Why not petition SDK improvement ? on a side note, has anyone seen the updates to mudbox ?! the new topology tools look pretty sassy !
On 10 April 2013 18:21, Eric Cosky <e...@cosky.com> wrote: > Just to add to the discussion, when I made EssGeo (a geometry plugin, > http://www.boundingboxgames.com/tools/essgeo) I had to disable a couple > of features due to problems with cluster management. Namely, a random > cluster op and a greeble op. I think they could have been pretty useful > tools.**** > > ** ** > > I made an imgur gallery that showed the features I had to abandon, check > it out here: http://imgur.com/a/5T7v0**** > > ** ** > > It’s been a while since I wrote all this, but if memory serves I couldn’t > find what I needed for cluster management in the C++ API, so I had to > resort to this sort of thing which basically builds a VB command to make > changes to clusters:**** > > ** ** > > > CString prefix = cluster.GetFullName() + L", " + pname + L".poly[";**** > > > if(ecount > 0)**** > > > {**** > > > CString arg = prefix + L"0-" + CString(CValue(pcount-1)) + L"]";**** > > > args[0] = arg;**** > > > status = Application().ExecuteCommand(L"RemoveFromCluster", args, val);*** > * > > > DEBUG_ASSERT_OK(status);**** > > > }**** > > **** > > ** ** > > This of course is pretty slow compared to a proper native API, but it > generally solved that part of the problem and in practice wasn’t a > performance problem since clusters didn’t change all the time.**** > > ** ** > > Unfortunately I had to yank these features because I could never get them > stable. Changing anything that changed cluster data had the potential to > crash Softimage. I usually assume when something doesn’t work that it’s my > fault, because there always seems to be a tendency for things to actually > turn out to be my fault the moment I point the finger somewhere else, but > in this case I wound up convinced the cluster management internals of > Softimage was buggy and there was nothing I could do about it. I’d love to > be wrong so I could fix my plugin but this thread suggests otherwise.**** > > ** ** > > If someone on the Softimage dev team would like the source code to this > plugin along with an already-built version with these features enabled in > order to fix the crash and see a specific example of what cluster API > changes would be useful, I’d be happy to provide it.**** > > ** ** > > -Eric Cosky**** > > ** ** > > ** ** > > ** ** > > *From:* softimage-boun...@listproc.autodesk.com [mailto: > softimage-boun...@listproc.autodesk.com] *On Behalf Of *Sebastien Sterling > > *Sent:* Tuesday, April 09, 2013 5:40 PM > *To:* softimage@listproc.autodesk.com > *Subject:* Re: Softimage 2014**** > > ** ** > > If these things are to hard to accomplish for third party people, then > what realistic chance is there that they will ever be implemented ? is what > i want to know, Autodesk don't exactly have a good track record of treating > there customers as a valid source of input... is there some secret ballot > where this stuff gets decided ?**** > > Also, is the problem that the SDK is just too archaic ? does it need a > complete rewrite ? or are aspects of the code unavailable or illegal to be > changed to/by scripters ? if so does Autodesk have the ability to make the > code available ?**** > > ** ** > > On 9 April 2013 09:27, Eugen Sares <softim...@keyvis.at> wrote:**** > > Oil on my fire. > That cluster SDK restriction really really sucks. It is the reason why > there never were any good topology/modelling addons from 3rd parties, which > leads to stagantion if there aren't any new "factory" modelling tools > brought also. > In 3ds max or Maya, all kinds of plugins are available, completely > natural. Not so in Softimage. > The few ICE modelling tools like Cap are nice, but slow. Native code is > nice and fast. > > Luc-Eric mentioned once, ICE was meant to be the "new SDK", that's why > this cluster update mechanism has been implemented for ICE already. > Imho that's an excuse. ICE complements the SDK, it is NOT a replacement! > Cluster updates should be supported by the SDK as well, even if it is > complicated, and thus somewhat of a challenge for a 3rd party dev. > Try us! Provide a good code example alongside, and we'll do fine. > > Be wise and do it. Please. > > > > Am 09.04.2013 09:08, schrieb Piotrek Marczak:**** > > Just give us proper SDK and let community do the rest. **** > > ** ** > > "**** > > Softimage currently does not fully support custom topology operators. The > problem is that any cluster or cluster property will not properly update > when a topology operator adds or removes points that belong to the cluster. > In the worst case Softimage may crash. Hence custom topology operators > should only be used in the more limited scenario of objects that do not > have any clusters. Once the geometry is ready it would be possible to > freeze the object to remove the custom topology operators (but leave the > result of their evaluation), then to add the clusters and other operators. > **** > > "**** > > ??**** > > ** ** > > 2013/4/8 olivier jeannel <olivier.jean...@noos.fr>**** > > They do modo for birds ? > > Le 08/04/2013 20:27, pete...@skynet.be a écrit :**** > > I’m pretty sure there’s neither gentlemen nor ladies on this list.**** > > **** > > as for Modo vs SI – a little bird tells me there’s more important issues > at stake than selection.**** > > **** > > **** > > *From:* Rob Chapman <tekano....@gmail.com> **** > > *Sent:* Monday, April 08, 2013 3:35 PM**** > > *To:* ron...@toonafish.nl ; softimage@listproc.autodesk.com **** > > *Subject:* Re: Softimage 2014**** > > **** > > now now gentlemen, there are ladies present on the list too! **** > > **** > > lets just say , when it comes to apps and selection methods, leave the > race courses for the race horses..!**** > > **** > > :)**** > > **** > > **** > > ** ** > > On 8 April 2013 15:32, Toonafish <ron...@toonafish.nl> wrote:**** > > ...but I prefer brunettes with bigger boobs. If you get the idea J > > That's prolly because bigger boobs aren't obstructed so much, so they are > much easier to select in shaded mode ;-) > > - Ronald**** > > **** > > ** ** > > ** ** > > ** ** > > ** ** >
