 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
An interesting thread on its own.
I think there might be some sigs in it that I haven't rendered yet as well, so
posting it here for general interest.
(I only got a 16 on the original test, haven't looked at the modified one(s) )
https://news.povray.org/povray.advanced-users/thread/%3C3c5977ae%40news.povray.org%3E/?ttop=444861&toff=1400&mtop=64348
- BE
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 2025-02-27 13:24 (-4), Bald Eagle wrote:
> What is probably a long while back now, I found it interesting to try and
> collect all of the different short-code signature scenes attached to forum
> members' posts, and render them to see what they looked like.
These are 20 that I've collected over the years. Some of them I've
added a #version or some wrapper code to. ron_parker-old.pov requires
COUR.TTF
I haven't looked at all of Bill P.'s files to match them against mine; I
named my files after the POVer.
Post a reply to this message
Attachments:
Download 'sigs.zip' (8 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Cousin Ricky <ric### [at] yahoo com> wrote:
> These are 20 that I've collected over the years. Some of them I've
> added a #version or some wrapper code to.
Yes, I've had a few that I needed to add commas to, so that 3.8 recognizes thing
like .8.8 as 2 different numbers.
I also got a weird warning rendering one that "POV-Ray 3.6.2" was trying to ...
do something .... :|
I'll have to see if I can replicate that one.
> ron_parker-old.pov requires
> COUR.TTF
Because he's a big cheater! :D
> I haven't looked at all of Bill P.'s files to match them against mine; I
> named my files after the POVer.
Thank you.
Some of these are pretty neat.
Update in next post.
- BW
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Bald Eagle" <cre### [at] netscapenet> wrote:
> So, I have so far managed to backtrack my way from the beginning, all the way to
> the first number: 11117333955
> But I need to add a 10-decimal place "correction factor" at each step.
In Excel and OpenOffice Calc,
> 0.0311342675 does the trick.
POV-Ray 3.8 needs 0.031130802
Now the interesting part.
I managed to hunt down a lot of information about solving for a particular N
that makes the two modulo operations work together, which gave me confirmation
about a "starting value" and the core multiplier.
I thought, "maybe the spreadsheet stuff that I have is only needed to backtrack.
Perhaps those very very specific values aren't needed, and freshly computed
values forward will give me the same decoded points when I run the macros..."
NOPE. Warp's algorithm is sensitive to even the last digit.
His number 11117333955 gives a nice, straight, crisp "W"
^
Even using 11117333956 makes it lean off to the right and just look wonky.
^
So if this whole thing is SO sensitive to the value of a fudge factor in the
generation algorithm, and the decoding algorithm doesn't even give results that
are close when the encoded value if off by a single digit - how the Hell did he
figure it all out and get it to work forwards and backwards?
I mean, I have some weird operations going on, where I did the (bad) thing of
letting the answers guide me, so I have no real idea why they're necessary to
get the proper intermediate values - but I don't really have any "first
principles" to guide me in place of that.
So I'm thinking that there needs to be a "cleaner" more reliable way to do the
encoding than what I'm doing.
- BE
Post a reply to this message
Attachments:
Download 'warpsig1.png' (48 KB)
Preview of image 'warpsig1.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Bald Eagle" <cre### [at] netscapenet> wrote:
> Now the interesting part.
I remember trying to crack it many years ago. This time I didn't try, I asked
Claude. It even generated Python code to create letters this way in POV-Ray. If
you like, probably not, I can post the link to the discussion (I think).
ingo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"ingo" <nomail@nomail> wrote:
> I remember trying to crack it many years ago. This time I didn't try, I asked
> Claude. It even generated Python code to create letters this way in POV-Ray. If
> you like, probably not, I can post the link to the discussion (I think).
I'm not sure I'll make any more enlightening progress on my own, so sure.
Did you see anything that was especially notable?
Were there fudge factors?
It seems to me like a highly complicated interweaving of integer math and very
small/specific fractional values.
So I f Warp didn't have AI available, I wondering how he puzzled this all out a
priori.
- BW
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Bald Eagle" <cre### [at] netscapenet> wrote:
> you like, probably not, I can post the link to the discussion (I think).
>
> I'm not sure I'll make any more enlightening progress on my own, so sure.
https://claude.ai/share/9e537cf6-2633-402d-91c0-974a3b6873a5
ingo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"ingo" <nomail@nomail> wrote:
> "Bald Eagle" <cre### [at] netscapenet> wrote:
> > you like, probably not, I can post the link to the discussion (I think).
> >
> > I'm not sure I'll make any more enlightening progress on my own, so sure.
>
>
> https://claude.ai/share/9e537cf6-2633-402d-91c0-974a3b6873a5
>
> ingo
Thanks for this.
I looked it all over, 2 or 3 times, and I think the AI glosses over the details
of the iteration. Thanks for the link - perhaps I'll be able to implement some
more versions of this and see if there's anything that I've missed so far.
(one thing that I've been trying to take advantage of with AI is getting it to
write critical pieces of code for things we've been asking for, but writing
macros from scratch is devilishly difficult. Singular Value Decomposition, data
structures, algorithms, FFT, patterns, all sorts of stuff.)
- BW
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Bald Eagle" <cre### [at] netscapenet> wrote:
> one thing that I've been trying to take advantage of with AI is getting it to
> write critical pieces of code for things we've been asking for[...]
>
I tried a few things in the past but most of the time it was clueless. Claude
seems to work better than OpenAI. Recently I had a chance to play for a short
while with a payed version of Claude 3.7. It's a lot better than earlier ones.
When one can keep the discussion on an aspect small and focussed, Claude 3.5 is
mostly fine.
ingo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"ingo" <nomail@nomail> wrote:
> I tried a few things in the past but most of the time it was clueless. Claude
> seems to work better than OpenAI. Recently I had a chance to play for a short
> while with a payed version of Claude 3.7. It's a lot better than earlier ones.
>
> When one can keep the discussion on an aspect small and focussed, Claude 3.5 is
> mostly fine.
I was able to play with M$ Copilot for a bit yesterday, and it's possible to get
a lot more out of it if you don't ask it to do it all at once.
(I was actually surprised how much you were able to get with your simple initial
prompt!)
Ask it to do a small bit, then the next bit, then the next bit, then ask it "can
these all be done together in one scene / macro?
So I got it to spit out some Householder transformation code, and then some QR
decomposition code, and Hopefully I can use that as a guide to get the SVD
working in SDL.
I'd also like to get the red-black data structure hammered out, and then get
Fortune's algorithm working for a Voronoi diagram.
Maybe we can't get everything we want implemented in source (yet), but we can
get a bunch of features implemented via macros - however slow they might be.
I'm wondering if a concerted effort to leverage all of the AI's out there might
let us start to unravel all of the parser stuff and get that all sorted out so
that POV-Ray 4.0 can start to be worked on.
- BW
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
