Most wanted features in a Render

Kerkythea is a new player in the renders arena. Altough the main code derives from PHOS it is a brand new program. It is free and also has a nice blender exporter enhanced by ideasman.

The developer (Giannis) is always looking for ways to enhance the render so we would like to hear which are the most wanted features for the ideal render for Blenders. Full integration with Blender would not be posible for the time being but Kerkythea will soon be available for Linux as well. So please let us know the most wanted features you want!!

Remember K is free!!


ok, let’s see how this polling goes!

I 'm really interested to see what will be the final
result, but i have to say that “volume lighting & effects”
is a big topic by itself.

So, perhaps another poll may be needed (for example,
finding out what are the most wanted “effects”) :wink:

btw, this is something that i have to say; please register
in K forum:
this is the least you can do to support the development.

Your comments and ideas are always welcomed.

you forgot vectorial motion blur and DOF.

What’s OpenFX? The only OpenFX I know of is the Open Source 3D package that never seemed to ever take off.

eh!? i don’t know either… perhaps OpenEXR? but this should
go along HDRI then…
hmmm, i think that a second poll is needed more detailed like:

  1. HDRI & OpenEXR support
  2. Depth of Field
  3. Motion Blur
  4. Volumetric Lighting
  5. Halo Effect
  6. Smoke Effect
  7. Fire Effect
  8. Physical Sky
  9. Metal Shader
  10. Plastic Shader
  11. Anisotropic Shader
  12. Spectral Lights
  13. Spectral Materials

and perhaps others… in this way we can focus on
a particular thing to implement and also the order of the results
will yield the sequence of implementation… what do you say?

Renderer look promising, but when do you plan to release linux version?

Oh gosh… another renderer. I think that’s number 5389 this season… it’s almost like The Bold And The Beautiful.

Anyhoo, don’t let yourself be distracted by the fact that you aren’t the only one’s around. All the power to you!

Actually Phos has been around for sometime now… Kerkythea is a rewrite of the original Phos code.
The important thing is not how many renders are out there but how many will survive, and how good they are. K is a very promising!!
Linux version will be available August 25 I belive.

not exactly :frowning:
next release of K will be again only in Windows around 25Aug,
and the linux version is planned at early September.
Linux version will have a gui from scratch, better than that of
Windows, but the first versions will miss all these cool dialogs
settings the rendering type, materials, etc… i will build all these
step by step.


ha-ha :smiley:
I agree with you. Most developers like to make a ray tracer
because of its linear difficulty. You may start tracing just spheres
with a couple of days implementation and then you are so taken
by the images that you go on with other features.

It’s fun and as long as there is time and interested users, this
can go on quite well. But feedback is necessary for small apps!

ha-ha :smiley:
I agree with you. Most developers like to make a ray tracer
because of its linear difficulty. You may start tracing just spheres
with a couple of days implementation and then you are so taken
by the images that you go on with other features.

It’s fun and as long as there is time and interested users, this
can go on quite well. But feedback is necessary for small apps!

I know the feeling. I was making a small chat application in PHP for a school website. It’s in its BETA stage right now, and that only because I got so absorbed by it that I just had to keep adding features upon features. Because of that, I’ve somewhat lost the overview; there are two bugs that were spawned at some point, and I have no clue where the hell they originate from. One is a logoff-bug that kicks you for no reason, which leads me to think that something is wrong with session handling; the other is a bug that I attribute to the lack of flock() on Win32, which corrupts data files when you are either doing multiple things at once or chatting with someone else (multiple threads accessing/editing the same file).

But I digress.

I wish you all the best of luck with a sort of YafRay (Yet Another Free RAYtracer), if I may call it that :wink: .

hum, not there… a per object (edge >25 degree, pointnormals of edge
angles with camera for 87+ degrees, etc.) edge rendering similar to
lightwa*e one, with tickness control (nothing too advanced, just 1 control
per object).


i wonder which freeware renderer most blender users use.
is it yafray? do they stick with the internal engine?

Blender’s internal is the most used. I believe that is because it’s far better integrated; if a feature exists in Blender, you can be sure that Blender’s internal supports it. For example: Yafray doesn’t support Blender’s Diffuse- or Specular-only lights. Most Yafray renders that I see are “explicit GI-renders” and I don’t know how to make anything else with it :expressionless:

Good luck with Kerkythea!

BTW, does “Spectral lights” mean Maxwell-style rendering? That would be awesome!

yes Spectral Ligth would bring Kerkythea closer to Maxwell… beside other optimization needed.
I am very much in favor of that myself, Giannis has the last said on that.

i thought that we need also spectral materials as well…
and making everything spectral costs in memory and time.

i really don’t know, but i see that a lot of people want this.
so, i guess this will be planned for some of the next releases,

I would really like an improved particle system. If there was some way that you can group seperate particle emiters (like let’s say, fire, smoke, sparks, and chunks of rocks from an explosion), group that into one paricle system, and you can just tell Blender, “I want an instance of this particle system to happen here at this point in time,” then I’ll be happy.

I’m working on some visually complex action scenes, and the thought of having to duplicate groups of overlapping particle emiters, and change each one of them to start and stop at a different time, is sounding really tedius and messy.

And maybe some other improvements on the particle system as well would be nice, but as it stands now, you can still make some pretty cool results. The idea of taking a bunch of particle emiters and turning that into one particle system, where you can say “instance here, here, and here, at this point of time” will make my life much easier in the future.

improved particle system

anisotropic specular

better AA

i think the rest of blender is already quite great.

so, i don’t get it completely,
what you want is the particle system of lights,
or the instantiation ability of such a system?

doesn’t Blender use any hierarchies for the objects in the scene,
so that you can copy-paste them anywhere you like?

Basically what I want is a way to easily duplicate complex particle systems (where you are using more than one emiter to make the right effect), have them grouped and simply tell Blender, “I want this particle system to play here, here, and here and play it back at these points in time.”

In other words I want particle systems in Blender to work the the same way that it works in video games, where you just say, “Explosion happens here”, and it will make an instance of your particle system there.

Otherwise, which is the way it works now, I have to select all the particle systems, duplicate them, and then change the start and stop times of every single one of them. For the scene that I’ll be working on in the future, we are talking about hundreds of bullets hitting walls and making chunks of rock and sparks fly everywhere. Multiply 4 (which is the number of emiters to make the blast), by 100, and you get the idea. Not fun. That’s hard to manage and will make a bit of a mess.

The copy and pasting isn’t as big of an issue as it is manually editing the start and stop number for every emitter in the effect.