The Persistence of Vision Raytracer (POV-Ray).
This is the legacy Bug Tracking System for the POV-Ray project. Bugs listed here are being migrated to our github issue tracker. Please refer to that for new reports or updates to existing ones on this system.
Attached to Project: POV-Ray
Opened by Jaap Frank - 2010-12-06
Last edited by Thorsten Fröhlich - 2011-08-23
Opened by Jaap Frank - 2010-12-06
Last edited by Thorsten Fröhlich - 2011-08-23
FS#176 - Raise maxpower of the Poly Oject to 16.
At the moment in the Poly Object the maximum power is 15. The mathematics for converting the three parametric equations for x, y and z into a formula for the Poly Object require that the equations are squared several times given max-powers of 4, 8 and even 16. I’ve one eqaution that needs power 16. At the moment this is just one power short. Please raise this to 16. That’s all I ask for.
Can you please provide an illustrating equation of order 16 ?
(I'd like to see it, assuming it's not a trivial squared lower order; it would serve as test case)
You are aware that order 15 already need 680 coefficients, and order 16 will go to 816. I wonder if an alternate syntax would not be welcome (but I need a real non trivial equation). (ala mesh2... same object, different syntax)
Would you stop at 16, really ?
Short notice about a lazy solution: raising MAX_ORDER might be sufficient, but the computation of the binomial would be redone each time (and they are very slow, as we need to compute x!/1) without actually computing x! (which goes over the roof very quickly). But at least the code is there (as long as the parser let it enter).
Minimal optimisation would be to extend the binomial array from 15 x 15 to higher values.
(the factorial code so far is dead code, a caching mechanism would be better, and we drop that array too ?)
Also the reducing of factor could be more effective with a list of primes instead of testing all the odd numbers.
Comments in code talk about factoring up to 2^15, but as the code is so old, it might have been a dependency on the local int architecture of that time (with 16 bits signed int). Anyway, coefficients from pascal triangle could also be computed differently (as sum of two previous cells), so that nice factoring code can probably be removed (a bit of shame, it's a very nice trick) and the array of MAX_ORDER x MAX_ORDER automatically filled on demand (well, if someone is ready to enter the (n+1)(n+2)(n+3)/6 parameters of the order n poly.
Bonjour Grimbert,
I've posted a zip file in povray.binaries.scene-files for you. I didn't know you could put files here too.
You will find the same files here now.
Read my comment file for you first.
Greetings,
Jaap Frank
Evolution made in beta 41, with an additional syntax to avoid entering all the coefficients (when they are 0) and hopefully easier to read & write (to check with the original developed polynomial (see release-note until documentation get updated)
I'd appreciate some clarification for the docs:
http://wiki.povray.org/content/Documentation:Reference_Section_4.2#Poly.2C_Cubic_and_Quartic
Is that where changes go? If so would you mind working out something I can add? You can either do it on a talk page or you have the juice to do it direct ... whatever your comfort zone is I'm OK with either choice
Yes. Updated section in talk page. (Proof reading of added text by a native English speaker is welcome, i'm too tired to detect my errors at this time of the year.)