normally i don’t post so much in blenderartists.org but with the risk
of sounding like i advertise something (which is not my way of doing
things), there is something that should be mentioned.
There’s a topic that looks like a contradiction to me, and
that is while a lot of users are complaining about poor/non existing
integration of Blender with external renderers, still there’s only
a little feedback.
I am refering to Blender2KT developed actively by Jean-Michel Soler (JMS),
that integrates Kerkythea on Blender. Perhaps you can give some
feedback to JMS post where there are only 15 answers so far,
something THAT IS VERY frustrating for a developer. http://www.kerkythea.net/phpBB2/viewtopic.php?t=1551
Well if you gave more updates on the progress here you’d get more feedback from the members who only post here (as well as it being very likely to be picked up by BlenderNation because I know they read these boards on a regular basis).
And as far as the integration goes, I’m all for it!!!
What i haven’t worked out yet is the ‘whole’ Blender2KT thing regarding the new render api for the purpose of ‘connecting’ to external renderers.
Blender2KT just looks like a fork of blender, my ignorance i’m sure but perhaps when such a project as Blender2KT can say 'hey look you can have any version of blender you want, even a CVS build, custom build and then use 'your ‘add on’ render engine like Podium in SketchUP that uses Kerkythea and render seamlessly (i know you can with Blender2KT) but i mean from within any blender build, and the same for any other render engine like Sunflow for example and that you can get blender artists juices flowing by telling them 'hey you will be able to render your AO pass or GI pass or what ever with Kerkythea your so and so pass with internal etc etc instead Blender2KT just looks like a fork thats going to need maintaining seperate to blender BF.
Probably talking crap but that’s why i haven’t bothered with Kerkythea, Yafray’s intergration feels more seamless even though it’s external and yet to intergrate with composite nodes.
Why is the Kerkythea interface stuff not in CVS as a build option like Yafray is or is it?
I’d prefer, i think to use Kerkythea over yafray as it appears to have far more to offer, but it’s also not open source and perhaps there’s a little resistance to building your workflow around an app that users could possibly end up having to pay for.
hi, i am unsure as to the purpose of ‘intergration’ when a decent script should do the job well. With the work being done to upgrade Blenders Open GL support, the internal renderer should become more than capable. And should make people wonder what they were doing with 30 hour renders in Indigo.
Already Blender has support for: Yafray, Yafaray, Indigo, Povray, Radiance, Kerkythea,
COS426 Raytracer, Nividia Gelato & I probably missed a few. How many raytracers does Blender need, 10, 20, 30. All i see that needs doing is to improve the current internal renderer, then worry about proper tight scripting to add support for 3 or 4 of the most popular best higher end renderers. Rather than call for intergration of this months render flavor. Each new renderer that comes out is soon followed by a script that is followed by calls for intergration.
As i have said before, if as much effort was put in to Blenders internal renderer than is put in to keeping up with each new renderer there would be little need to use any but a few choice external renderers via full scripting support.
I don’t have anything against external renderers, but with a few tweaks will Blender need them?
Thanks M.A.
You are right, it’s a branch of Blender. In principle, you are also true
that this needs more maintenance effort since the developer has to
update every time a new Blender release is available.
Then again, you 'd have to convince every other user that it is good
to have all renderers/exporters integrated on Blender CVS and
instantly available.
Most probably not!
But there is a point here; writing rendering code requires different
specialization from developers than making an integration. I don’t say
that the former is more difficult than the latter, i am just saying that you
can’t so easily “transfer” the effort from integration topic to
Blender internal renderer topic.
Just to clarify this:
" I don’t have anything against external renderers, but with a few tweaks will Blender need them?"
Anyone not agreeing that Blender strongly needs the external renderer doesn’t use Blender for serious work. I’m not doubting that the internal one will have the features and speed one day, but external renderer are used for special tasks. Some use a own shading language, some not. Some can be tightly integrated into a companies workflow and network structure and so on…
What we do need is a refacture of the rendering system, and a plugin structure like MAX. Or scripts, if you prefer.
Anyone who has used Brazil and VRay for MAX knows what you gain by external renderer.
For have massive feedback i think you have to make a proposal for a official project integration like yafray and not a “apartheid”,why not you try to integrate jms works in 2.5 release?
Well I visit kerkythea site (but not register) to see how its going for a while now and even jms seems to have stopped development on it on in recent months…or at least he stopped posting news of it at both places
I ask him before here in past twice how mature it is and if it is a good time to learn and he ignores me. I also ask if this will work with KT new version and I get no reply so…
I really dont know why he gets cold feet at this point though.
As I remember he announced and took this big mission on himself and was doing quite ok with it. As I understand there are still quite a few things to implement so I wait patiently in the background to see how it pans out.
If he gets it all functioning for 1.3.5 i’ll take a serious look at it. Ideally Ton gets his renderer api done soon and KT gets native support in blender. This is another factor that puts me off -at the moment. I seem to have builds for everything, everywhere and it gets confusing.
Because there are so many renderers and other stuff competing for my attention at the moment if I am to adopt it I absolutely need it to have good docs and for it to be working well
Also I need to see it does a better job than another renderer .Because I only have so much time I will need to give up following my existing interest in another in its place. So far I cant really distiguish a winning advantage to persuade me to do that.
Having come this far jms may as well finish it but really its up to him.
Can someone explain why an upgrade to OpenGL support will magically give blender internal the features in Indigo, Kerkythea. Will better OpenGL give GI, IBL, path tracing etc etc.
Already Blender has support for: Yafray, Yafaray, Indigo, Povray, Radiance, Kerkythea,
COS426 Raytracer, Nividia Gelato & I probably missed a few. How many raytracers does Blender need, 10, 20, 30. All i see that needs doing is to improve the current internal renderer, then worry about proper tight scripting to add support for 3 or 4 of the most popular best higher end renderers. Rather than call for intergration of this months render flavor. Each new renderer that comes out is soon followed by a script that is followed by calls for intergration.
What was the point of the render api changes? wasn’t part of it to make the link between blender and external renderers more seamless than a script. How do you script passes, node intergration, render layers etc etc.
As i have said before, if as much effort was put in to Blenders internal renderer than is put in to keeping up with each new renderer there would be little need to use any but a few choice external renderers via full scripting support.
I don’t have anything against external renderers, but with a few tweaks will Blender need them?
Thanks M.A.
if it only needs a few tweaks then surely someone would have done it by now even Ton himself while he was deep involved in the internal rebuild, the way your talking he could have fitted those few tweaks in surely.
sorry, that was not intended to start a war, i did say i have nothing against external renderers, i use yafray on occasion, but mainly rely on the internal renderer. I have tested a few others only as far as to see if they work compatibility wise.
It could be good to have a few choices in the renderer window.
Also the internal renderer could benifit from an upgrade to keep up with some newer features. ( i would hope that it would’nt need a complete rebuild, but it may) If it was decided to replace it then i would use the new one whatever it was.
IMO, Blender, being a SW used in so many areas, has also
users with quite different specializations and opinions.
From past posts, i saw many users wanting to have some good
integration with external renderers besides yafray. Perhaps i
understood that wrong…
I don’t want to speak on behalf of jms, but
i am pretty sure, that if a person like jms invests his time
without getting paid for developing something, he will do
that for the sake of blender community and the feedback he
will receive. If there’s no feedback, i can easily imagine that
this project will freeze.
You can use all the new features of the original Blender in Blender2KT. The last update was made the 11/01/2007 .
I ask him before here in past twice how mature it is and if it is a good time to learn and he ignores me. I also ask if this will work with KT new version and I get no reply so…
Twice, are you sure ? I found the firt one but not the second.
