• The only 'puzzle' parts of this are the links, titled 'Set X', 'Inscription X', or 'Constellation X'. Everything else is flavortext, apart from this readme section, although the flavortext does tell you that the meaning of the inscriptions is not supposed to be deduced until you have finished everything else.
  • DO NOT even LOOK at the links out of order. They have been placed in this order in order to make the solving experience as interesting as possible.
  • DO NOT move on from a set until you have solved all 5 puzzles in that set, and have ensured that your interpretation of the rules arrives at a unique solution for all 5.
  • Every puzzle has a unique solution. If you believe you have found a puzzle which does not have a unique solution, or has no solutions, please comment on this post and I will see if it needs to be fixed.
  • No puzzle arrives in a solved state. There is always a distinct SOLUTION that must be input into each puzzle.
  • The mechanics and meanings are one hundred percent consistent, even between sets.
  • These are plain grid-deduction puzzles, apart from the fact you must work out the mechanics yourself. Once you know the meanings, they are simply pen-paper puzzles in the vein of Japanese puzzles such as Sudoku, Hashi, Nurikabe, Slitherlink, etc.
  • Feel free to post partials, but SPOILER everything please!
  • Do not read spoilers unless you have really given up.

The old monk sips quietly from his tea, before placing the cup beside him on the floor.
'So, you wish to climb the Witless Mountain?'
'I do.'
'I am too weak, now, to come with you to the summit. But I can guide you some of the way.' He picks up his staff, and leads the way into the snowstorm, away from the shelter of his shack.
The summit looks so distant - the climb will not be easy.

The snow falls with a certain ferocity, sparkling white pricks appearing on the robe hugged tight against the monk's body.
'Do you see the rockface right up ahead?' His voice is barely audible over the wind.
'That's your first challenge.'
On the black cliffside, a giant set of shapes and symbols are carved:

Set 1

A few minutes pass.
'I think I understand.'
The monk smiles. 'Very well. Let us continue.'
A second rockface rises around the corner.

Set 2

The monk ushers you along as soon as he sees that the puzzles have been understood.
'It is getting cold. I think this is as far as I can go. I will leave you here.'
'Hang on, aren't you going to tell me what THIS is?'
On the third rockface, a huge row of symbols glows blue beneath the puzzles.
'Ah. You want to know what the blue inscriptions mean?' He laughs wryly. 'You'll be seeing a few of those. I wouldn't worry about them now, if I were you. They're something to think about on the return journey.'
With that, he begins making his way back down the path.
You're on your own, now.

Set 3
(Inscription 3)

Set 4
(Inscription 4)

Set 5

Set 6

Set 7

Set 8

Set 9
(Inscription 9)

Set 10
(Inscription 10)

Set 11

The peak is so close...

Set 12
(Inscription 12)

The final climb awaits.


It is dark, now, and as you look up into the sky, the stars begin to shine...

Constellation 1
Constellation 2
Constellation 3

The visions fade, and you return to your senses. You look back down the mountain, past the cliffs, to the gently glowing image of the monk's shack far below. You laugh, recalling his last piece of advice -
And begin the journey down.

EDIT: All 12 sets and constellations have been solved. No inscription solutions have been posted yet - the bounty is for whomever solves the most inscription puzzles!

EDIT 2: Inscription 5 has been taken down for maintenance, sorry to all those who have spent a while on it!

  • 4
    $\begingroup$ Very cool puzzle! +1 $\endgroup$ Jan 12, 2017 at 8:24
  • 2
    $\begingroup$ From the title and nature of the puzzle I'm guessing that this is the inspiration: en.wikipedia.org/wiki/The_Witness_(2016_video_game). $\endgroup$
    – ShadowCat
    Jan 12, 2017 at 17:53
  • 1
    $\begingroup$ @ShadowCat right on the money! No knowledge of that game is required though :) $\endgroup$ Jan 12, 2017 at 20:49
  • $\begingroup$ How can we be certain that our solution is correct? $\endgroup$ Jan 12, 2017 at 21:13
  • $\begingroup$ @greenturtle3141: Having unique solutions to all puzzles within a set. $\endgroup$
    – Deusovi
    Jan 12, 2017 at 21:43

4 Answers 4


Partial Answer - Sets 1-12 solved, Summit partially solved


  • Try the puzzle before reading this solution. Please, it's worth it.
  • If you can't figure out a mechanic, take a break and come back. Only look at this guide if you absolutely have to - for example, you've been trying for days and you haven't been sleeping well, or the only way to pacify the three bears currently mauling you is to tell them the solution to this puzzle.
  • This first section of my answer contains hints for all the sets' "teachings". The first hint is written normally; the second is rot13ed so you don't reveal both at the same time. Only go for the second hint if you've thought about the puzzle with the first for a while and still haven't come to a conclusion.

Set 1

Draw lines on the gray borders. // Fcyvg gur flzobyf rirayl.

Set 2

Last time, you divided the segments into two "equal" sections. // Znxr gur gjb frpgvbaf unir gur fnzr ahzore bs qbgf.

Set 3

You shouldn't need a hint for this one. // Vg'f gur fnzr guvat nf orsber. Rzcgl fdhnerf nera'g "jbegu" nalguvat, gubhtu.

Set 4

You've been making an assumption that you shouldn't have. (I may have reinforced that assumption - sorry!) // Ybbx ng gur frpbaq chmmyr. Vg unf svsgrra qbgf, juvpu lbh znl abgvpr vf abg qvivfvoyr ol gjb.

Set 5

The square mechanic does not interact with the dots. // Ybbx ng gur znexf. Gurl nccrne gb or cbvagvat gb jungrire'f cnfg gur rqtrf.

Set 6

Squares don't need to be separated from each other. // Nqq.

Set 7

It's like a dot, only... different. // Vg'f cnegvnyyl n qbg, naq cnegvnyyl abg n qbg.

Set 8

Again, the symbols are very representative of what's going on. // Ubj qb gur gjb fznyyre gevnatyrf eryngr gb gur ovttre bar?

Set 9

The big triangles can be treated independently. // Gurl whfg fnl gung lbh unir gb or noyr gb nffrzoyr gurz sebz gur unys-gevnatyrf - abguvat zber.

Set 10

Eyes see things. What can these eyes see? // Bar rlr ol vgfrys jbhyq or hfryrff.

Set 11

Come on, really? No, you don't get a small hint for this one. // Ybbx pnershyyl ng gur rlrf gung frrz gb or gur fnzr.

Set 12

These eyes are like the ones from before, only... different. // Vs V chg bar bs gurfr rlrf fvqrjnlf, vg jbhyq abg punatr gur fbyhgvba ng nyy.

I was a budding archaeologist, looking to learn more about the the Witless. The ancient writings had always intrigued me, but I'd never had the chance to make a pilgrimage to the mountain... until now.

As the monk showed me the first rock face, I took out my transparent sheet (so I wouldn't disturb the carvings themselves) and marker, and got to thinking.

That first one was interesting. It appeared to be a domino of some sort, but... that line in the middle was only barely noticeable. Almost as if it was begging to be traced, to be completed.

Yes! That had to be it! But what about the others? Did they work the same way?

enter image description here
They did! I could divide all of them into two "equivalent regions", so to speak.

Ignoring the monk's smile, I moved on to the second rock face. More inscriptions, this time with dots.

So, before I had divided them into "equivalent" regions - could I possibly do the same thing now? The symbols aren't exactly the same, but they could be equivalent in some other way... like the number of dots, maybe?
enter image description here

That wasn't too bad, right? Alright, let's move onto set 3. It seems to be more of the same.

enter image description here

Yep, more of the same. Is this going to get any harder?


The First Roadblock (Set 4)

enter image description here
There's definitely no way to split that into groups of 6.

But wait...

who said it had to be two groups of six?
enter image description here

There we go. Does it work for the rest, too?

enter image description here

Alright, it does! So maybe I was wrong about

the rule being to split into two equal groups - it seems to be "split into any number of equal groups". Cool, so far, so good.

The Second Roadblock (Set 5)

Okay, what.

What is this.

This is not a dot. I liked dots. Dots were sensible, and logical, and they taught me things. But the square is not helping right now.

So what could it be? Maybe...

a negative dot?
enter image description here

Nope, that fails at the fourth carving.

What about...

being "worth" the area of that region?
enter image description here

Nope, that fails at the fourth carving too.

Wait, maybe I was wrong in both of those ideas.

Maybe it's not worth anything at all! Maybe it tells us something else.

But if so, then what is it?

It looks like it should be something related to the borders... specifically, the things outside the borders...


It's the number of regions that that region borders!

And that works for all of them!

enter image description here

Let's continue to the next set - it works the exact same way.

enter image description here


Okay, this symbol makes sense. It's like a dot, only... not a dot. I guess that means that...

the dots are optional?
enter image description here

Yep, seems right to me. Moving on.


Ooh, new symbols. Okay, so we have two small triangles and one big one.

It seems like the two triangles fit together to make the big one. Maybe that logic carries over somehow to the regions themselves?

Let's see if that works:

enter image description here

It does! And it also tells us

that we can't rotate the half-triangle pieces, since the last two puzzles would have multiple solutions if we could.

Oh, this time there are multiple big triangles!

I guess it works the same - each big triangle gets its own copy of the two half-triangle region shapes. They can be combined in different ways, too, it looks like.
enter image description here

The Third Roadblock

Well, that's... interesting. Now we have eyes.

Maybe I should turn around before I accidentally activate some sort of ancient curse.

No, no, I came here to get to the top, and that's exactly what I'm going to do.

I just don't want to look at the eyes too much.

Anyway, eyes look at things. (Or so I've heard.) Maybe that's somehow related to what they do?

Maybe what they "see" is important?

But how?

It can't be what they see in their own region. The third puzzle rules that out, since the eye in the middle can't see everything in its own region. (We can solve that one without even knowing what the eye does, which is nice.)

Maybe it's that...

they all have to see the same thing?

Yep, that seems to work. And it gives more unique solutions!

enter image description here
This also teaches us that we can't have any "dangling" walls - all walls have to be between two different regions.

So, the next set... it seems to have the eyes again, but rotated and reflected.

That just means what they "see" needs to be rotated and reflected too! This isn't so bad.
enter image description here

Alright, one set to go! And it has the swirly eyes. I guess those just mean...

that we don't have the direction. (But hey, at least the reflection is still given!)
enter image description here

Wait a minute. I was a fool.

How could I think those original puzzles were so simple? Of course the symbols meant something. No, I have to go back there with my newfound knowledge and do it right this time.
enter image description here

The Summit (Partial)

It's cold up here.

I have no idea what I'm doing.

(Okay, I have a vague idea of what I'm doing, but it doesn't seem to be helping much.)

enter image description here

  • 1
    $\begingroup$ Sorry to be the one breaking the bad news but 11.3 is incorrect (bottom left eye sees different stuff) $\endgroup$
    – Wen1now
    Jan 12, 2017 at 23:16
  • $\begingroup$ Have you proven that your coloration of the dots in the summit (so far) is correct? I feel like there are more ways to divide them. $\endgroup$ Jan 12, 2017 at 23:16
  • $\begingroup$ Sorry (again) but 12.1 seems to be wrong $\endgroup$
    – Wen1now
    Jan 12, 2017 at 23:32
  • $\begingroup$ How did you deduce green and dark blue on the summit? $\endgroup$
    – Wen1now
    Jan 13, 2017 at 0:01
  • 2
    $\begingroup$ Thoroughly enjoying the narrative without peeking at the spoilers! Here's to another 21,055 characters before the limit. $\endgroup$
    – humn
    Jan 13, 2017 at 7:45

You smile as you recall the last few hours, in which you finally solved the Summit...

Also on a kind of side note, I haven't posted the solutions to the sets because I beta-tested them, so I knew the mechanics already. And somehow @Deusovi still managed to beat me to finish the sets. Anyways...

spoilers of mechanics are not spoilered, but actual steps are so be careful!

After a long walk you finally trudge onto the summit. And what a spectacular view it is! But as you glance up you realise it will be night in an hour or two... you look down and see the final challenge, a single puzzle etched into the mountaintop. By the time it gets dark there is no way you'll be able to solve it so you quickly get to work...

Part I: semi-dots

Well doing a simple dot count (which is overpowered) I find that there are between

56 and 62 dots

Now we can get rid of the majority of these because

by the top right 6-pack, there are at least 7 regions, and by the bottom right, there are at least 7 dots per region. Now we can quickly eliminate the primes and the primesx2 to get 56 or 60 dots total. However 60 cannot be expressed as a times b where a, 6 are at least 7. Therefore we use exactly 56 dots.

Quote from @TheGreatEscaper:

The semi dots get used in the first logical step of realising that none of them get used

Anyways that means we have

a 7, 4 and 1, five 5s, five 2s, and three 3s.
Can we make 7 regions of 8 or 8 regions of 7 dots? 8 dots a region means that 7 goes with 1, and each five must therefore go with a 3. But there aren't enough threes! Therefore we deduce that there are exactly 8 regions of 7 dots.

We are forced to pair them like so:

(7), 5 groups of (5,2), a group of (4,3) and a (3,3,1) group. Also the (3,3,1) group must contain the very top right two 3s, and therefore the top right 2 and 5 are joined.

Okay, that was the easy part. Here's a diagram if you got lost/are not bothered to read.

Notice that I've drawn in a few edges. Also, the numbers for the filled-in regions don't matter; we know that they must be the same colour. Lastly I'm sorry if you're colour-blind, tough luck. Just read through and hopefully you get it. enter image description here

You quickly scrawl down your working on a piece of paper. It seems like you're making good progress, until you try and do the next step... and get stuck

Part II.i: eyes (spooky)

This part is much harder.

Can eyes have three adjacent edges?

Seems unlikely. In fact

Looking at the top right-most region of eyes, one can deduce a contradiction (in essence, the top right most must face left, the one down must face down and therefore the top right must look like this:
enter image description here
which fails because the yellow going into the eye means it can't get back out, if you see what I mean.

Now question two:

Can an eye have exactly one adjacent edge?

This gets killed pretty quickly:

The top right eye cannot be part of the three dots and five dots.

Okay so we know that

eyes must have exactly two adjacent edges.

Now consider the

middle right eye

If it is not part of the green region then it has

three adjacent edges so it must be part of the green

But wait! We can deduce that the last adjacent edge must be below it. Also since it connect to green it must be distances of

2, 0 and 0 (and ???).

Now fill in the rest of the top right:

We know that an edge opposite an edge that's coloured in is NOT coloured in. Therefore the top right eye must connect to the yellow and to the six.
enter image description here
Can we have distances of (0,0,2,2)? No because then the other eye fails. So therefore they must be distance of (0,0,2,3)!!

Okay then finish that corner off:

enter image description here

Let's move on before those eyes get to us...

You glance around the frozen mountaintop. The sun's heat has almost all been taken away as night approaches

Part II.ii: eyes (left side)

Ok let's look at the left side:

Specifically, the purple eye:
enter image description here

Now there are two orientations possible, and

one of them stuffs up the eye next to it

So we must have this:

enter image description here

Okay now the

eye pair underneath that.

Now we have two non-trivial cases

one of the trivial one gets eliminated because the red 3 gets blocked off

First case (the failed one):

enter image description here

It's not obvious (at least not it wasn't immediately obvious to me) how this fails until you consider

the squares

The pink square is NOT part of the other square. But the other square is then adjacent to pink and red, which are NOT the same region! But it's a single square which means that it is next to one region, and there's no way it can connect to the six square group. Contradiction.

So the second case (which must be correct by assumption that this puzzle has a solution):

I have skipped ahead and made some more semi-trivial progress
enter image description here

Then we realise that wait a second!

The two squares must be connected because they're not connected to the six square thing.

And the red's not connected to any of them! This is going well.

Also red is not connected to the eye because if it is then obviously something bad happens.
enter image description here

As you look up you realise that the sun is setting fast now.

Moving on to the last bit of the eyes, look at the top left unsolved eye. Since it's kind of obvious that the 5 dots and 2 dots next to it connect, and must connect through that eye, we can solve for it:

enter image description here

where pink is the same region but unknown.

Part III: triangles

I'll skim through the mundane counting and get to the real stuff fast.

Firstly: the pink cannot be part of any triangle stuff. The pink region knocks out at least 13 squares.
Secondly: the bottom left eliminates a further at least 9 squares that cannot be part of a triangle (the eye has two possible orientations, and in either case at least 9 squares are knocked out)

Therefore we have

90-13-9=68 squares left, over five triangles which means at most 13 squares each.

But also

The yellow region is at least 15 squares

So we can subtract that as well!

68-15=53 squares left in triangles.

Therefore triangles have at most

10 squares

BUT think about where the light green fits in the red.

It must fit in the bottom right because the red is quite close to the limit of 10 and it can't fit anywhere else.

So we know what shape the other piece is! Adding this on our map, after some trial and error, we have:

enter image description here

Now we're almost finished! Now for a weird step:

Consider the rightmost triangle.

Which piece of the two small triangles covers it?

It can't be the tetromino we just made there because that would overlap with the other triangle

We can make another step!

It must be the light green polyomino there! But after a quick consideration, we realise that the light green cannot extend further right.

Then it's just basic logic to deduce the rest (about as hard as an easy/medium puzzle from the sets)

You finally finish the Summit puzzle! Just in time too, because the sun has finished setting. It's getting dark now... Smiling, you begin to relax and as you look up into the sky, the stars begin to shine...

And here's the final finished result:

Click for final image

  • $\begingroup$ Outstanding effort! It's not an easy puzzle. $\endgroup$ Jan 13, 2017 at 1:57

Partial Answer - Constellations

In the spirit of @Deusovi's answer, I will precede the actual solutions with hints. Again, two hints, with the second rot13ed.

Constellation 1

What have you been doing in all the previous sets? // Lbh bayl arrq gb havdhryl svyy va gur yvarf.

Constellation 2

Is there something missing, or is there just nothing there? // Sbe bar bs gur tevqf, lbh qba'g arrq gb svyy nal bs gur pryyf.

Constellation 3

Something else is gone. // Gur flzobyf ner va beqre bs gur tevqf va juvpu gurl ner cynprq.

It is dark, now, and as you look up into the sky, the stars begin to shine...

Hmm, those constellations look strange yet oddly familiar, and from here, even the stars of the Milky Way look like they're gathered in recognisable groups.

That first grid - you recognise it from the very first you solved on the base of the mountain. Except it's missing something. You look below it - there it is! A neatly formed dot, enclosed by a clear night sky.

You look more closely at the band of stars below the grids.

Looks like each group of symbols has exactly the number of symbols missing from the corresponding grid!

Eagerly you begin to solve the other grids.

That second one trips you up though. Both ways of putting the symbols in lead to a unique solve.

You look again at your transparent sheet (which has proven even more useful than you thought, now that you have the stars giving you puzzles).

Of course! Only the lines matter - that's what you've been doing all along, right?

As you complete the set, you look back at the stars, and realised they've shifted from before. Something looks off though.

There are only four groups of symbols. Maybe one grid doesn't need any symbols put in it?

Once again, the stars have shifted. Now something looks really off.

There's one group of three symbols, but all five grids only have one empty cell?

As you think upon it, you suddenly realise,

Maybe you have to fill in the lines in the band as well?

You try that, but there are too many possibilities...

Maybe the symbols have to go into the grids in order?

Satisfied, you look up again. The stars are back to normal now. Time for the journey down.

  • $\begingroup$ What? But... that's just - I thought I already tried that! Good job btw. $\endgroup$
    – Wen1now
    Jan 16, 2017 at 5:27

Inscriptions Partials

(No hints because so far I haven't found a new mechanic)


Red - Segmentation
Green - Which new piece goes where?
Pink - Hypothetical arrangement
Blue - Segmentation of inscription

First one:


Inscription 12

Second one:


Inscription 10

Third one:


Inscription 9

Fourth one:


Awaiting TGE

Fifth one:


Inscription 4

Sixth one:


Inscription 3

  • $\begingroup$ Nice, you scored the two easiest inscriptions. They do, of course, generally get more difficult the lower the set, in line with the 'return journey' narrative. Set 4's inscription is the hardest, IMO. $\endgroup$ Jan 24, 2017 at 8:34
  • $\begingroup$ @TheGreatEscaper - Yeah, made the mistake of starting from set 4 and getting bogged up in seeming contradictions about prime numbers... But it did have a darker blue bar so I thought maybe... Working on 9 now. $\endgroup$
    – boboquack
    Jan 24, 2017 at 8:50
  • $\begingroup$ @TheGreatEscaper What does IMO mean? (I've done set 9) $\endgroup$
    – boboquack
    Jan 24, 2017 at 9:14
  • $\begingroup$ In my opinion :) Not the one you're thinking of. $\endgroup$ Jan 24, 2017 at 9:39
  • $\begingroup$ Wow, nice! Not sure if the fix for number 5 is an easy one, so it looks like you're done :) enjoy the rep $\endgroup$ Jan 24, 2017 at 23:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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