unforum

[thebruce]

well there's 7 consistent numbers (not 8 - the first number, before the counter spins, is random on each page load). At 4 digits each, that's 28 numbers. There are 36+/- circles. can't see how the two would recognize by placing numbers in circles... perhaps some other way, don't know.

The numbers again (they were posted earlier)

[resistor]

I'm starting to think this puzzle is WAY too complicated for me.

[Kayberry]

Okay, my current theory, which will take me some time to disprove, involves what a couple of you said about the size of the circles indicating depth. If we take all of the circles and lift them off the page in proportion to their size (Larger circles go higher) we will get a 3-dimensional rendering of the whole neural network. Then, after reconnecting all of the lines, we look at the whole mess from different angles and hope that the connecting lines either form a message or something obvious, or fail to do so and go back to forming tetrahedrons.

I'll be back soon.

Edit - Part 1 is done, all of the circles and straight lines that connect to something. The hard part comes next with those curved lines. It looks pretty from a side view, but I don't see anything significant yet.

Edit - Part 2 completed, and in my professional opinion it looks like a mess, not a message.

[thebruce]

go for it Kayberry. If I had a 3D modeller still installed I'd have done it a while back. Part of a past email to Dyson Trust mentioned that idea but that it's a fairly tedious process. The response didn't refer to that idea at all... so it could go either way. Let us know what you come up with!

[ckuchy]

I am thinking that some of the circles may be duplicated from one triangle onto another, if that is the case, then we may get down to 28 circles after all . . .

[MrToasty]

Just one more observation. The final piece we were given was a negative color scheme compared to the others - white lines on a dark background, vs. dark lines on a light background. Not sure if there is any significance. It does seem like the kind of thing you might say shine a light up through if printed on paper.

[thebruce]

ok, got another reply from DT to our status update. They said "Why not just construct it as a basic pyramid shape? I'd be curious to see what that looks like."


So I sent them my little photo of my transparent pyramid. *shrug*

[HopeOfTheFuture]

[thebruce]

interesting... what are those clips from? and what is the pyramid that's depicted, it's function?

Also, welcome to UF HopeIsTheFuture!


eta: also, here's an updated compilation... I took the pieces and adjusted them for how the pyramid would look top down with each triangle as the base (as if the pyramid were folded up around each piece on a table)
Nothing seems to stand out (red lines are the extended lines through the cap piece as shown earlier)

[Kelis]

Pyramid data compression?

[HopeOfTheFuture]

[Rogi Ocnorb]

Is there any way we can level-set our terms with them?

I know they said "3D" in an earlier comment, but I'd just really like to know for sure when we say pyramid that we truly mean pyramid (4-sided base) and not tetrahedron (3-sided base) or simply triangle.

[Made@Dark]

[enaxor]

[Update] Enitech Blog Feb 8

[MrToasty]

PMs - "Guys, it's a tetrahedron already! Just do it!"