This puzzle belongs to the puzzle series: hyper-modern art

Some other place in the gallery...

"You know, I really like this piece of art."

"Yes, it's colourful and blinking and stuff. Finally, something which looks really modern to me!"

"No, I didn't mean that. Yes, the paining is nicely augmented with those flashing LEDs and the rotating rings at the bottom, but that was not what I was referring to."

"So what is it then?"

"I like, how the painter has once again managed to cleverly hide the title of the image in the image itself. That's so hyper-modern-art-ish!"

"Hmm, without my HUD I can not see the title anywhere in the image. It is somehow fitting, though. Would you care to explain it to me (once again)?"

enter image description here ( static 1st frame )

The task of the puzzle is once again to find a secret keyword (the title of the painting) from the image. Note, that the static image (link below animation) is in theory enough to solve the puzzle. The keyword has more than 4 and less than 12 letters.

  • $\begingroup$ I've split the gif into individual frames here for people to scrutinise. There appears to be a 0.3s delay between each. 1drv.ms/f/s!ArQEzxH5nQLKhtlnKpZwCbR1PJlypw $\endgroup$
    – niemiro
    Jun 17, 2016 at 21:01
  • $\begingroup$ Some thoughts so far: All of the clocks have 24 potential dot positions, with some of them removed. There are also 24 dots around the bottom. There are eight lights, eight clocks, and eight grey rings. Sometimes, lights show up more than once in the lower section. In the sixth and seventh frames, the same pattern of lights are on (all dotted but green), but the lower section is different. $\endgroup$
    – Deusovi
    Jun 17, 2016 at 21:49
  • 1
    $\begingroup$ ususual clocks - both hands seem to be the same $\endgroup$
    – Jasen
    Jun 17, 2016 at 23:47
  • 3
    $\begingroup$ This definitely has something to do with Flag Semaphore (due to the bottom right thing) $\endgroup$
    – Deusovi
    Jun 21, 2016 at 18:08
  • 1
    $\begingroup$ @Deusovi I tried that before but couldn't get anywhere, as there were unmatched patterns. With the new hint you're probably right. :O $\endgroup$
    – cyberbit
    Jun 21, 2016 at 20:31

4 Answers 4


The piece is titled


Credit to @Deusovi for the final decoding! My original method was overthought and didn't produce readable text, so look to his post for how that portion is done. I'll leave my observations here, as they may serve the wary puzzler looking for more clues.

  • The colored dots around the main "clock" never change. The center ring is the only one that doesn't move with the others. It is colored orange, and the sectors match up with the position of the visible orange dot around the clock. Thus, you can assign the dots on the outer ring like this:

    sectors with provided dots plus orange

  • There is a specific pattern to the "blinking" lights mixed with the clocks. This pattern can be used to assign colors to the rings from the inside to the outside like this:

    sectors with orange dots and rings colored

  • The patterns formed by dots of the same color match a pattern formed by the missing dots of one of the smaller clocks when rotated. This information can be used to assign colors to all the dots on the main clock.

    rotated clocks with colored dots
    Note that the bright green clock matches with three different orientations. Since the rotation doesn't actually matter in this puzzle, knowing the color association is sufficient.

  • The patterns of white bars in the inner grey rings of the main clock also match these missing dot patterns, and can be arranged to line up with the colored dots. This results in every sector (think pie slice) having exactly one highlighted zone.

    sectors with dots and rings colored solved

  • OP said the solution can be found using only the first frame, which indicates that the specific movement of the inner rings between frames is probably not necessary, but simply hints that they should be moved somehow.

  • The weird symbol thing looks like a clock, a colored dot, and a pair of semaphore flags (image provided for reference):

    key image overlay semaphore flags

  • Using this information, lines can be drawn connecting the colored dots to the matching clocks, and when the orientation is corrected with regards to the key, letters from the semaphore alphabet can be read from the hands of the clocks, and read (from left to right) as


Check out @Deusovi's answer for more details about the semaphore translation!

  • $\begingroup$ Valid observations all around, and quite useful. Just in the end remember that the still frame is all that is fully needed to decipher. $\endgroup$
    – BmyGuest
    Jun 18, 2016 at 6:03
  • $\begingroup$ @BnyGuest Does that mean that the other images are completely unnecessary? $\endgroup$
    – cyberbit
    Jun 18, 2016 at 11:54
  • $\begingroup$ Not saying that. Some info above clearly was discovered because of the animation, but no crucial information is hidden 'between the frames' and if you print out frame 1 you have enough info to decipher the title. $\endgroup$
    – BmyGuest
    Jun 18, 2016 at 13:10
  • $\begingroup$ @BmyGuest What about any other frame? Is the first frame the only one with sufficient information? $\endgroup$
    – cyberbit
    Jun 18, 2016 at 13:19
  • 1
    $\begingroup$ My First quick answer would have been "any frame" is enough, but I think that is not fully correct., so no, first frame is verified to be enough, other individual frames might be, but not all of them. $\endgroup$
    – BmyGuest
    Jun 18, 2016 at 13:47


The puzzle's title is


(I'm going to refer to the lower middle assembly of rings as the "orrery", since I need a convenient name for it, and it vaguely reminds me of one.)

Step 1: The Rings

The rings of the orrery all can rotate, as the GIF shows. We have eight rings, eight colored lights, and eight clocks, so the natural first thing to do is to try to associate them. The missing dots from the rings around each clock correspond to the "holes" in the orrery's rings, so that gives each ring a corresponding clock. Then, we can use the lights to align the orrery rings in a way so that every light is used once. With the colored lights given (and one ring already given to us), there is only one possible way to do so. Lord of dark's answer explains the logic in full detail.

Those associations give us this image, helpfully provided by Lord of dark. (Ignore the hands drawn on the lights: they're unnecessary.)

enter image description here

Step 2: Decoding

Now we've used the orrery to match up each clock with a colored light. The bird tells us what to do: those are the standard flags in Flag Semaphore! If we take the direction of each clock's corresponding light as "up", we get letters in Flag Semaphore from the clocks' hands.

Here is the image, with lines from the "head" to the "body" drawn in. (Arrows are used so that lines don't go straight through the clocks.)

enter image description here

Now we can read off the correspondences left to right in order of the heads to get a letter from each one.

Dark blue is north and northwest, which gives T. Red is northwest and southwest, which gives I. Lime is southwest and east, which gives M. Orange gives E. Magenta gives L. Yellow gives E. Cyan gives S. Green gives S.

So the puzzle's title is...


  • 1
    $\begingroup$ I was just about to update my post. Nice work! I started off reading the flags upside down, ended up with KKYQXGLY. Thought there was another cipher or something, silly me had my head turned around. XD $\endgroup$
    – cyberbit
    Jun 22, 2016 at 0:58
  • $\begingroup$ @cyberbit: You also used the wrong order: it's left to right by head location, not by rings. (I did that at first too.) $\endgroup$
    – Deusovi
    Jun 22, 2016 at 0:59
  • 1
    $\begingroup$ Yea, I noticed that too. I just realized the pattern of lights turning on and off represents the order of the rings, as well. Seems there were a lot of shortcuts you could take with this puzzle! $\endgroup$
    – cyberbit
    Jun 22, 2016 at 1:04
  • $\begingroup$ The answer is correct, the process only 90% as I imagined. I will maybe summarise in a community wiki post. As for accepting, I m going to give @cyberbit the check, as he was first to post something and needs the rep more, OK? $\endgroup$
    – BmyGuest
    Jun 22, 2016 at 5:46

Partial answer based on observations by @cyberbit:

The white patterns int he rings of the big clock all match a pattern of the missing points in the small clock.
The middle ring is orange so we can guess that every ring matches a color and we will find the same patterns in the color around the clock. Let's number the rings from 1 (in the middle) to 8 (the outside one). Here is the reasoning to match the rings with the colors :

  1. The orange ring cannot be moved without overlapping an other color so we know where to place the 2 other orange points.
  2. The only color which can match ring 4 is light green
  3. dark blue can only match ring 6
  4. The only color that matches ring 5 is red
  5. Dark green can only be ring 8
  6. Ring 3 can only be light blue
  7. Pink is 7
  8. Yellow is 2

enter image description here

This is the final big clock :

enter image description here

We can now associate each small clock to its pattern color. I have also reported each clock arrows to the corresponding color point.

enter image description here

From here I don't know what to do, I've tried to extend the arrows but it doesn't look like anything...

  • $\begingroup$ @BmyGuest I've mixed together the white from the 12 pictures. But I may have forgotten one picture in the process, I'm not sure. $\endgroup$
    – Fabich
    Jun 18, 2016 at 17:17
  • $\begingroup$ @BmyGuest I've edited the beginning (I hope I've made no mistake ^^ ). $\endgroup$
    – Fabich
    Jun 20, 2016 at 23:06
  • $\begingroup$ I'm wondering if the if the figure to the right of the the "color wheel" (with 8 grey rings) is supposed to be overlayed over the clocks (the one with an arc going through it). $\endgroup$
    – HAL 9000
    Jun 21, 2016 at 1:16
  • $\begingroup$ Interesting. You found an alternative determination route than I had in mind, but I don't see any fault with it. My intention was rather like @cyberbit s last image, I.e. That there is only one rotation where none of the whites overlap which assigns a color uniquely as well. But as said, I think your answer is as valid (and identical) $\endgroup$
    – BmyGuest
    Jun 21, 2016 at 5:28

Answer as intended by OP

The puzzle has been well solved by cyberbit and Deusovi (with some insights from Lord of dark), but as the path to the solution somewhat deviated from what I had originally in mind, I want to give a 'complete' answer to the puzzle as I thought it would be solved.

The wheel

The animated, turning wheel shows eight rings of which seven seem to move freely. The innermost, ring remains static and is the only ring which has a colour (orange). The wheel is bordered by dots of which only few show colours.

On closer inspection one can notice:

- The whole circle is split into 24 radial segments of 15degree span each
- There are 21 white patches plus 3 orange ones

Which leads to the hypothesis that one possibly could...

..rotate the 7 rings such that one patch lies in each of the 24 segments.

Matching clocks to the rings

One also notice that the clock faces all show a whole circle of 24 black dots spaced 15degree apart, from which a few dots are removed on each. One can easily notice...

that each ring has segments in a radial distribution which matches the radial distribution of the omitted 'dots' on the clock faces:

enter image description here etc.

Thus, each clock can be uniquely assigned to a single ring.

Matching colours to the rings

The orange color matches to an orange dot on the outside to which one of the ring-sections is pointing. The other two sections point at black, 'unknown' dots. It seems that there should be a connection between rings an colours.

Seeing that only the 7 white rings move, we can arrange them according to our earlier hypothesis:

enter image description here

We now note,that each ring has one patch pointing to a color and the others to black. Dark green is an exception here, with both of its sectors pointing to dark green, verifying that this rotation is a 'match':

enter image description here

With this, we have uniquely assigned a colour to each ring. We also notice that

the sequence of color-flashing of the leds matches the sequence from inside to the rim. (orange->yellow->cyan->green->red->blue->pink->dark green)

Combining the information

We now have

- Clocks assigned to rings
- Colours assigned to rings

so that we can directly get

Colours assigned to clocks

enter image description here

Arranging the clocks

The strange drawing in the bottom right indicates that

enter image description here
- clocks need to be attached to their handle (center connection)
- clocks can rotate along their handle (semi-circle)
- clocks 'hang' downwards from their handle (orientation)

If this is done for all of the clocks, one gets from:

enter image description here


enter image description here

As verification one can notice that

All clock-hands now point towards 45 degree - spaced angles exactly.

Decoding the word

With the clocks now all being arranged, the last hint stems again from the little drawing:

enter image description here visually hinting at enter image description here

which is a symbol of the semaphore flag alphabet

Using the hands of the arranged clock faces, this directly translates:

enter image description here

  • 1
    $\begingroup$ Pretty pictures. :) $\endgroup$
    – cyberbit
    Jun 23, 2016 at 12:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.