Yes, Volume VOPs I have been using a lot and seem the right way to go for the noise functions and simple range fit, clamps, gain, plus/minus/mul etc.
For anything point related which we would do in ICE previous sounds like all wrangles (minus the noise stuff you mentioned). On Wed, Apr 26, 2017 at 4:05 PM, Andy Nicholas <[email protected]> wrote: > VEX Wrangles nearly all of the time. Usually the only reason I go into > VOPs is if a) I need to use something like the Anti-Aliased Noise VOP > (which can’t be called as a function directly from VEX), or b) if I’m doing > shaders. For some reason I find shaders much easier to deal with in VOPs. I > suspect it’s partly because I just haven’t used them enough, plus > testing/debugging VEX is generally a bit of a nightmare. > > FYI, I’m working on a solution for making noise much easier to use in VEX > and VOPs as part of the siLib initiative. I know it doesn’t exactly sound > exciting, but from my experience working with existing noise functions in > Houdini, I think a lot of folk will find it extremely useful. I’ll explain > more when I finish it. Show, not tell :) > > > On 26 Apr 2017, at 23:47, Steven Caron <[email protected]> wrote: > > OK, great... I can see how to customize it from that snippet. > > Just to understand how the power users are using this. Are you using > wrangle nodes with vex snippets 100% of the time or are you using the VOP > sub graph for somethings? > > Steven > > On Wed, Apr 26, 2017 at 11:38 AM, Andy Nicholas <[email protected]> > wrote: > >> If you have a look at the docs for pcfilter, it gives you some VEX code >> that’s the equivalent functionality. You can tweak to suit your needs. >> >> Pasted in from the docs: >> >> - - - - >> >> float pcfilter(int handle; string channel) >> { >> float sum, w, d; >> float value, result = 0; >> while (pciterate(handle)) >> { >> pcimport(handle, "point.distance", d); >> pcimport(handle, channel, value); >> w = 1 - smooth(0, radius, d); >> sum += w; >> result += w * value; >> } >> result /= sum; >> return result; >> } >> >> pcfilter takes the points that were opened by the point cloud and >> produces a filtered value. The following equation shows how the individual >> points are weighted. >> >> w_i = 1-smooth(0, maxd*1.1, d_i); >> >> maxd is the farthest point, and w_i is the weight for a given point at >> distance (d_i). Points that are closer to the center will be weighted >> higher with that formula, rather than it being an average. >> >> >> On 26 Apr 2017, at 19:25, Steven Caron <[email protected]> wrote: >> >> So pcfilter has weighting built into it? >> >> I guess I want to customize this weighting should I not use pcfilter? and >> use pcfind and loop over the particles? >> >> >> On Wed, Apr 26, 2017 at 11:18 AM, Andy Nicholas <[email protected]> >> wrote: >> >>> Yep, I’m afraid I’m a complete VEX convert now. I never used to be! >>> >>> Cris, don’t mind me posting stuff like that, I’m just doing it in case >>> it’s useful to anyone who’s trying to get into VEX. Actually, when I was >>> learning to do all the point cloud stuff, I found it useful to see the VEX >>> when I was trying to understand how to hook up the pointcloud VOPs. I >>> (still) don’t think it’s obvious how to do it, especially when you start >>> thinking about if it’s necessary to use pcclose() and where that should be >>> wired. >>> >>> It’s one of the reasons I moved to VEX, things like “loops" and “if” >>> statements are much easier to understand in VEX than VOPs. >>> >>> >>> >>> On 26 Apr 2017, at 19:07, Steven Caron <[email protected]> wrote: >>> >>> Thanks guys! I didn't want to use VEX even though I might need to in the >>> long run. >>> >>> On Wed, Apr 26, 2017 at 10:57 AM, Andy Nicholas <[email protected]> >>> wrote: >>> >>>> Yep, what Cris said. Here’s some VEX you can drop in a Point Wrangle if >>>> you want to try that approach: >>>> >>>> float radius = 1.0; >>>> int maxpts = 50; >>>> int handle = pcopen(0, "P", @P, radius, maxpts); >>>> @P = pcfilter(handle, "P"); >>>> >>>> >>>> On 26 Apr 2017, at 18:37, Cristobal Infante <[email protected]> wrote: >>>> >>>> pcopen > pcfilter (P) will give you the nearest positions. >>>> >>>> >>>> >>>> On 26 April 2017 at 18:27, Steven Caron <[email protected]> wrote: >>>> >>>>> i hate to do it but i gotta ask this group because of our shared ICE >>>>> knowledge... >>>>> >>>>> what is the proper way to get closest points and average their >>>>> position and update the point position? >>>>> >>>>> pcfind gives me an integer array, but how do i look up those indices >>>>> and get their point position? >>>>> >>>>> pcopen, then pcimport, do work, then pcexport seems like the right >>>>> thing to do, should i just ignore pcfind? >>>>> >>>>> thanks >>>>> steven >>>>> >>>>> ------ >>>>> Softimage Mailing List. >>>>> To unsubscribe, send a mail to [email protected] >>>>> with "unsubscribe" in the subject, and reply to confirm. >>>>> >>>> >>>> ------ >>>> Softimage Mailing List. >>>> To unsubscribe, send a mail to [email protected] >>>> with "unsubscribe" in the subject, and reply to confirm. >>>> >>>> >>>> >>>> ------ >>>> Softimage Mailing List. >>>> To unsubscribe, send a mail to [email protected] >>>> with "unsubscribe" in the subject, and reply to confirm. >>>> >>> >>> ------ >>> Softimage Mailing List. >>> To unsubscribe, send a mail to [email protected] >>> with "unsubscribe" in the subject, and reply to confirm. >>> >>> >>> >>> ------ >>> Softimage Mailing List. >>> To unsubscribe, send a mail to [email protected] >>> with "unsubscribe" in the subject, and reply to confirm. >>> >> >> ------ >> Softimage Mailing List. >> To unsubscribe, send a mail to [email protected] >> with "unsubscribe" in the subject, and reply to confirm. >> >> >> >> ------ >> Softimage Mailing List. >> To unsubscribe, send a mail to [email protected] >> with "unsubscribe" in the subject, and reply to confirm. >> > > ------ > Softimage Mailing List. > To unsubscribe, send a mail to [email protected] > with "unsubscribe" in the subject, and reply to confirm. > > > > ------ > Softimage Mailing List. > To unsubscribe, send a mail to [email protected] > with "unsubscribe" in the subject, and reply to confirm. >
------ Softimage Mailing List. To unsubscribe, send a mail to [email protected] with "unsubscribe" in the subject, and reply to confirm.
