Any suggestions as to what to look into in the meantime?
There was a post in RSA #14 that I think everyone missed. >>8408
This was from before Chris started talking about q in detail and people actually did anything with it. I seem to personally remember thinking it wasn't actually Chris who posted it, since it had a questioning tone to it. Chris actually started mentioning the idea that we could multiply c by a bunch more primes in order to produce more information before I even did the main portion of Grid Patterns:
>It is easier to break down the less factors there are as there are less solutions.
>The answer is easier to come to the more factors there are as there are more solutions.
So I think this other post is almost definitely Chris, in retrospect. This is what it says:
>Thinking outside the box, what if actually multiplying c by certain value or values, forces the result to be where we want, to make the lookup easier.
>We could theorectically do it in two steps, getting information from the first product, qc, and introduce a second factor, v, again forcing the result to give us a deterministic result for e and -f. This product, vqc, could then be used to "triangulate" a lookup somehow?
>The chosen numbers for v and q may depend on the type of c we are using but I would speculate they would be from a limited set…
So when a couple of us were pondering earlier in this current thread whether we'd be multiplying qc by another variable called v, it would seem that he actually already told us that we would be and nobody paid any attention.
Now, he says we're meant to "get information from the first product, qc", in order to find v. This variable v that we're meant to multiply qc by is somehow related to qc. Here's where I think it gets interesting. Look at this >>8880 post, within this current thread:
>The column at -1 is significant.
>At n=1, the values of a[t] and d[t] that if you subtract one from the series of squares with sides (2vv)-1 : 1, 7, 31, 49,..
The values in (e,1) and (f,1) are based around 2tt and 2t(t-1). They're based around the t values. But for some reason Chris decided to call it v here, just out of the blue, having only ever mentioned a variable called v in that other post.
In short, I think we're meant to multiply qc by a t value in (-1,1) that somehow relates to qc. I'm not sure which, and I'm not sure what we'd then do with it, but it seems fairly obvious to me that this is something we're meant to do to find the solution, and given nobody else seemed to notice I thought I'd point it out here.
>In column 1 row 1 at x = c (for odd c) a-1 and d represent something, but is it of any use?
This is from RSA #14 I'm pretty sure. d'-(a-1)' = c+1. Their gcd also equals c+1. I don't think that was the point of this clue though. Where else are d and a-1 paired up? If they're meant to "represent" something then I would think this is less about a neat element where c+1 turns up and more about applying a concept to this cell. Any ideas anyone?
Words words words words words words PUNCHLINE