ArtiCAD V21 Sun and Sky vs TurboCAD's

Just another set of tests.... I think the sun is not bright enough or the sky is to bright... To get similar Sun Brightness to TC the Sky rays overpower and add way to much noise and why we sometimes need such high render settings and the horrid horizontal sky ray lines!
It also shows clearly how much better the Spherical Sky option Is!
This also explains why the sky colour is so strong in ArtiCAD sometimes.... Can we please have the Sun Brightness separate from the Sky Brightness!

All Materials (apart from the glass) are 187RGB Grey with 2% Sheen

[You must be registered and logged in to see this link.]

Do you have Compusoft Winner as well to be able to compare?

I use Joe's 'Scottish sun' settings on my renders. 


(Disclaimer, the above post is a little over my head)

No, I don't have Winner but seeing some of the renders they have it set to Spherical as the ceilings get some sky rays and not just relying on GI to fill in the gaps...

Perhaps this explanation and a larger comparison explains it better.

1. Top image ArtiCAD V21. Notice the sky rays just go horizontal so it is making the window frame top opener rail to obvious with a band all the way round the room.
2. It creates way to much noise on the robe doors because the sky is to bright compared to the suns brightness. We should really have more control over the sky brightness and the suns....
3. Not using the RedSDK sky for the background. you can clearly see a nice blue sky tone in the TurboCAD glass. 
4. So as per V20 we have to add more lights to compensate for this bad skylight.

[You must be registered and logged in to see this link.]

It's frustrating with some room scenarios trying to get enough light into certain rooms and you have this fight between lit appearance and the GI. I have that horizontal, unrealistic light banding all the time.

Mingerz wrote:It's frustrating with some room scenarios trying to get enough light into certain rooms and you have this fight between lit appearance and the GI. I have that horizontal, unrealistic light banding all the time.

I did the above and below tests as I asked Peter today about how the option for the Spherical Sky was going... He told me Alek was working on it but was finding the sky was casting to much blue green colours onto the materials.... It's because the Sun is not bright enough to compensate for it....
Should hopefully have it set to Spherical by default as Interiors require it soon...

[You must be registered and logged in to see this link.]

You would have thought, that this would be a more important issue than tweaking a materials editor interface!

Mingerz wrote:You would have thought, that this would be a more important issue than tweaking a materials editor interface!

It is very important.... the trouble is we have been relying on using extra lighting and the default point light to fill in the missing "above window" light cast.... GI does try and do the job but not as good as the true light bounce from the ground that we get using the Spherical option.

Probably many of you are getting bored with this but to clarify further why a Spherical Sky should be default:

The following article is taken from this PDF - [You must be registered and logged in to see this link.]


6.3.4 The Ground “Glow”: An “Upside-Down” Sky

Although it might seem too self-evident to point out, we should remind ourselves that at the horizon the sky “meets” the ground. An actual ground plane of finite extent, say, a disc of radius r, will always fall short of an “infinite” horizon. For any given view toward the horizon, we can make the gap (a black void) between the edge of the ground and the sky appear smaller by using a larger r. However, we can never make them meet. Furthermore, there are good reasons not to introduce an actual ground plane of inordinately large size: the resolution of an ambient calculation will be dependent on the maximum dimension of the scene. To get around this problem, we use an upside-down sky to represent a luminous ground. 

The glowing ground behaves differently from a glowing sky. Although the same modifier is used for both, Radiance can distinguish between the two by testing the z component of any ray’s direction vector. Above the horizon, the sky-model brightness distribution is applied, but below the horizon, a constant brightness value is used.18 Note that as with the sky, the ground brightness is achromatic. In fact, a sharp-cutoff mixing function ensures a continuous transition from ground to sky radiance value that will be used for the ground brightness was determined by the gensky program. It is based on two factors: the sky’s (diffuse) horizontal irradiance and the “average ground reflectivity.” The horizontal irradiance is either supplied as an argument to gensky or evaluated from the zenith radiance. The “average ground reflectivity” may also be supplied as a gensky argument (-g refl); otherwise, a default value of 0.2 is used (as will be the case for us). The value 0.2 (or 20%) is a typical value for ground plane reflectance. We can check the gensky-supplied value for ground radiance very easily using Equation 6.8, since the ground is in effect a luminous “hemisphere” of constant brightness. Execute the gensky command as it appears in the scene file: % gensky -ang 45 0 -c -B 55.866 Recall that the last number of the gensky output for the CIE overcast sky is the ground radiance, which here is shown to be 3.56e+00 w/m2. The illuminance from a hemisphere source of this brightness is π(3.56 × 179) = 2001.9 lux, which is 20% (or 0.2) of the horizontal illuminance due to the sky. We shouldn’t worry too much about using an “upside-down” sky for the ground, but we should be aware of the practicalities. Although the ground radiance is based on the sky’s horizontal irradiance, putting something between the sky and the ground will not affect the brightness of either (Figure 6.5). In other words, no matter how built-up the model becomes, with nearby tall structures and so on, the ground radiance (where it is visible) will be the same as for an empty scene. By the same token, a single building is an obstruction. Therefore, all scenes should include a local ground plane that participates in the interreflection calculation. This will ensure that the ground plane brightness is a function of both the sky brightness and the local environment. 

[You must be registered and logged in to see this link.]


Figure 6.5 The luminous “envelope” describes luminance as a function of incident direction. 499-06ch.fm Page 359 Friday, 

Summary 


The scene we have constructed thus far is a seamless luminous envelope. The brightness of this envelope is based on a combination of a mathematical sky model and a ground plane reflectance model. We can specify the absolute brightness of this environment using physically meaningful quantities. Environments of this type will contain the rooms, office spaces, and so on, for which we wish to predict daylight quantities.