END OF AN ERA, FRACTALFORUMS.COM IS CONTINUED ON FRACTALFORUMS.ORG

Luca-
  Should I use the CL register for loops if I save ECX somewhere, or should I use AL instead since ECX is used for z? 
  Would this code snippet (assuming I have to use AL, and use defined Option work?

  UPDATE  CUBE TYPE WAS WRONG.. WILL COMPLETE UPDATE LATER!!! SORRY FOR THE SHODDY WORK!!!


 Thanks!   Also... would the code that I posted work?  I guess I have to try.  


  Here Luca..   I put pixel multipliers in these, so that we don't have to use an additional formula to remove pixel components.

  The Square Type does what it says in the description: T1, unSpheric (sphere to cube), Pine Tree, Spheric so you don't have to use a bunch of formula tabs.  Use it alternating with the other formulas for cool stuff.

  If you do a few T1s first, square type, a couple of Pine Tree 1, loop it....   it really squares it up.  

  So you can take out the "no yz" versions of Pine1, Pine2, and PineX and PineXYZ.  

I added an option to the T1 with Pine Tree, so that you could do multiple T1s in one formula.  It allows you to do some of the things I've been doing recently.  


  Next project is to work on the polygonator and non-polygonator- extend the spheric functions to allow a little bit more stuff.  Still have not figured out the correct regular Tetrahedron- I might have done it.. but I can't tell.

  I'm probably going to have to ask Knighty for help, after I clean up the code a bit.  It might sound silly, but I really want to combine a Sierpinski tetrahedron with a Menger cube... so need cube to sphere (we have that), sphere to tetra (we have that, but the tetrahedron is not regular!!!), tetra to sphere (I think I can do that), then sphere to cube.  


  Also, made a 4d hyperspheric, 4d unhyperspheric, although I think the math is incorrect so will hold off on posting til I check them- might be my assembly, maybe I did the incorrect order of operations, or maybe it just doesn't really work that well.

  4d Transform 1 is also included (it works), with an optional version that preserves the original magnitude of x,y, and z, while still altering details a bit.  I don't think either does a lot, but it makes the extension of Tn (T1,T2,T3...) to higher dimensions a little easier to understand, since they all use magic angle rotations.  It does change the fractals a bit, and maybe mixing it with other 4d fractals will be interesting.  

 @Luca-  I'm not sure if I did the division correctly at the end of the 4d T1 with xyz clamp.  I saved the original magnitude in EBP-20, then need to divide the stack st(0) by it at the end, so do I use DC75E0 or DC7DE0?

  I would think DC7D, but I don't know if the reference (Art of Assembly) I'm using is correct (or if my interpretation of it is!), because I had to reverse a whole bunch of divisions in another formula to make it work.... I can't tell by looking at the fractal.

  Don't know how to run the script in Ollydebug to check the math, which would make it so much easier, if you have any tips for me.  Thanks for all you've done so far. 

4dT1:

3dT1:

You can not execute a full script in olly but you can test something manually modifying the floating stack then executing the code (saving - fstp -not always works! Except if you are able to modify EBX EDX ECX to some writeable memory... argh)
You then highlight where you want to run then run line by line... watch what happens.

  Has rotation at specific iterations in the right place in the formula, so you don't have to do T1, rotation, pine tree (3 formula tabs).

  Going to make one with the squaring thing on specific iterations too.  I'll try how to do it every 2 iterations or something like that....

  Still have the polyhedron transforms to add to this!  Lots of work. 


  <a href="https://www.youtube.com/v/ypxR2sQAnws&rel=1&fs=1&hd=1" target="_blank">https://www.youtube.com/v/ypxR2sQAnws&rel=1&fs=1&hd=1</a>

Ok.  I put the spheric, after the Pine, after the cube, after the 1, probably because I swallowed a fly.  And I don't know why I swallowed a fly.  Perhaps I'll die.

  Anyway.  Here is a rotatin'pineCube and a rotatin'pineSquare fer yins.

  Square just squares it up.  Pretty sharp.  The cube one, well, it sort of stretches it out a bit.  Kind of nice too.  I'm liking the square though.  Going to make polygonal and polyhedral formulas for it, and add them, without the regular tetrahedral transform (just the general one I did for the Menger).  

  Then maybe make a square every iter??  // scratch that.. not too pretty.  Maybe set it up so that it does it every few iterations?  Or just have a formula that always does squares and mix that with formulas that don't.  

Square:
 
  
Cube:

  

Random programmer comment/observation: if you'd like to "future proof" these formulae, or simply make them usable in other systems, there should also be a plain C (or Pascal, or whatever) version besides the assembly code. I think x87 code written in 2015 (!!!!!) has way too high chance of being "lost in the sands of time", and this seems like a good effort so it would be a shame.