🧵 puzzle 3. Difficulty 3/7
John Puzzle at Thu, 20 Feb 2025 04:58:44 UTC No. 16592161
Instructions:
In each puzzle, 2 panels are missing.
You must select 2 of these panels—in the correct order—to complete the pattern. ( 46 is different from 64)
It is possible for the same panel to appear twice (e.g., “11,” “22,” etc.).
None of the puzzle require external resources. You do not need any special knowledge to solve these items. None of the patterns are cultural.
Lastly, you need to provide both the answer choice and the explanation. Feel free to use drawing tools to aid with your explanation. I encourage it so that it is clear to everyone what you are saying.
Previous post:
>>16591260
John Puzzle at Thu, 20 Feb 2025 08:56:01 UTC No. 16592320
There was more activity yesterday on the other 2.
Maybe this one is a bigger leap in difficulty?
Anonymous at Thu, 20 Feb 2025 09:09:23 UTC No. 16592335
>>16592161
Not science or math. >>>/tg/
Anonymous at Thu, 20 Feb 2025 09:36:25 UTC No. 16592364
John Puzzle at Thu, 20 Feb 2025 09:37:48 UTC No. 16592365
>>16592335
That doesn't look like an appropriate board either. Maybe I will post in random.
I just picked sci cuz I thought people here might be interested in puzzles like these and be clever enough to crack them
John Puzzle at Thu, 20 Feb 2025 09:41:43 UTC No. 16592374
>>16592348
2 of the later puzzles I will be posting took me 1 hr each and I have experience with this shit.
Don't feel bad about missing tricky matrix puzzles. It is fine to take more time to crack them as well.
Anonymous at Thu, 20 Feb 2025 09:42:47 UTC No. 16592376
(1,2) or (2,1)
2d shadows of a 3d object rotating in space
if it's wrong uhm actually I hate you and your test is shit fuck you, but if it's right it's actually a really good puzzle for smart people.
John Puzzle at Thu, 20 Feb 2025 09:55:01 UTC No. 16592399
>>16592376
I guess you're going to hate me.
Also whatever your logic is it must distinguish between xy and yx, 12 can't be the same as 21.
It isn't the shadow of a 3d object, I would say the pattern here is a bit better than something like that.
To not waste people's time. The pattern is entirely 2d as far as I can tell. If you manage to link it to some projection of a 3d object, gj to you, but it isn't necessary.
Anonymous at Thu, 20 Feb 2025 10:07:03 UTC No. 16592415
>>16592161
3, 4
John Puzzle at Thu, 20 Feb 2025 10:09:41 UTC No. 16592421
>>16592415
logic?
Anonymous at Thu, 20 Feb 2025 10:27:07 UTC No. 16592438
>>16592421
i work on emotions
John Puzzle at Thu, 20 Feb 2025 11:19:38 UTC No. 16592506
>>16592438
Your vibes are off this time
Anonymous at Thu, 20 Feb 2025 13:46:58 UTC No. 16592687
>>16592320
4>2
1st row the lying L is rotated into cube. Then the lying L is moved right (or the one on top is moved left).
2nd row the left shape is moved 1up 1right and then they're mirrored/rotated. Then bottom shape is rotated and moved to fit into top shape (1 right, since I rotated it by the middle of the small 2x2 square).
So with 3rd row I took the left shape, rotated it, moved it similarly to 2nd row, then rotated the whole to acquire shape 4. And then, following earlier example, I moved it up and right (or left) and got shape 2 , then I rotated it.
I don't care if it's right or wrong, or if it makes sense. I'm almost out of break time and I'm free at last.
John Puzzle at Thu, 20 Feb 2025 13:54:28 UTC No. 16592703
>>16592687
Not right and doesn't make much sense, but thanks for trying
Anonymous at Thu, 20 Feb 2025 14:23:19 UTC No. 16592741
>>16592161
Is this meaningful?
John Puzzle at Thu, 20 Feb 2025 14:31:19 UTC No. 16592753
Anonymous at Thu, 20 Feb 2025 14:35:45 UTC No. 16592763
How do AIs perform on these types of questions?
John Puzzle at Thu, 20 Feb 2025 14:40:53 UTC No. 16592768
>>16592763
I think it is currently beyond the ability of any AI model even if they were trained on these types of questions.
The scope is way bigger than on raven, it isn't a format that only uses the logic gate operations mixed in with shifting and grouping.
I think classical raven questions can already by solved by specialized AI and quite possibly models like o3.
So far, o3 has not been able to get a single question from my collection
John Puzzle at Thu, 20 Feb 2025 14:43:34 UTC No. 16592771
>>16592768
Edit, I didn t try these on O3 I was just tripping when I wrote that part. But I doubt it would be able to get them.
Only had success with easier verbal items and numerical sequences
Anonymous at Thu, 20 Feb 2025 14:48:38 UTC No. 16592775
>>16592768
>>16592771
That's interesting to know. Thanks
Anonymous at Thu, 20 Feb 2025 14:52:11 UTC No. 16592778
>>16592161
The answer is 3 and 1
John Puzzle at Thu, 20 Feb 2025 14:56:00 UTC No. 16592781
>>16592778
Why?
Is it physically debilitating for you to write down a logic?
Anonymous at Thu, 20 Feb 2025 15:01:56 UTC No. 16592786
>>16592781
The bottom square in the first column holds strong resemblance to the first square.
Answer three and four hold resemblance to the first square in the 2nd column, but answer three feels morphed in a more similar way to the bottom square in the first column.
Answer one feels right.
I stared at this problem and messed around with it for an hour, flipping things on their axes, etc, and I don't do puzzles ever so I have lost interest because there are too many red herrings.
John Puzzle at Thu, 20 Feb 2025 15:16:50 UTC No. 16592803
>>16592786
I can give a small hint
if you were to divide the black shape into 1x1 unit squares you would always have an even number of such squares. Why?
This is a symptom of the pattern.
John Puzzle at Thu, 20 Feb 2025 15:57:57 UTC No. 16592862
Quite surprised this is still unsolved because I was going to send significantly harder problems in the future posts.
Anonymous at Thu, 20 Feb 2025 16:15:27 UTC No. 16592885
>>16592161
This is definetely not it because it's such an stupid answer. Also a flawed one
But maybe the shapes are evolving as more squares are added? Left column has shapes that all have some sort of entrance
About the shape on the center, yes, it kinda fucks up with all my logic, but let's ignore that for now
Right column has a shapes that resembles stairs
So I would choose 3 and 2. Not 4 and 1 because 4 has two entrances, and not 1 because it seems as if the shapes flip each time
Anonymous at Thu, 20 Feb 2025 16:27:44 UTC No. 16592906
>>16592803
If you divide them into 1x1 unit squares you can notice that all of these are made out of two similar (or mirrored) shapes.
1 is 2 Ls = 4x2 blocks
2nd is 2 5-block "stairs" = 5x2 blocks
3rd is like 2 qs without top lines = 6x2 blocks
But I don't really see how top left turns into top mid, that would be consistent with any other changes between fields.
I'm sure there's a logical explanation that I'm not seeing, but it feels a bit like somebody made up some bullshit just to laugh at others for being unable to solve it.
John Puzzle at Thu, 20 Feb 2025 16:29:17 UTC No. 16592910
>>16592885
No, it's quite far off from the correct idea.
Let me restate the hint I gave and put more emphasis:
-divide the shape into 1x1 unit squares
you will notice 2 things:
1) each shape is made of an even amount of 1x1 squares
2) the shapes in the same row have the same amount of unit squares.
What could cause such symptoms?
Another hint is that you should look row wise for a logic, looking at the columns won't help here ( it didn't help in the previous matrix puzzle I posted either ).
Explore ideas that can explain 1) and 2)
John Puzzle at Thu, 20 Feb 2025 16:32:48 UTC No. 16592915
>>16592906
Oh correct, you got that part right. Let me attach a drawing to people can see what you mean
John Puzzle at Thu, 20 Feb 2025 16:34:41 UTC No. 16592919
>>16592915
Shouldn't take long for people to solve it from this point
Anonymous at Thu, 20 Feb 2025 16:35:57 UTC No. 16592921
>>16592915
That's what I initially did. Then I tried rotating them, mirroring, moving etc. but while I can see how to make 1->2 or 2->3, I don't see how it relates to how 4 transforms into 5 or 5 into 6. I'll think some more.
John Puzzle at Thu, 20 Feb 2025 16:38:34 UTC No. 16592926
>>16592921
Maybe you don't need to do any of that.
You are stuck with the idea that you need to transform something. You just need to identify one property that is changing.
Anonymous at Thu, 20 Feb 2025 16:45:02 UTC No. 16592939
>>16592926
In that case, my answer would be 2->1 because of how many lines they connect with.
1 2 3
3 4 5
3 4 5
if I'm counting right
I was also thinking about how many blocks touch, but then I wouldn't be able to finish the bottom row.
Doesn't sit completely right with me because the pattern is different in 1st row while the same in 2nd and 3rd, but I'll go with it for now.
John Puzzle at Thu, 20 Feb 2025 16:48:07 UTC No. 16592943
>>16592939
This is the right logic. Good job.
I have no idea with it is 3 4 5 twice, but when I scored my submission it said i got it right.
John Puzzle at Thu, 20 Feb 2025 16:49:38 UTC No. 16592947
>>16592939
What do you think of the puzzle overall?
Anonymous at Thu, 20 Feb 2025 16:51:39 UTC No. 16592955
>>16592947
Now that I know the answer, I don't think it's bullshit, obviously. I think it's a nice puzzle, I often fall into the trap of trying to notice a pattern of transformation which in cases like this leads to nowhere. I liked it, I think it's widened my scope if that makes sense.
John Puzzle at Thu, 20 Feb 2025 16:54:17 UTC No. 16592962
>>16592955
That's nice to hear. I have more puzzles that quite conceptually distinct. I think I will not increase the difficulty for now until people get more used to this level.
Anonymous at Thu, 20 Feb 2025 18:16:30 UTC No. 16593268
>>16592943
das pretty insane ngl