project "Houdini, a great modeler"

   156507   609   9
User Avatar
Member
210 posts
Joined: Jan. 2014
Offline
McNistor
I never liked the marking menus inside Maya and besides, in Houdini there's no such thing and it will never be because marking menus are patented AFAIK.
Unfurtunatelly you are right but it doesn't has to be marking menus. These marking menus are actually just fancy designed context menus. Currently in Houdini, rightclicking in empty viewport gives you context menus which are rarely usefull because what you can reach through these context menus is
1. quite a bit far down in my prioritylist, I discribed earlier.
2. different on every node that is currently active.
Rightclicking in empty viewport and getting a list, where the first four entries are the component types (points, edges, prims, object), then a seperator followed by a list of the most commonly used commands. And now making it badass: Because no one knows what your most commonly used commands are, you can edit this list and arrange it on your own. So you get a list to choose the component you wanna work on followed by a custom list of your favourite commands (when I have the time, I'll prepare a screenshot for that idea). That solves two problems at once:
1. My hand can stay on the almighty triangle because it doesn't have to reach for 1, 2, 3, 4 all the time
2. No TAB searching anymore for each node
Cinema 4D has something similar. Righclicking gives you a list of all modelling tools (extrude, edgeloop, bridge, knife, ringselection, loopselection…and so on). Throughout my entire time working with C4D I think I've never used the UI to call these tools.


Edit:
I now took the time to prepare a screenshot for my rightclick idea. As I said befor, the list below the seperator is customizable so don't argue that the list on the picture doesn't fit your needs. And something I've already mentioned somewhere here in this thread should, of course, work as well the same way in this rightclick menu: Calling a command from here shouldn't require confirmation by pressing enter all the time. That's annoying.

Attachments:
quickSelect.jpg (81.2 KB)

User Avatar
Member
3925 posts
Joined: June 2012
Offline
For those wishing to enjoy a custom ‘toolbelt’ today, similar to as described before, you can drag each node into a shelf and then assign hotkeys to each tool by shift-control-option-LMB on each one.

Attachments:
Screen Shot 2014-05-27 at 10.49.53 am.png (13.1 KB)

User Avatar
Member
3925 posts
Joined: June 2012
Offline
jeff
RMB on any geometry in the viewer.
Select Geometry > Delete History option
Locks the current SOP and deletes the entire upstream SOP tree.

Still there.

Just tried this ‘delete history’; it's awful and shocking to reopen the scene and have no ‘diagram’ of the models construction.

For everyone advocating destructive modelling - try it out to cure your yearnings. Better node network organisational tools would take care of the gazillion nodes issue.
User Avatar
Member
1592 posts
Joined: March 2014
Offline
From all this talk about modeling in Houdini and how to “clean” its history I still don't know if there is a simple and/or efficient way of getting rid of modeling ops after one finishes a model, like Softimage (FreezeM) or even Maya (delete history).
So, is there a way to do this in Houdini or not and how if there's?
User Avatar
Member
151 posts
Joined: Dec. 2009
Offline
Just lock the last sop by clicking the node's red flag (the second one beside the bypass flag) in the chain and delete all the other operators.

Or export a file and load it back in with a file sop.

If you just want to tidy up the look of your network, you can encapsulate all the sops inside a subnet - without loosing the history.
User Avatar
Member
3925 posts
Joined: June 2012
Offline
More imagination is needed than Subnets, they are a blunt force tool as organisation toolsets, network boxes are bit better but are fidley; that is nodes get lost behind them, and they aren't fully robust for an artists workflow.

Sorting, grouping, scaling of node groups would help a lot. Think of packing UVs but for nodes. Efficient use of screen real estate should be one of the goal, readability too.
User Avatar
Member
151 posts
Joined: Dec. 2009
Offline
100% agreed.
My post was just meant as an answer to the question above.

Cheers
User Avatar
Member
6788 posts
Joined: July 2005
Offline
MartybNz
More imagination is needed than Subnets, they are a blunt force tool as organisation toolsets.

I must confess that I have no imagination. So what exactly are the disadvantages of using subnets to alleviate the need to delete history?
User Avatar
Member
6788 posts
Joined: July 2005
Offline
McNistor
So, is there a way to do this in Houdini or not and how if there's?

Yes, use the Delete History menu item as per jeff's comments above. The other comments were trying to dissuade you from doing so because it's not necessary to do so in Houdini, and trying to suggest better workflows.
User Avatar
Member
3925 posts
Joined: June 2012
Offline
edward
I must confess that I have no imagination. So what exactly are the disadvantages of using subnets to alleviate the need to delete history?

The main disadvantage is Subnets nests the nodes away- it's just one small step up from hiding nodes with shift-D. A goal is to go way beyond that to add efficiency and bring much nodal delight.

A Beyond The Subnet (BTS) system is related to the organisation of nodes in the network pane. Only the mad people, read non-enlightened-Houdiniless people, have a love of deleting history; most posts have a complaint with the unwieldily amount of nodes created.

BTS includes The Network Box v2 :

Smarter Network boxes - be inspired with UVLayouts box centric packing uv shells - in the movie below substitute uvshells with nodes:

http://www.headus.com.au/doc/uvlayout-expert/videos/UVLayout-Packing-2.mov [headus.com.au]

The idea is to layout network boxes and fit nodes into them. i.e. draw network box - select nodes and RMB on network box to select ‘pack nodes in here’.

Node layout controls/handles - select and scale nodes aka Houdini Channel Editor ‘Show Scale Handles’ As with keyframes, selected nodes could be moved and scaled.

Move node into network - a node can quite easily get stuck behind a network box currently, RMB menu on selected nodes would bring it into the box.


Houdini's FX toolset is unmatched- Let's raise the nodal toolkit to equal footing! Hurrah!
User Avatar
Member
6788 posts
Joined: July 2005
Offline
MartybNz
The main disadvantage is Subnets nests the nodes away- it's just one small step up from hiding nodes with shift-D. A goal is to go way beyond that to add efficiency and bring much nodal delight.

The only real differences between the concept of netboxes and subnets are that netboxes do not modify the path to the nodes. For the cases that the “delete history” feature is desired, I cannot see why that matters. Don't get me wrong though, I'm all for better implemented netboxes. However, I don't see how we do not satisfy the “delete history” people 100% today.

Perhaps this has only been implied thus far, so let me make it clear (mostly for McNistor). The workflow problems that “delete history” purports to solve are:
  • Performance, and
  • Complexity (Out of sight, out of mind)Collapsing nodes into subnets solves (2) and locking the subnet solves (1). Furthermore, collapsing subnets improves upon “delete history” in the sense that it is undoable because history was never actually deleted. And it is ok to have these nodes exist in your scene because they're lightweight behind the locked subnet since they do not cook.
User Avatar
Member
1592 posts
Joined: March 2014
Offline
Whoa, I inadvertently touched a sensitive cord within senior Houdini users apparently.

Now, I don't think wanting to delete modeling history is a sign of madness (I read it as you intended it Marty, don't worry).
The reasons why someone would want to do that were already mentioned above by edward.

If collapsing and locking does the same thing as FreezeM in Softimage (or delete history in Maya) in terms of performance gain as well as simplifying things while keeping a non-destructive nature, then I'd have to agree that it is a sign of madness to want to get rid of those nodes for good just because you feel uncomfortable knowing they exist somewhere.

Having said that, I know that “madness” and I understand because I've experienced it too. I remember when I first got into Softimage from 3dsMax I was bothered so much by the operators stacking up as I kept modeling that I worked in immediate mode for a few days and only had quit because for some reason the interface becomes non-interactive when having a PPG open in that mode.
I was a beginner in Softimage and 3d in general and now that memory makes me realize how silly I was and I chuckle inside.
User Avatar
Member
3925 posts
Joined: June 2012
Offline
Ya, if ‘freeze history’ rmb command was added, that subnetted and locked, that would probably satisfy.
User Avatar
Member
361 posts
Joined: Nov. 2008
Offline
MartybNz
Ya, if ‘freeze history’ rmb command was added, that subnetted and locked, that would probably satisfy.

Why? You proposed 2 clicks solution (First click, for rmb menu, second click for choosing the command).

But there is currently already 2 clicks solution: Click on subnet button (next to sticky notes) in top Network View menu and then click on lock icon on that new subnet.

You can even do it with one click shelf tool with a little bit of python code.
There is no need to clutter RMB menu for perfectly functional solution that is already there.
User Avatar
Member
3925 posts
Joined: June 2012
Offline
quite simple - there are many posts asking for freeze type functionality, if it's already there then the labelling is not clear enough for the common user.
User Avatar
Member
361 posts
Joined: Nov. 2008
Offline
MartybNz
quite simple - there are many posts asking for freeze type functionality, if it's already there then the labelling is not clear enough for the common user.

I think that new users are just too lazy to read forum or help (manual) first.
Look up in this thread, new users ask for questions while answers are just one post above their questions or on the second page of the manual.

Better and more detailed manual always helps a lot, but if user doesn't read it at all no matter how big any label is, it makes no difference.

And Delete History (that was asked) already is in RMB menu. :wink:
User Avatar
Member
3925 posts
Joined: June 2012
Offline
Ya- it's like RPN; it's meant to be quicker for the computer and people to use but people stick with infixing.
User Avatar
Member
1592 posts
Joined: March 2014
Offline
pezetko
I think that new users are just too lazy to read forum or help (manual) first.

True, but so many times reading the manual does nothing to help and digging the forums with search is time consuming and sometimes doesn't hit the nail exactly on the head which wouldn't be a big problem if you were experienced enough to adapt the found solution to your own problem.

And then there's the fact Houdini tries to be different than anyone else for no benefit. I'm obviously not talking about the things that are different and also making Houdini the powerful package that it is.
Fortunately those backwards things are not that many that if they were to be changed would break Houdini's philosophy which is proceduralism really, everything else being details.
User Avatar
Member
210 posts
Joined: Jan. 2014
Offline
MartybNz
More imagination is needed than Subnets, they are a blunt force tool as organisation toolsets
IMHO the same is true for most of the networks (SOP, POP, DOP, ROP…)
That's why I'm crying for a, I think it was called “heterogenous network”. Subnets indeed are not a very graceful artist tool, because they encourage one to end up with a subnet in a subnet in a subnet in a subnet…and thus the pathes to nodes are erratic. Kind of the same is true for all the different networks. They raise more problems than they solve. That's why they should be abandoned.

McNistor
pezetko
I think that new users are just too lazy to read forum or help (manual) first.
True, but so many times reading the manual does nothing to help and digging the forums with search is time consuming and sometimes doesn't hit the nail exactly on the head which wouldn't be a big problem if you were experienced enough to adapt the found solution to your own problem
+1
I would like to add that one big problem is, that searching for something becomes an issue very quickly when you don't know what to search for.
As a simple example:
If I was to search for saving a selection for later use, I'd have never come up with the idea to search for “group” or “grouping”. As I said, this is a trivial problem but I don't even know how many compareable issues are somewhere in the deepest and darkest corners of Houdini, probably because I don't know the right term That makes searching the help and the forums more often than not hard, annoying and frustratingly time consuming.
User Avatar
Member
3925 posts
Joined: June 2012
Offline
For the selection vs groups - it's simply a matter of moving from concrete to abstract reasoning; a skill that is essential for all creative work. If one looks for “A” and it's not there, what else could this functionality be called?

It was quite funny to read almost the exact same conversation on groups from some years back:
http://community.thefoundry.co.uk/discussion/topic.aspx?f=32&t=65575 [community.thefoundry.co.uk]

Edit: clarity
  • Quick Links