A couple of weeks ago I picked up a book called, “How are intelligent are you?: The Universal IQ Test” by Victor Serebriakoff cheap from a book store. Its pretty interesting but on the latest test (Spatial Analysis) there is a segment that I failed every question on but unlike the other questions I’ve got wrong I don’t really understand why.
I’ve scanned the page in below, copyright to the publishers of course, if this causes any problems I’ll take it down.
I’ll leave it up for a while so people can guess then I’ll give the answers and we’ll see if everyone is as confused as I am.
SA21 = A
SA22 = D, F
SA23 = B
SA24 = E, F
SA25 = C, D
And having sat down and looked at it I can understand where the answers are coming from now! It just took one of those ‘3d pictures’ moments before I could visualise it properly.
The top angle of A looks like it’s the angle you are looking for.
I found another version of the same question (as in, it has the same shapes and same questions), see page 16. I am more easily able to figure out the angles on a few shapes (although A looks the same to me), e.g. SA22 and SA23. Also the size of shapes look correct, whereas the one Disposable Hero gave it looked to me that one had to ignore the size of the shapes.
No it doesn’t. Take a right angle, say the corner of a piece of paper, and hold it up against the screen. One angle in SA21 matches, and none of the angles in A match.
I’ve cut and paste A on Paintshop Pro and you’re right that apart from the right angle the other angles are similar but not the same, the only other option F, is distorted by the scanning process and is an even poorer match when checked against the original book. I think they did intend it to be A but its definitely not an exact match
I wasn’t sure if the question meant find any set that work or find all sets that work. Obviously wer’re to assume we can rotate the shapes, but it is not clear if we’re allowed to flip them or not. Finally it says shape and not size and shape.
In any case
ETA Sorry I forgot how to make a spoiler tag and am running out of edit time.
Avert your eyes if you don’t wish to see.
[spoil]21 is A by itself rotated
22 looks like and A and F. (The top corner is clearly larger than 90 degrees so it’s not a C or E by itself.)
23 looks like a D stacked on a C back to back with another C stacked on D. Ditto F stacked on E back to back with another. Both of those are larger of course.
24 I’d say you can make with 2 As
25 is F by itself (and D by itself if you allow smaller). Also four Ds will make one about the same size by putting a triangle to the right and top of a square made with two Ds.[/spoil]
What the question fails to clarify is whether a shape can be rotated, reflected, or enlarged/reduced. So it requires a mind reader to know what the composer of the question had in mind. The examples show a rotation, but does not address the other two possibilities. If nothing else, that leads the solver to what may be a lot of time-wasting trial and error.
That’s a really poor quality puzzle. The shapes only barely resemble the claimed solutions.[spoiler]For instance, the leftmost angle in SA22 is well under 90 degrees. The only way for D+F to remotely fit together is for D’s right angle to be on the left, but the resulting shape is distorted.
SA24 isn’t much better. The end angles are about 45 degrees each, while the angle on E is more like 30. A significantly better answer is C+C, with the shapes flipped. I think it’s still a bit long but the angles are much closer.
And SA23 can’t be B, even with scaling. SA23 is obviously an isosceles triangle. B is not. The closest match is with the upper-right side rotated to be on the bottom, but it’s still very far off.
Like Peter Morris said, you can use a piece of paper to verify all of these. SA21 and SA25 are the only ones that remotely match the answers.[/spoiler]
Here’s something strange. I printed the page, to try cutting them out and fitting them together. For some reason the angles change on paper. A does have a right angle on paper, but not on the screen.
I think it is a poorly copied puzzle. The version I linked to in post 10 looks right to me (and looks like it solves the specific problems you mention). In any case it is a lot better.
Then your display is set up to have non-square pixels. Some poor quality monitors are this way naturally–in particular, ones at 1280x1024 resolution. Otherwise, you may be running at a non-native resolution which does not have the correct aspect ratio, such as running at 1024x768 on a screen that is natively 1920x1080.
Ahh–that is much better. But I don’t think it’s just a poor quality copy. Ordinary distortion can’t account for SA23. I think someone tried to recreate the puzzle rather than just copy it (maybe that’s what you meant by “copy”).