Learning Houdini...

[SamHowell]

I've come across some tiny tuts for Houdini that might be of use for beginners. They are more like quick little tricks than full blown tutorials but there is some cool little mograph type things here to try out.



http://www.tokeru.com/mayawiki/index.php?title=Houdini

[MauricioPC]

Thanks Sam, the site looks good. Bookmarked.

[Bullit]

Well my first dwelling with testing Houdini confirmed what i was expecting. Anything but intuitive. It was just a simple fluid to fill a cup.
Well i could not found a way to make a dynamic rbd static.

[SamHowell]

Well my first dwelling with testing Houdini confirmed what i was expecting. Anything but intuitive. It was just a simple fluid to fill a cup.
Well i could not found a way to make a dynamic rbd static.

[Bullit]

Thanks Sam but my issue was just to convert a dynamic rigid body to a static. I am used to have that either in Softimage and 3dsMax explicit.

[pezetko]

Hi Bullit, that is not that hard when you dive into houdini a little bit deeper (yes it's deep black ocean, but it worths it imho).
I added example file with comments inside dopnet.

Best area for this type of question is on the sidefx.com forum, very helpful community.

Good luck and have a fun with Houdini

[NNois]

Hi Bullit, that is not that hard when you dive into houdini a little bit deeper ...

I confirm that too

[Bullit]

In kind of work environment here i can't push Houdini or even use it if is some kind of black box.

I cannot be an Houdini guy exclusively, so my brain can't be all for it and also cannot be only one guy that touches Houdini, so i'll have to convince someone else. Even not taking account the price difference, if Realflow is easier to understand and Houdini is not explicit in most things it doesn't stand a chance and i am only talking about fluids.

[Mathaeus]

Even not taking account the price difference, if Realflow is easier to understand and Houdini is not explicit in most things it doesn't stand a chance and i am only talking about fluids.

Well, Realflow is x times easier to understand, but I think you already know that. Should be a little bit harder with RealFlow to get the liquid related things (foam and so) to work in another app, than to launch the render directly from H, anyway.
I'm also not fan of their DOP configurations, perhaps because in my mind, actually any simulation deserves to exist only if there isn't any other way - if I really have to use sim, it has to be simple as possible... Since last summer, passed through many phases, including 'it's not that hard when you dive into houdini a little bit deeper', which is probably true, but constantly feel complete 'workflow' as simply wrong, idiotically overcomplcated. Also, what to do with app which can't animate in a way of Maya or SI, can't do direct modeling like any 3d app, still has viewport slower than Modo (not to even mention SI or Maya), has overcomplicated simulations to the sick level, some of them nowhere close to compete to Maya or SI (cloth).
So for now it's for simple scene assembly and rendering, some experiments with procedural modeling, and that's it...

[nodeway]

Also, what to do with app which can't animate in a way of Maya or SI, can't do direct modeling like any 3d app, still has viewport slower than Modo (not to even mention SI or Maya), has overcomplicated simulations to the sick level, some of them nowhere close to compete to Maya or SI (cloth).

Special Effects .

You are trying hard to make from Houdini something that it never wanted to be. Viewport in Houdini is very fast, but not when animating characters. Simulation part is aimed at shows that needs to create each time some new crazy shit no one done before, that's why it allows so much customization. And its all integrated very well together allowing you to pull data and modify it at each time step.

You can do direct modeling in Houdini. I consider it a lot better modeler than Maya (at least till 2015 version), but if you don't like nodes when modeling, you will hate it.

As for Realflow, did it get possibility to use nodes instead of scripting to get some more complicated effects? Last time I checked it you had to do ton of Python scripting for anything more than simple glass filling sims.

Lastly, Matheus, you really hate Houdini workflow and it shows. Move on with your life. I hate Maya, so I moved on and I don't look back. Do yourself a favour and don't look back on Houdini. Maya 2016 awaits, if that's what you want. For character animation it will be better choice. Don't waste another year on it.

EDIT:
But if you really insist and you will spend another year with H, just recently SESI fixed FBX export of rigs. I made simple test (which failed one month ago) but it looks that finally you can stay in Houdini if you would like to rig and animate for Unity/UE4. I will test it more in next couple days/weeks.

[Mathaeus]

Well it's definitively true that I don't like Houdini that much. If I do, I'll write much more of negativity. Also, at this point, I think H definitively needs a personality, anything else than usual set of phrases - this is actually worst, when everyone have a nice words, but no one anything exact, as only a few people uses it in real world.
If 3d app has something to show, doesn't matter is it positive or negative comment, both is actually positive, I think this isn't a secret. For example, Maya....
However, it's really is interesting, how even small negative comment or request for change, always causes reaction in 'move on' style. Sorry, but, won't go.

[nodeway]

Sorry, but, won't go.

Than stop yelling on a forum of a dead application about it. Have some problems ? Than Move your lazy f*cking ass to oDforce and SESI forum and ask there. Yelling will not help you. There are riggers in Houdini, proffesional, but if you don't ask them about your problems , and how they solved them, I will be always thinking about you like some sneaky little beyotch that is just yelling and yelling and yelling... without really doing anything useful.

Have you even asked even once about any rigging thing on any Houdini forum?

;)

Cheers.

BTW.
Only you guys think that Softimage was so good application. Believe it or not, in 2010 I played with Softimage and not even once, thru the time I played with it, never came to my ming idea that this application is superior to Houdini in any way. Actually I felt like I'm sitting on Maya, just with more child looking interface.

[MauricioPC]

Let's calm down people. There's no need to get offensive.

mantragora, if Matheus wants to complain about Houdini he can. You can also just ignore it.


As for Houdini, yeah, it's a great tool. But as an artist, I can do much more in Softimage than Houdini. And if Houdini was so much better, than why is SideFX listening to Softimage users on how to improve workflow, etc?

And I'll say you this ... Houdini got lucky because if Autodesk has actually worked on improving performance for 3ds Max, you would see a real competitor there. Sure ... you need to buy lots of plugins and spend more money, but the experience of using Max is more "artist driven" than Houdini (for FX).

In the end of the day, it just boils down to personal preference. I have artists that I look up to and they are Maya users.

[nodeway]

mantragora, if Matheus wants to complain about Houdini he can. You can also just ignore it.

Sorry, but, won't go. I checked his posts on oDforce and he never asked about rigging, so for me that means he's only complaining for complaining. If he is so good at rigging, I would love to see him asking questions and see them answered. This would actually help.

As for Houdini, yeah, it's a great tool. But as an artist, I can do much more in Softimage than Houdini. And if Houdini was so much better, than why is SideFX listening to Softimage users on how to improve workflow, etc?

See, there is a problem. I don't like those changes. I started from Houdini, Modo, Fusion, Cinema4d, Unity3d and then I came to Autodesk world. I didn't liked even one Autodesk app after all those I started from.

And now we got those new selection in Houdini... and they suck compared to more 'artistically' driven application like Modo/Cinema4D. At this point I don't remember how selection in Softimage worked, but if selections in Houdini are really inspired by Softimage ones, than I don't think that Houdini will become good modeler anytime soon.

And I'll say you this ... Houdini got lucky because if Autodesk has actually worked on improving performance for 3ds Max, you would see a real competitor there. Sure ... you need to buy lots of plugins and spend more money, but the experience of using Max is more "artist driven" than Houdini (for FX).

And I'll say you this... they had this 'artistically' driven application called Softimage, and it's dead. Because it sucked as a hardcore TD tool, and was not good enough as a artist tool.

BTW. More than 20 years of luck of Houdini. In the meantime this 'artistically' driven, with better performance application called Sotimage didn't become real competitor.

In the end of the day, it just boils down to personal preference. I have artists that I look up to and they are Maya users.

I have people who came from Maya and they never locked back after been touched by Houdini.

[Hirazi Blue]

Can we please try to get this thread back on track without the passive aggressive turf war?

edit: to elaborate slightly, it's not so much the passive aggressive turf war I have problems with,
it's the derailment of a thread innocently named "Learning Houdini..."

[nodeway]

Can we please try to get this thread back on track without the passive aggressive turf war?

edit: to elaborate slightly, it's not so much the passive aggressive turf war I have problems with,
it's the derailment of a thread innocently named "Learning Houdini..."

Ok, I'm moving on. Just one last word, about Maya this time ;)

In 2012 (or 2013?) When they introduced .NET extension for it, I thought that after so many years of trying to like this app this will be final nail that will do it for me. I could finally use the programming language I like the most (C#) inside it. So I started digging. And I digged to situation where I wanted to get selection of polygons and delete it. Not one polygon, but many. It turns out that to accomplish this with good performance I had to use MEL inside C# code to remove them. There was no possibility to delete all selected polygons at once from C code. You could delete them one by one, and update point numbers after you deleted next one, but when you done it, it was slow. Put a MEL inside C code and in split of a second you can delete them. I was really surprised that you cannot do this operation fast with C code. Her is a topic => http://forums.cgsociety.org/showthread. ... ge=1&pp=15.

And it's only one of the many reasons why I will never go to any Autodesk application. It really doesn't matter if they fixed it or not.

So think about that guys. This is something that awaits you in their world. Glued s*it barely holding together.