I've been going over this VQC post VA brought up on Discord as well as the fourth image in >>6695.

From the attached image:

>The final steps involve finding a base using (f-1)/8 for each of the eight triangles (some a bit bigger than others by a unit in length) which is larger than (n-1)/8 BUT smaller than (x+n)/8.

>The choices reduce ‘exponentially’ and with less complexity than find the root of C, so the last steps do not push the overall complexity Big Oh above Big Oh for finding the root.

From the image at the beginning of the thread:

>Since f can be divided by 5, we can make a base for each triangle which is made of (f/40), still with 4 left over.

Are we meant to be factoring f recursively? We're meant to "use" (f-1)/8 to find the base of the triangles in nn. f can be smaller or larger than n, so we can't just "use" f to find it, since it doesn't seem to scale with any of the other related variables. But if we multiply a factor of f by 8, we can find a base with a remainder less than 8 that works. And I remember when he was talking about that people were asking where this arbitrary 5 came from. Maybe we're meant to find the factors of f. f is made of 2d and e, which also have square roots in the latest code snippet.

I've spent way too much of today putting together a program that I intend to have do the following:

>use the nested i and j loop from the grid to generate a bunch of cs (only semiprimes in this case, because they only have two n values and I haven't accounted for varying numbers of possible ns)

>brute force every single number related to c that we've been told about

>using this big list of variables, see which ones are equal, and do some arbitrary math (like mods and additions and thing) to pairs of variables to see if there are any relationships we aren't aware of

>do this to every pair of i and j, generating a big list of relationships for each one and storing them in an array

>find the relationships that are common to every single pair of i and j

VQC said enumerate the patterns, so that's what I'm trying to do with this. When I tried to do it with an Excel spreadsheet, he said I was getting very close. That Excel spreadsheet didn't go anywhere because it was far too clunky and hard to use. I've been trying to figure out how to do this in the most efficient way possible since then. I would post my code, but the number of things I've done in really stupid or lazy ways is probably embarrassing, and it's 579 lines of code anyway. Pic related is what it currently outputs. These are all pretty obvious and well known to us now (or they're just stupid redundancies that I haven't bothered to code out yet), but I've only implemented basic stuff so far. I think 55 is the number of i and j semiprime pairs it went through. I'm about to implement the first three cells where a[t]=c in (e,n) and the an, bn, a(n-1) and b(n-1) cells. Then I'll start adding other variables.

We keep going through this cycle (I think I've mentioned this before) of working on getting a big list of the patterns together, seeing a new VQC post mentioning something we weren't aware of already, and then spending all of our time on the new VQC thing or something else he didn't mention that we think will be useful and forgetting about our main objective here. If anyone would like to help me to revive the whole "enumerate the patterns" thing, please, I would really love to not be the only one doing that. If not, I'll still do it myself. Another thing that would also be useful, if anyone reads this, would be math ideas. I'm thinking there might be obscure things we aren't thinking of, like d of the na transform cell being equal to the x from the cell where a=1 and b=n or some weird bullshit like that, but I'm also thinking maybe there's a less obvious mathematical relationship like the thing I mentioned about (BigN-n)/(b-1)=(a-1)/2 (which is less obvious just looking at it). So I'm thinking if I make a big list of every variable mod every other variable, or every variable mod (every variable +1), and so on, we might find something. So what else? Maybe (variable+1)/2 and stuff. It will be more useful to first work on getting every known pattern together but I'm still thinking it might bring up things we wouldn't have otherwise discovered.