kev2 Posted March 3, 2017 Share Posted March 3, 2017 Is it me or have the group nodes mulitplied like rabbits? After unifying noise, shaders, etc, I don't understand how node multiplicity makes sense elsewhere. Maybe SESI could weigh in on the thinking? Quote Link to comment Share on other sites More sharing options...
Guest tar Posted March 3, 2017 Share Posted March 3, 2017 @michael yep exactly - copy/paste works well but it would even be more fluid, and exciting, to be able to drag-copy a node or bunch of nodes of the same type between networks. Quote Link to comment Share on other sites More sharing options...
michael Posted March 4, 2017 Share Posted March 4, 2017 23 hours ago, kev2 said: Is it me or have the group nodes mulitplied like rabbits? After unifying noise, shaders, etc, I don't understand how node multiplicity makes sense elsewhere. Maybe SESI could weigh in on the thinking? there were many people telling us that the 'one node to rule them all' was often confusing and they had trouble figuring out where/what was going on, and there was interest here to split them up to make fixes/features easier to do in simpler code. 19 hours ago, marty said: @michael yep exactly - copy/paste works well but it would even be more fluid, and exciting, to be able to drag-copy a node or bunch of nodes of the same type between networks. personally I find copy/paste faster, easier and far less error prone than dragging nodes around. and there would have to be lots of logic for what you want to do after the drop - copy of nodes, move nodes, make reference etc RFE it if you think it's worth it 1 Quote Link to comment Share on other sites More sharing options...
Guest tar Posted March 5, 2017 Share Posted March 5, 2017 Warning peoples - h16 Boolean has 4 aliases in the viewport!!! run for the hills as y'all be confused and annoyed by this duplication - hahaha Quote Link to comment Share on other sites More sharing options...
Werner Posted March 6, 2017 Share Posted March 6, 2017 Move, rotate and scale components independently on their local axis...as per Mcnistor's feedback. 4 Quote Link to comment Share on other sites More sharing options...
kev2 Posted March 6, 2017 Share Posted March 6, 2017 20 hours ago, marty said: Warning peoples - h16 Boolean has 4 aliases in the viewport!!! run for the hills as y'all be confused and annoyed by this duplication - hahaha Rabbits. Quote Link to comment Share on other sites More sharing options...
nisachar Posted March 8, 2017 Share Posted March 8, 2017 Make the poly split node handle multiple splits under one node. Just have the split operations appear as a list in the node's Parameter window, or display them in the group/expression slot ( 1-2, 4-7, 12-63 etc ). Color code the newly generated edges when the node is selected and if the user hovers the cursor over any split edge, the group list automatically highlights the connected points/edges/faces. The user can then remove them from the list and thus undo that particular split ( without having to invoke a blast node ) Quote Link to comment Share on other sites More sharing options...
Werner Posted March 8, 2017 Share Posted March 8, 2017 (edited) 'Ignore point connectivity' in Edit SOP is a mess. We need something more intelligent along the lines of edge detection. It's very hard to create nice mouth and eyelid blend shapes on high density geometry with the current method. This is a very long standing RFE (ID: 72902) Old thread about this.https://www.sidefx.com/forum/topic/5114/ Edited March 8, 2017 by Werner image upload 2 Quote Link to comment Share on other sites More sharing options...
kev2 Posted March 9, 2017 Share Posted March 9, 2017 A way to temporarily disable a merge node connection (w/o disconnecting it). Alternatively, discrete merge node inputs with additional ones created on the fly. Quote Link to comment Share on other sites More sharing options...
Guest tar Posted March 9, 2017 Share Posted March 9, 2017 @kev2 the yellow bypass toggle does that. Quote Link to comment Share on other sites More sharing options...
kev2 Posted March 9, 2017 Share Posted March 9, 2017 9 hours ago, marty said: @kev2 the yellow bypass toggle does that. Oops. I meant a single merge node connection, not the entire merge node. I realize that the switch does that but I often want to audition different inbound pipes to a merge node without running around setting the display flag. I edited the post to make more sense. Quote Link to comment Share on other sites More sharing options...
angelous4x Posted March 9, 2017 Share Posted March 9, 2017 skip rendered frames? 1 Quote Link to comment Share on other sites More sharing options...
sebkaine Posted March 16, 2017 Share Posted March 16, 2017 ALT+E in a wrangle open the editor define in $EDITOR directly. At the moment to open your editor you have to use the redundant 2 * ALT+E. 1 Quote Link to comment Share on other sites More sharing options...
jonmoore Posted March 16, 2017 Share Posted March 16, 2017 Just now, sebkaine said: ALT+E in a wrangle open the editor define in $EDITOR directly. At the moment to open your editor you have to use the redundant 2 * ALT+E. I'm torn by this one as it always used to annoy me that I had to go through an interim step of opening the Houdini text editor before Sublime Text but as I've become more adventurous with VEX I've come to rely on the error messages in the Houdini text editor. Maybe an option to open both windows at the same time when using ALT E. Quote Link to comment Share on other sites More sharing options...
sebkaine Posted March 16, 2017 Share Posted March 16, 2017 (edited) in fact jonathan that might be better to be able to use an other var like $SNIPPET - to allow people who like to keep the current state in snippet to use only $EDITOR, - and those who want to open sublime directly in a wrangle to define also $SNIPPET to sublime. Edited March 16, 2017 by sebkaine Quote Link to comment Share on other sites More sharing options...
jonmoore Posted March 16, 2017 Share Posted March 16, 2017 @sebkaine, maybe. As long as it's on sesi's radar as a workflow that needs improving, I'm happy to see what they come up with, as I'm sure there are other considerations too. Quote Link to comment Share on other sites More sharing options...
Thomas Helzle Posted March 16, 2017 Share Posted March 16, 2017 ...or the internal editor could be brought up to snuff, would be even better... ;-) (sorry, couldn't resist) Cheers, Tom 2 Quote Link to comment Share on other sites More sharing options...
jonmoore Posted March 16, 2017 Share Posted March 16, 2017 14 minutes ago, Thomas Helzle said: ...or the internal editor could be brought up to snuff, would be even better... ;-) (sorry, couldn't resist) Cheers, Tom Agreed. I'd still want a fluid path to a third party editor, but I'm hoping the internal editor is given some love too. 2 Quote Link to comment Share on other sites More sharing options...
Guest tar Posted March 18, 2017 Share Posted March 18, 2017 It would be nice to iron out the delay in Mantra starting a new frame. This slight pause adds a lot of accumulated time when the actual frame is quick to render. GPU renderers excel at this. They kick off the next frame instantly. Quote Link to comment Share on other sites More sharing options...
McNistor Posted March 19, 2017 Share Posted March 19, 2017 (edited) I'd very much like to see the new Hair system continuing to be improved in both performance - reduce the lag of brushes, decrease cooking speed, etc - as well as features. Regarding features for example, the new system no longer allows the user to use a texture-map or a paint sop for using as a multiplier for the thickness ramp, might be possible following Jeff' suggestions (on sidefx forums) but it's akin to masochism as far as I'm concerned. I'm looking for high-level tools to help do the thing quickly, while also allowing for going in depth to low-lvl for TAs that like tinkering, but Houdini's already good at this. The old Hair system also had a feature called "guard hairs" which is no longer present as far as I'm able to figure. Might be able to achieve the same effect by randomizing length, don't know, but the old method felt a bit more straight forward. I've started working on a model with which I intend to "drive test" the new system so that I'll be able to give a more detailed feedback while learning more about it. Something else I'd like to see in the very next H version, is a foot on the gas pedal regarding implementing modeling and viewport features. If the current speed is maintained, 2023 seems like a realistic ETA for when I'll be able to use Houdini as proficiently as XSI for modeling and other viewport centered work. Speaking for myself here obviously. Edited March 19, 2017 by McNistor Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.