Future of Twilight Render Engine

[eph]

Hi,

How do you guys see the future of the TL engine evolve?
As I believe, some evolution of the Thea render engine will make its way into the Kerkythea engine..?
Then, will Twilight follow this engine upgrade as well? Is this easy to accomplish (in other words, will we have to wait long)?
And lastly, will TL help in the engine development, or is your task/goal focused on UI and SU integration..?

Trying to understand the more and more complicated picture here
Looks really nice though, will try out this afternoon

[archigrafix]

suprised nobody of twilight answered this question, cause its a very interesting one... Fletch, chris are you there???

[Frederik]

suprised nobody of twilight answered this question, cause its a very interesting one... Fletch, chris are you there???

Come on... The question was posted 10:40 am CET...
Chris is based in US, only Fletch would be able to comment at this time of the day, since he's now living in Europe...

How do you guys see the future of the TL engine evolve?

It is indeed an interesting question...

As I believe, some evolution of the Thea render engine will make its way into the Kerkythea engine..?

I'd say that it's going to be really difficult, since Thea is build from scratch...
Although Thea is based on A LOT of knowledgement gained on developing KT, they're two seperate applications and can't be mixed...
The Thea Render will be able to import KT models and convert KT materials, but Thea models and materials can't be used in KT...

From previous discussions (which I can't disclose here...) I'm aware of some future plans and improvements of KT, but they have absolutely no direct connection to the Thea Render...!

Then, will Twilight follow this engine upgrade as well?

Yes - most likely... (sorry - can't be more specific than that...)

Is this easy to accomplish (in other words, will we have to wait long)?

First of all... All development resources are currently aimed on the Thea Render...
Development of KT hasn't stopped, but let's acknowledge that the pace in developing KT has dropped significantly, but I'm sure it will catch up after the first release of Thea...

I'd also like to see some more development of Twilight...
I.e. we're still missing an option to use the instancing brush and Network rendering in TWR...
Although SU is very poor at handling high poly components, there's room for adding proxies/"dummies", which in the render process will get replaced by high poly instanced components in TWR...

And lastly, will TL help in the engine development, or is your task/goal focused on UI and SU integration..?

Currently Chris and Fletch (and the rest of us) have been focussing on the UI and SU integration...
I don't see any immediate change in this until we see the development of KT catching up again, but I'm sure we will have ideas, which will improve the development of the KT engine...

[eph]

As I believe, some evolution of the Thea render engine will make its way into the Kerkythea engine..?

I'd say that it's going to be really difficult, since Thea is build from scratch...
Although Thea is based on A LOT of knowledgement gained on developing KT, they're two seperate applications and can't be mixed...
The Thea Render will be able to import KT models and convert KT materials, but Thea models and materials can't be used in KT...

From previous discussions (which I can't disclose here...) I'm aware of some future plans and improvements of KT, but they have absolutely no direct connection to the Thea Render...!

That was based on Giannis' post Announcing Thea Render:

And of course, there will be some long-promised features, that now make their appearance in Thea Render, but will also become available in a next Kerkythea edition.

I thought he meant backporting some of those features and improvements.

I don't see any immediate change in this until we see the development of KT catching up again, but I'm sure we will have ideas, which will improve the development of the KT engine...

That sounds ok.. still, I'll wait for the public beta of Thea before deciding where to put my money, I guess...

[Chris]

Hehe, it's just now 8 AM here...

Anyway, Frederik's response is pretty comprehensive. As he said, some Thea stuff may, independently, be ported back to KT but I can honestly say I have no idea what or when. However, if it is then yes, we intend to make full use of it!! The time it takes to incorporate changes is usually small but that just depends on the change involved.

As far as contributing to the KT core, we make suggestions when it's appropriate, but development continues to be done only by KT's original creator.

As far as developing things for Twilight, there's a lot we have planned that isn't reliant on the render core. We've never published a schedule or list of planned features because when and in what order those things get done, and even if the can be done, is rarely well known. Plus, everyone likes a surprise... right?

[archigrafix]

sory chris and fletch...purpose wasn't to wake you up...of course LOL. I just wanted to get your feeling on the subject, and I could wait for some days...but you guys are so reactive its really a pleasure!
Yes I feel there's still many inetresting things to do improving TWL integration: Instances, dummy components and most of all network rendering... that said its already a great product and great support from you and all users
thanx

[Fletch]

I can say that Twilight has already compensated/added some things that were wanted/needed in KT. there are many things actually

I'd say the future for Twilight is shining BRIGHT!

You must keep in mind that Twilight's main goal is to be a high-quality and easy-to-learn render engine for Architects designing in SketchUp.

It is extremely flexible and powerful, and can therefore do much more than this, technical renders such as splitting light thru crystals, etc... but most Architects don't really need this sort of thing.

Net Rendering is of course a high priority and known need. Have no fear.

If you peruse the news section and see the features/fixes since our first release until now you will see the impressive pace of development, so I think you can be confident that as long as there's SketchUp, and us, there will be Twilight.

[Pilou]

I have seen in the new Blender version 2.5 that the internal renderer has won from 3 to 10 speed factor
Does this possible in next versions of Twillight?