Octane 1.5 features revealed!

[gabrielefx]

will ies lights perform better than the 1.18?

[Karba]

will ies lights perform better than the 1.18?

I expect - yes. Did try yet.

[MOSFET]

Looks great Karba

[bepeg4d]

wow this great, thanks Karba for sharing, night shots should be no more a nightmare in the near future
ciao beppe

[UnCommonGrafx]

Though I don't do games just yet, I have to add Kudos to Otoy for leveraging their technology, particularly in the directions they've shared.
Realtime rendering, with the beauty of Octane, is to be desired. Speed, baby, Speed!!

I believe that as their knowledge of the GPU-renderer matures, we will see the benefits more quickly. We are at the infancy of this technology and theory.

Faster, cleaner, smoother is what I want. As well as all the other accoutrement I can get with what would be termed a modern day renderer.

I can get complicated images in 41 mins vs 3 hrs for my regular renderer. I laugh that now I think the 41 mins is too long.
Looking forward to the speed updates. Loong over-due, I tell ya.

[Refracty]

Many great features are coming. Visibility options and speed improvements with tons of emitters are useful.
Hope to see motion blur soon after the 1.5 release.

[Elvissuperstar007]

Many great features are coming. Visibility options and speed improvements with tons of emitters are useful.
Hope to see motion blur soon after the 1.5 release.

Hope to see motion blur soon after the 1.5 release.[/quote] +1000000)

[ROUBAL]

One question that I have asked many times without getting any answer is :

Why not doing first the bunch of simple little things that would bring in a short time many usefull features to most users ?

As an example : the negative scaling in transform node !

There are most probably many little useful features in the users wishes list that could be implemented in few weeks or even less !

[oguzbir]

One question that I have asked many times without getting any answer is :

Why not doing first the bunch of simple little things that would bring in a short time many usefull features to most users ?
As an example : the negative scaling in transform node !

I totally agree with you ROUBAL.

First,
I really want to congradulate you guys for the 1.5 roadmap. Things looking great. Well done.
1.20 is a really solid version. I love octane. It's really the best. It beats the hell out of the other engines which others are so sluggish and problematic and slow. I earn money using octane as a renderer. So please read my lines not as any angry fella. I'm never angry but I really like Octane to perform well.

Here is my few thoughts about the current situation on my side.
I really would love to know why MB is realy hard on Octane, if it's no secret.
I mean if it's hard to implement I get it. But please let us know.
I just feel like it's like a candy bar, you keep it until you're ready and boom..

Below are my thought on the new features which might be done without waiting for the next
release. These are 3DsMAX centric but applies to all package plugins.
I'm never been a +1 guy. So please read...

FWIW Not a programmer myself but I think an override material can be implemented more easyly than MB. It's really a pain when you're producing images for commercials or ads.

For a simplest scene. When I try to render passes. I tend to save the file for each pass just to be safe when I back and forth to photoshop. When there is a feed back from the customer. I have to redo the whole process, just to have the render pass.
Wouldn't be aweasome to have a preset manager. Where we could save presets internally and without loosing data..And a Override material.

Another thing that bothers me most is the parametes for Texture Environment.
Wouldn't be nice to have just a single material where you can choose your hdri and change gamma and rotation values at the same time underthe same simple UI. It's really bugging me I must say.

If there is no MB at the moment maybe A button would be nice to convert the materials to Vray or scanline Mray or even iray, so that we can render out MB sections in other engines. Believe me when I say this, that would realy help. It's really pain to redo the same materials in vray just to have a MB. That's simply because of the different material layout of octane.
If that "Convert Material button" can be reverted to convert materials to vRay or other engines that would be aweasome.

Maybe it's really time to clean up the UI and the workflow as well. Maybe add couple of things that may really give speed to the current and loyal users.

The Matter material properties is lacking its' use.. It cathes shadows, reflections. No solid workflow there. Shadow cathing is problematic. That really should be solved.

Being unable to transfer material properties between standalone and MAX (or any other plugin) is a must. I generally tend to use standalone as it's so much faster to use and explore.

Last year Autodesk started to release SAT (Small Annoying Things) updates for 2013 product line. Those were simple but a number of annoying problems we faced while using.

I really congradulate you about the API through scripting thing. That's really going to fill a gap. At the moment it's really pain when you use lots of lights and being unable to control them via lightlister. I couldn't achieve to do that in script, but I tried. It's hard to get all the control code from the maxscript listener.

I mean there are still undo problems through out the UI. You can undo material properties but not render properties. It's somewhat annoying when you want to go through parameters via using TAB
But it goes reverse and sometimes it jumps.

These tiny little things might really boost and give the solid feeling o this amazing software.

Best,

[abstrax]

Finally some sensibility came back to this thread. To say thanks, I will try to answer two of the questions (need to get some work done...):

Why not doing first the bunch of simple little things that would bring in a short time many usefull features to most users ?

As an example : the negative scaling in transform node !

There are most probably many little useful features in the users wishes list that could be implemented in few weeks or even less !

It's economically not feasibly to just spend time on the "simple" things, because "small" changes don't sell a product and people forget about them as soon as they get released. Just have a look at versions 1.12 - 1.18 which fixed and improved heaps of small things, but nobody really cared.

-> It's best to work on "big" improvements and along the way add/fix smaller things. As a consequence, not every "small" feature is implemented as soon as it could have been done, if we would work only "small" improvements.

I really would love to know why MB is realy hard on Octane, if it's no secret.
I mean if it's hard to implement I get it. But please let us know.
I just feel like it's like a candy bar, you keep it until you're ready and boom..

Well, it's actually quite simple:

Where in Octane do you find animated data? Nowhere - beside the turn table animation and daylight animation tools. And these tools effectively don't create animated data either, but modify the camera/daylight environment, render a frame, modify it again, render and so on and in the end, set the nodes back to what they were. I.e. there is currently no way to store/manage animated data in Octane.

-> To allow animated data (and solve lots of other problems), we are currently working on two things: Refactoring the node system and providing a way to get animated data into Octane via Alembic and the API. And believe me both tasks are big. It doesn't have to do anything with CUDA or fancy algorithms. It's just good old boring software development.

Then there is the actual rendering: The old acceleration structure of Octane was not suitable for motion blur. With 1.11 we introduced a new one, which is more suited for motion blur. It didn't support motion blur yet, but adding motion blur to it is fairly straight-forward. Of course, writing a new acceleration structure wasn't done in a day. It actually was developed over several months leading up to the release in version 1.11. And similar to that development we have been and are working on improvements and features that will be required on future features.

Beside of the stuff above we have to maintain the API (we just recently ported it to different platforms), have to keep compatibility with OCS/OCM files, the LiveDB, have to work towards cloud rendering, fix bugs and keep the whole thing as stable as possible and do lots of other stuff...

And as a side note about the previous pages in this thread: This whole discussion about future features confirmed my opinion that we shouldn't release any future road maps at all...

Cheers,
Marcus