Houdini 17 Wishlist

[sebkaine]

glTF 2.0 exporter 

https://www.khronos.org/news/press/khronos-releases-gltf-2.0-specification

[Wooshum]

A full implementation of a weighted straight skeleton operation similar to the one developed by Tom Kelly:

https://github.com/twak/campskeleton

http://theses.gla.ac.uk/4975/ 

Roofs, procedural building models, etc. At a SOP level this would be epic, PolyExpand2D with way more functionality

Edited July 20, 2017 by Wooshum

[mestela]

Just now, Wooshum said:

A full implementation of a weighted straight skeleton operation similar to the one developed by Tom Kelly:

https://github.com/twak/campskeleton

http://theses.gla.ac.uk/4975/ 

Roofs, procedural building models, etc. At a SOP level this would be epic, PolyExpand2D with way more functionality

+1000, so many uses

[Guest tar]

Would like an 'expert' mode for the Network View where you don't need to select a node to dive into it, you simply have the mouse over the node when you preemptively 'i'.  The select then press 'i' is taking too much time 

 

Submitted RFE #84036.

Edited July 24, 2017 by tar

[animatrix]

18 hours ago, marty said:

Would like an 'expert' mode for the Network View where you don't need to select a node to dive into it, you simply have the mouse over the node when you preemptively 'i'.  The select then press 'i' is taking too much time 

+1, this is a nice RFE. I wrote shelf scripts that does this for Display flag so it's possible to script this action as well. I can send you the code if you want that you can bind to a hotkey.

 

My RFE: In H16 if you copy some text inside Houdini such as code from inside an AttribWrangle and then select and copy a node tile, and then paste inside an AttribWrangle, Houdini will paste the path of the node inside the text editor.

Before 16, it didn't do this, it preserved node tile copy clipboard separate from textual clipboard.

If this is a new feature then it should have its own shortcut not alter the default copy/paste behaviour people have used to for a long time IMO. [ID# 84001]

Edited July 24, 2017 by pusat

[Guest tar]

@pusat - very cool tip about the copy node path thingy - learnt something new. thx!  

[LaidlawFX]

I never thought I would be using cops this much... but please, oh please give this a serious under the hood fix... 

I never thought it would be so buggy, and a pain to work with... ARGGHHHHHHH

[Guest tar]

On 27/07/2017 at 3:13 AM, LaidlawFX said:

I never thought I would be using cops this much... but please, oh please give this a serious under the hood fix... 

I never thought it would be so buggy, and a pain to work with... ARGGHHHHHHH

So not that much different from Nuke then  Which bugs are you hitting?

[pxBomber]

4 minutes ago, marty said:

So not that much different from Nuke then  Which bugs are you hitting?

Regardless of bugs you really cannot compare to nuke. I'm not a compositor - but even for simple slapcomps you have to be really lucky to render what you want from COPs. 

Sad is that everyone I know can see the potential (VOPs, same env paths, super easy pipeline integration for hou-based studios, etc.) but it's not wise to use it i production when you have a deadline.

To ask which bugs you mean is the same like to ask which features are you missing in MS-paint . (no offence) 

[LaidlawFX]

59 minutes ago, marty said:

So not that much different from Nuke then  Which bugs are you hitting?

It's something to share over a beer. The funny thing in all of this, is Houdini used to sell it's compositing package as Halo. Which I'm willing to bet lots of money on, was where DD based Nuke off of, and when Nuke started selling, Halo stop selling. 

 

[symek]

14 hours ago, LaidlawFX said:

It's something to share over a beer. The funny thing in all of this, is Houdini used to sell it's compositing package as Halo. Which I'm willing to bet lots of money on, was where DD based Nuke off of, and when Nuke started selling, Halo stop selling. 

 

There was a Charlie and Rayz (by Silicon Grail), with the latter one using Halo technology licenced from SESI, they were much more serious about compositing than SESI ever was even before Nuke was released. Halo has lost with Shake mostly (not to mention binding compositor to HDK with all its limitations back then, wasn't very promising idea). 

[sebkaine]

@symek just a small erratum . I think it was Chalice and Rayz.

http://web.archive.org/web/19990127211724/http://www.silicongrail.com:80/ 

Silicon Grail Corporation is a joint venture between Ray Feeney and Side Effects Software, with involvement from Richard Hollander, Greg Hermanovic, Kim Davidson and Wyndham Hannaway. Silicon Grail is dedicated to creating software solutions to meet the needs of the film professional. The first product from Silicon Grail is Chalice, a 2D image composition and adjustment product.

The product was buy and kill by apple like shake ... 

http://www.linuxjournal.com/article/5933

Back on topic i agree with you guys some cops love would be nice !

- 2d tracker 

- ofx plugin support

- better RGB curves

- better HSV curves

- good / fast defocus

basically something that give us the exact same feature as shake 2.5 would be good enough. 

Edited August 3, 2017 by sebkaine

[Mandrake0]

I have made long time ago a RFE for enhancing the Composition part and also integrate also NLE so you get all functions in one tool. Time for more RFE's 

[animatrix]

Before any new serious COP features though, SESI needs to fix some core bugs that are as old as the system itself. I reported many but I get the feeling that there are no plans to address any of them. There are threading issues, deadlocks, round trip (COP to SOP to COP) limitation, GPU image artifacts, anti-aliasing issues using the exact same code in VOPs vs VEX, etc.

Otherwise I really like COPs too. I made an OpenCL Wrangle COP recently that proves to be very useful for image filters.

[symek]

1 hour ago, pusat said:

Before any new serious COP features though, SESI needs to fix some core bugs that are as old as the system itself. I reported many but I get the feeling that there are no plans to address any of them. There are threading issues, deadlocks, round trip (COP to SOP to COP) limitation, GPU image artifacts, anti-aliasing issues using the exact same code in VOPs vs VEX, etc.

Otherwise I really like COPs too. I made an OpenCL Wrangle COP recently that proves to be very useful for image filters.

That's the whole point and the problem SESI is facing as much as I understand. Cops engine would require serious rewrite to make it reliable. Personally I don't mind its current limitations features wise, just make it work for +2K resolution with +10 layers of multi-raster images is a obligatory requirement before going further. I'm afraid this is not something SESI is willing to undertake (and neither most of the users would appreciate if that would caused slow down of 3d development). 

[LaidlawFX]

On 8/4/2017 at 1:01 PM, symek said:

 I'm afraid this is not something SESI is willing to undertake (and neither most of the users would appreciate if that would caused slow down of 3d development). 

It's getting to the point where they either need to overhaul it or remove it. It's almost a pointless archaic tool now. I tell people to just use another application, it's not production capable anymore. It's kind of like chops too.

I would vote for a slow down of some 3D development if that were the choice, over removing the feature set. They updated pops after all. 

[jonmoore]

1 hour ago, LaidlawFX said:

It's getting to the point where they either need to overhaul it or remove it. It's almost a pointless archaic tool now. I tell people to just use another application, it's not production capable anymore. It's kind of like chops too.

I would vote for a slow down of some 3D development if that were the choice, over removing the feature set. They updated pops after all. 

I still find a place for Cops but not for compositing workflows, simply for live image assets referenced into to other contexts. I'm not overly bothered about it's lack of development but certainly wouldn't want to see it sacked altogether.

My bigger bugbear at the moment is the single threaded nature of Chops and the majority of Sops modeling nodes. I very much fall into the proceduralism over simulation camp and was very excited by the introduction of the new compile blocks in H16 but so few Sop mesh manipulation nodes that one would commonly use in for/each loops have been 'verbified' (compile friendly, in SESI parlance), it's hard to find many everyday use cases for compile blocks. And Chop's seems to have been the brainchild of the TouchDesigner folk so it's been in a frozen state for far too long. The new Chop's constraints functionality in H16 offered a ray of hope (all the new nodes are threaded) but I'd like to know the roadmap for other Sop's/Chop's nodes. It seems that the current priority is very much driven by the Sop nodes commonly used in partnership with Dops.

I seems to me that Houdini is increasingly being used outside of it's core VFX customer base and I think much of what attracts these new artist's to Houdini has little to do with Dop's. Sop Solver driven generative antics from the likes of Entagma seems to have inspired a new segment of artists to Houdin's procedural prowess, so a little focus from SESI to optimise the mesh manipulation tools that are so central to generative design workflows would most definitely find an appreciative audience.

[Guest tar]

The Topobuild 3.0 sneakpeek shows a fun cool new shelf tool, top right. 

 

 

[kev2]

And, from Magee's talk, two new contexts, LOPs and TOPs, next year. 

[Mandrake0]

3 hours ago, kev2 said:

And, from Magee's talk, two new contexts, LOPs and TOPs, next year. 

TOPs => Texture 

LOPs => ??? Light / Logic / Layer (guessing)

Touchdesigner has TOPs => https://www.derivative.ca/wiki099/index.php?title=Operator