About " computation "

[jaydfox]

For what it's worth, I doubt I've come up with much original. Maybe novel restatements of previously discovered things. I think my most original contribution was my acceleration of Andrew Robbin's matrix method, and I honestly don't know how original that was.

Oh that was before I became a member I think.

Is that acceleration method in the thread of Andrew's tetration ?

I'd like to learn more.

Maybe I have seen it anyway , but the name does not ring a bell.

Matrix methods are far from understood imho.

I would love to learn more about this acceleration , but maybe its all written down already ... and proved ?

regards

tommy1729

Here's where I had my "Aha!" moment (see post #6):
http://math.eretrandre.org/tetrationforu...401#pid401

For more in-depth discussion (or ramblings, depending on whether anyone but myself can follow what I was saying), see this thread, especially posts 3, 4, and 5:
http://math.eretrandre.org/tetrationforu...php?tid=59

There's a broken link in post #5. I'm afraid to edit such an old post, so I'll re-link it here. I tried inserting it as an embedded image, but it didn't work, so you'll have to click the link to see it:
http://math.eretrandre.org/tetrationforu...php?aid=73

Finally, where I really get into the technical details of my acceleration technique, is in this thread:
http://math.eretrandre.org/tetrationforu...php?tid=63

And then I posted code, which is quite out of date:
http://math.eretrandre.org/tetrationforu...php?tid=74

There was some misunderstanding by some forum members about what I was doing. For example, Henryk (Bo) thought that if I was adding a pair of logarithmic approximations (similar to the regular slog), I would need to divide by 2 in order to remain a true Abel function (see post #13 in the thread linked below):
http://math.eretrandre.org/tetrationforu...553#pid553

One of these days I'll get around to posting a formal explanation of why we don't need to divide by 2.

Anyway, I think Andrew was the first to really "get it". See post #2 in this thread, where he does a decent job of explaining my method:
http://math.eretrandre.org/tetrationforu...14#pid1814