Austrian emperor crown

Post your tests, experiments and unfinished renderings here.

Moderator: coordinators

Austrian emperor crown

Postby loramel » Wed Mar 04, 2009 5:22 pm

I am currently in the process of modelling the Rudolfskrone, the crown used by he austrian emperors and starting from the mid 1800 being the official crown of the austrian empire. Being an austrian this seem to be fitting ... ;)

See here for a picture taken from the treasury in vienna where it is exhibited.

rudolf_07.jpg


This scene is quite demanding as it has lots, lots LOTS of jewels. Allthough my focus in this stage is to get the modelling right, I woudl more than welcome any tips on setting up an efficient render.

Right now I am using Bidir with a depth of 50 ( otherwise the diamonds do not look right) and chromatic dispersion. Environment is infinite but enclosed in a sort of cylinder and the test render uses 3 simple planes as mesh lights.

The most impact on render time is using the cylinder as environment. Without it I get around 110000 S/sec but as soon as I use the cylinder I drop to currently 45000 S/sec on my AMD Phenom X4 @2.4Ghz, 64bit Linux, 8GB RAM.

The renders below is just a test with absolutely no focus on materials, besides maybe th pearls, where I played around a little bit. This is also the first stge in modelling, too exact and smoth and has to be 'humanized' to make it look more like handcrafted.

krone_unten_03.jpg

krone_02.jpg


Do you have any tips how to get the best out of lux for this scene ?
User avatar
loramel
 
Posts: 198
Joined: Mon Aug 25, 2008 3:37 am
Location: Austria, Tyrol

Re: Austrian emperor crown

Postby Abel » Thu Mar 05, 2009 2:15 am

loramel wrote:...on my AMD Phenom X4 @2.4Ghz, 64bit Linux, 8GB RAM.

Sorry for reacting off-topic, but:

1. That's already looking great! Can't wait to see this project progress.

2. I've been curious about the performance of LuxRender on the Phenom, would you mind running the benchmark? (http://www.luxrender.net/wiki/index.php ... Benchmarks)
User avatar
Abel
Developer
 
Posts: 1433
Joined: Sat Oct 20, 2007 8:13 am
Location: Helsinki, Finland

Re: Austrian emperor crown

Postby psychotron » Thu Mar 05, 2009 3:18 am

very nice, looking good so far.. will closely watch for progress :)
User avatar
psychotron
Developer
 
Posts: 836
Joined: Tue Jan 15, 2008 4:04 am
Location: Pleiades

Re: Austrian emperor crown

Postby Poncho » Thu Mar 05, 2009 4:01 am

What about using path?

I always try to avoid big boxes around the objects because they slow down. If you want a white/grey reflection from the surrounding box, you could try an environment map or change the color of the environment in blender.

looking nice so far :)

btw, there is a velvet (cloth material) on material database you could perhaps use. http://www.luxrender.net/lrmdb/en/material/view/54
Cheers!
User avatar
Poncho
Developer
 
Posts: 1127
Joined: Sun Nov 30, 2008 11:47 am
Location: Northwood

Re: Austrian emperor crown

Postby loramel » Thu Mar 05, 2009 4:52 am

Abel wrote:2. I've been curious about the performance of LuxRender on the Phenom, would you mind running the benchmark? (http://www.luxrender.net/wiki/index.php ... Benchmarks)


Will do it this evening.


Poncho wrote:What about using path?

I always try to avoid big boxes around the objects because they slow down. If you want a white/grey reflection from the surrounding box, you could try an environment map or change the color of the environment in blender.


I think BiDir shouldd be better suited for this type of scene, but I will give it a try and post the results as comparison.
Using an environment map is worth a try, but it will limit my possibilities in tuning the background.

Poncho wrote: ...btw, there is a velvet (cloth material) on material database you could perhaps use. http://www.luxrender.net/lrmdb/en/material/view/54


As it happens this material was posted by me, so yes I will use it :D
User avatar
loramel
 
Posts: 198
Joined: Mon Aug 25, 2008 3:37 am
Location: Austria, Tyrol

Re: Austrian emperor crown

Postby rusted » Thu Mar 05, 2009 7:05 am

looking quiet good.in the photo gold is looking more saturated.any info on lighting setup and materials?
rusted
 
Posts: 84
Joined: Thu Dec 06, 2007 8:54 am

Re: Austrian emperor crown

Postby loramel » Thu Mar 05, 2009 7:16 am

rusted wrote:looking quiet good.in the photo gold is looking more saturated.any info on lighting setup and materials?


Not at the moment, as there is no real work done on both of them. Once the modelling is finished (which I hope to be this weekend) I will concentrate on the various aspects of material setup (texturing, bump mapping, dirt maps etc).

The gold currently used is a plain metal material set to gold. The lighting are 3 simple planes as mesh emitters.

Any materials developed during this project will be uploaded to the material database, provided they use only features supported by the database.
User avatar
loramel
 
Posts: 198
Joined: Mon Aug 25, 2008 3:37 am
Location: Austria, Tyrol

Re: Austrian emperor crown

Postby jeanphi » Thu Mar 05, 2009 7:28 am

Hi,

Pure gold is a pretty unusual material, I've only seen once a sample of pure gold. You can have white gold with >90% pure gold, and standard yellow gold with only 75% pure gold.
Also, I think there is too much roughness, the exponent should be set higher.
In practice you also usually have a thin oily film of those that touched it or due to the cleaning products used, ... which tend to bend light towards surface normal when it hits the gold material and saturates the color (this is the same effect than on oiled or varnished, or even wet, wood).

Regarding materials, I think that your pearl material is pretty good looking given the pretty basic materials lux currently has for this kind of objects. You might try to use the thin film feature to create more realistic interference patterns.

Jeanphi
jeanphi
Developer
 
Posts: 6624
Joined: Mon Jan 14, 2008 7:21 am

Re: Austrian emperor crown

Postby loramel » Thu Mar 05, 2009 7:41 am

jeanphi wrote:Pure gold is a pretty unusual material, I've only seen once a sample of pure gold. You can have white gold with >90% pure gold, and standard yellow gold with only 75% pure gold.
Also, I think there is too much roughness, the exponent should be set higher.
In practice you also usually have a thin oily film of those that touched it or due to the cleaning products used, ... which tend to bend light towards surface normal when it hits the gold material and saturates the color (this is the same effect than on oiled or varnished, or even wet, wood).

Thanks jeanphi. The explanation about the oily film is valuable. Do you have any tips how to simulate this with the current lux materials.


jeanphi wrote:Hi,
I think that your pearl material is pretty good looking given the pretty basic materials lux currently has for this kind of objects. You might try to use the thin film feature to create more realistic interference patterns.


I already tried this, but decided to focus on materials later in the project. The problem with thin film is, that its only available with shinymetal, mirror and glass, which leads to doing some tricks with proper mixing of materials, where I only use the thin film property.
User avatar
loramel
 
Posts: 198
Joined: Mon Aug 25, 2008 3:37 am
Location: Austria, Tyrol

Re: Austrian emperor crown

Postby jeanphi » Thu Mar 05, 2009 8:31 am

loramel wrote:Thanks jeanphi. The explanation about the oily film is valuable. Do you have any tips how to simulate this with the current lux materials.

You'd have to use the thin film feature of materials supporting it or the glossy material, but currently none of those really fits. More general layered materials are on my todo list after v0.6

loramel wrote:I already tried this, but decided to focus on materials later in the project. The problem with thin film is, that its only available with shinymetal, mirror and glass, which leads to doing some tricks with proper mixing of materials, where I only use the thin film property.


Same as above :) However for pearls you might try having 2 concentric spheres, the outer one with a glass+thin film material and the inner one with a matte or glossy material. The radius difference should be around 1mm.

Jeanphi
jeanphi
Developer
 
Posts: 6624
Joined: Mon Jan 14, 2008 7:21 am

Next

Return to Works in Progress

Who is online

Users browsing this forum: No registered users and 0 guests