - that is: all black.

Fun with Bits

[AP31R0N]

There is some combination of bits that will give us a picture of life on another planet, or Cleopatra giving Elvis a hummer. Maybe even me holding a lottery ticket with winning numbers. A combination of bits might give us a #1 pop song sung by no one.

Fun to think about.

Translation

[mattr]

FWIW it says take a photo with your cell phone, send it in and the response arrives.
They have variations in sepia, wine red, midnight blue, and a puzzle game -- but all are sold out.

By the way JIGAZO actually is how you read the three character name, which means "your own image" or maybe "self portrait image".

Here is what the separate white panel looks like:
here

And here is a blog, showing the process: "I made the Jigazo puzzle!"
I'm not going to translate it entirely, but he says the pieces are nice and thick, and well formed. Also there is a guide image on the back of each piece and also, a light version of it is shown on the front of the piece as well so it's easy to understand.
It comes out looking quite good, the key is to look at it as if looking far away, with squinty eyes. ;)
It took him 90 minutes to make his first one, then the second time he got faster.

Translation..

Use the 300 pieces in the box and you can make anybody's face.
First in the World, a jigsaw puzzle that can do your face!
Jigazo Puzzle

Can you believe it?!
All the sample images shown here were made with the same 300 pieces.
In order to make your own face...
When you send a photo by cell phone email a response image will be sent back immediately.
Just put the pieces together the way it shows and voila!

By repositioning the same 300 pieces you can make images as varied as these!
1. Send your image to the email address printed in the included instruction manual
2. A response image with the answer arrives in your cell phone
3. When you line up the 300 pieces according to that, then your face is complete. You can use anyone's face!

(small print)
Sending your photo, and receiving the response image will incur packet communications charges according to your cell phone service provider's contract.
If you are going to be using it repeatedly, we recommend you use your service provider's fixed price unlimited packet plan.
Note that the creation of the response image by our company is without cost no matter how many times you use it.

Set contents
300 piece puzzle. Color of pieces differs depending on the set type.
Piece set-up tray
Starch adhesive and applicator spatula (or something like that)
Manual. Includes instruction images for the Mona Lisa, Girl with Pearl Earrings, Natsume Soseki (the author), Kennedy (JFK), President Obama, Beethoven's face. You can immediately start playing with these faces without using a cell phone.

Other things you need:
A cell phone with camera functionality, from NTT DoCoMo, AU, Softbank Mobile (Disney Mobile and iPhone also included).
* When displaying your finished puzzle, please use separately sold 300 piece (white) panel specially for JIGAZO puzzles.

Price is 1890 yen, a bit over 20 bucks.

JIGAZO FAQ

Q1. When I send in a portrait photo with my cell phone, do you mail me back a finished puzzle?
A. No, that is incorrect. Your face is created only out of the 300 pieces in the box.

Q2. How do you do that?
A. There are 300 pieces in the box with slightly different tones.A program finds the tones closest to those of your face and sends back to your cell phone a response image specially for your face.

Q3. How quickly does the response image get sent back?
A. It depends on the state of the communications network, but in about 10 seconds it should be sent back and then you can immediately start playing.

Q4. How do you put the puzzle together?
A. Each of the pieces has a hint image on it, so you position the pieces as shown by the response animation. Enjoy watching your face gradually start to appear.

Q5. Can I only put the pieces together once?
A. No, you can redo it any number of times, and make anyone's face.You can pl