By rogersalph
#361874
Hi Guys,
I am trying to achieve the result of an exterior as I have produced in vray in maxwell.
http://www.rogersarchitecture.com/exten ... ting-house

The trouble is that each time I try to do that, I have noticed that the exterior shots in particular are always blurred.
http://minorityexpress.wordpress.com/20 ... ll-render/


I have also included a photomontage work which I have done with another house and I am having the same result with a different set of materials.
http://minorityexpress.wordpress.com/20 ... h-maxwell/


Is there anyway around this? I have had renderings go on for 2 days and they are still blur. Any ideas?
By JDHill
#361879
I'm not clear on what you mean by saying the image is blurred. The second image is a bit small to see exactly what's going on, but it just looks very blown out, and with very strong vignetting. The environment seems to be set up oddly, making it difficult to judge the material setup. Overall, I would probably recommend backing up, estimating the conditions (camera, mainly) in the image with which you are compositing, and then proceeding from there.

Regarding the last image & comment, I'm not sure what you're referring to; could you please elaborate on what you were expecting, vs what you have obtained?
User avatar
By dariolanza
#361890
Hello rogersalph,

If by "blurred" you mean the images are noisy, then it indicates they need a bit more render time (higher Sampling Level).

If you mean that they look out-of-focus, then you may need to adjust the camera target distance and diaphragm aperture (just like you do in a regular real camera) to ensure your object is completely in focus.

Otherwise, could you elaborate more in depth (and with more clear visual examples) what did you wanted to get and what are you getting? This will help us to faster isolate what is happening and how to solve it out. As JDHill is commenting, we can't see the problem you are describing in the pictures you uploaded.

Greetings

Dario Lanza
By rogersalph
#361915
Hi,
Dario -
I am referring to the fact that most of my exterior renderings appear out-of-focus. How do you adjust camera target and diaphragm aperture in maxwell and rhino? I know that you can do that in sketchup under scene manager>camera>focus and exposure> Auto(Spot), Auto(Average), Auto(Average+DOF). I usually toggle between Auto(Spot) and Auto(Average). I based this off the sketchup plugin manual. Is that what you are referencing to? I have realized that the answer to my question is to render the image at a much higher resolution as well as adjusting the sharpness to a value of 60. This has solved my query. I have posted my renderings here:
http://minorityexpress.wordpress.com/20 ... esolution/
http://minorityexpress.wordpress.com/20 ... n-maxwell/

The next thing that I need to work on would be to blend the environment of the rendering to fit in its surroundings. I often do that as a post production in photoshop using saturation. Would you guys often tweak the environment settings to attain/achieve the desired environment? That would have compromised with the quality of reflectance on the glass.


The other issue that I have had is that I find that I often would like to retain the perspective of an image. The reason why is so that I could montage the image with the same focal length into the photo that I have taken. Unfortunately, Sketchup's native camera does not cater to the focal length but works with a FOV instead. My solution to that would be to use a converter online to work out the FOV's :
http://kmp.bdimitrov.de/technology/fov.html

Of course cameras comes with different crop factors and they have to be accounted to the lenses to work out the actual Focal Length that it was actually taken. I can elaborate more about this but I think you get where I am getting to. The question is, how do I blow up the resolution of an image without changing the perspective of the image.

JD -
I think that I might have answered by question subconsciously. The image has to be rendered at a higher resolution - not necessarily longer SL's/time. The question that I should have asked is:
How do I render a close up shot of an image without necessarily changing the perspective of an image?

http://minorityexpress.wordpress.com/20 ... n-maxwell/

Say the image above for instance. If I were to stand say an additional 50 metres further away from the building, I would be seeing more context/site information. Therefore the subject (which is the house design) would appear smaller. If that is done in the renderings. I will have an issue putting it onto the site as the image would be really low resolution and therefore would appear out of place in say a high definition photograph. I will post an example later if my question is not clear enough.
By rogersalph
#361918
Hi Guys,
I finally worked out a way to simplify my question so that I don't keep you guys guessing.

http://minorityexpress.wordpress.com/20 ... esolution/

Say I've rendered two images set with the same focal length (20mm). I did one at 4000px x 3133px and the other at 5028px x 1974px. They are both rendered at a pixel aspect of 1. Does the filmback have any effect on the resolution of the image? I read in a couple of forums which states that the images needs to be rendered at 180 per inch (300 is preferable). And that A4 landscape should typically be rendered at 3508 x 2480 pixels, A3 landscape should be rendered at 4962 x 3508 pixels. As I normally work in metric scale.

1. Is there a formula which you can refer me to so that the image does not look stretched and blurred when it is being printed or merged with another high resolution photograph? How do I render to a resolution of 300 pixel/inch or 150 pixel/inch on say a custom format of 800mm x 1600mm, or a typical A2(841 x 594mm)?
2. What is the relationship between roll angle and target distance? I read somewhere in the manual which refers to the two as integral to establish Depth of Field as the background would be blurred. Is there a way to set that the focus to infinite? Like in real world photography, it is possible to set the focus to infinite so that everything appears clear and none of the image appear out-of-focus?
3. I have noticed that there are a few camera presets sensors that are available. EX: Nikon, Canon, Fujifilm, Kodak, etc. Take for instance the Canon camera which incorporates a point and shoot 2x cropped factor (IXUS 990IS, Powershot G10) , 1.6 cropped factor cameras (500D), mid range cropped factor cameras (50D,60D) , as well as a full frame cameras (5D, 1D). Does the crop factor affect the Focal Length as you would get in real life? For instance, a 10mm FL used on a EOS 500D would in turn be 10mm x 1.6 = FL 16mm / FOV 107.026. Or are they just sensors which varies in terms of its sensitivity to light? It would be great if more explanation about this could be elaborated.
4. Why does the devignetting affect cause the entire scene to become overexposed? I could troubleshoot this by increasing the shutter value or toggling around the apertures. Am I on the right track with this or is there something wrong with the way I set up my scenes?
5. I am still struggling with understanding the function of (sharpness) under tone mapping? What does this value do which affects the rendering?
6. And of course, there is the whole category under simulens which I completely have no idea what it does. Ex: Aperture Map, Obstacle Map, Diffraction, Frequency and Scattering.


I really appreciate the help which you guys have provided. They are both very prompt and direct to the answers which I am seeking for. The reason why I am asking so many questions is so that I can speed up my learning curve as I need the produce heaps of renderings in the coming months as jobs are flowing into our 9 month old firm. :D
By JDHill
#361919
rogersalph wrote:Hi,
Dario -
I am referring to the fact that most of my exterior renderings appear out-of-focus. How do you adjust camera target and diaphragm aperture in maxwell and rhino? I know that you can do that in sketchup under scene manager>camera>focus and exposure> Auto(Spot), Auto(Average), Auto(Average+DOF). I usually toggle between Auto(Spot) and Auto(Average). I based this off the sketchup plugin manual. Is that what you are referencing to? I have realized that the answer to my question is to render the image at a much higher resolution as well as adjusting the sharpness to a value of 60. This has solved my query. I have posted my renderings here:
http://minorityexpress.wordpress.com/20 ... esolution/
http://minorityexpress.wordpress.com/20 ... n-maxwell/

The next thing that I need to work on would be to blend the environment of the rendering to fit in its surroundings. I often do that as a post production in photoshop using saturation. Would you guys often tweak the environment settings to attain/achieve the desired environment? That would have compromised with the quality of reflectance on the glass.


The other issue that I have had is that I find that I often would like to retain the perspective of an image. The reason why is so that I could montage the image with the same focal length into the photo that I have taken. Unfortunately, Sketchup's native camera does not cater to the focal length but works with a FOV instead. My solution to that would be to use a converter online to work out the FOV's :
http://kmp.bdimitrov.de/technology/fov.html

Of course cameras comes with different crop factors and they have to be accounted to the lenses to work out the actual Focal Length that it was actually taken. I can elaborate more about this but I think you get where I am getting to. The question is, how do I blow up the resolution of an image without changing the perspective of the image.

JD -
I think that I might have answered by question subconsciously. The image has to be rendered at a higher resolution - not necessarily longer SL's/time. The question that I should have asked is:
How do I render a close up shot of an image without necessarily changing the perspective of an image?

http://minorityexpress.wordpress.com/20 ... n-maxwell/

Say the image above for instance. If I were to stand say an additional 50 metres further away from the building, I would be seeing more context/site information. Therefore the subject (which is the house design) would appear smaller. If that is done in the renderings. I will have an issue putting it onto the site as the image would be really low resolution and therefore would appear out of place in say a high definition photograph. I will post an example later if my question is not clear enough.
The Rhino plugin does not have a realtime auto-focus mode like the SketchUp one. However, in the plugin's Camera toolbar, you will find the commands: Auto-focus to Objects, and Pick Camera Focal Distance. The first prompts for an object selection, which is used to set the camera's fStop and focal distance; the second allows for setting focal distance by picking a point in the viewport.

Setting up the environment to blend with a photo is a pretty subjective process. Depending which type of environment you are using (i.e. Scene Manager > Environment > Type), you could use any number of strategies for doing that. When using Image Based environment, you can set each channel to use an HDRI texture, or the sky type chosen in the Sky Type dropdown. When using HDRI images, the only control you really have is the intensity of the texture, so if, for example, you need more saturation, you would need to edit the image itself in an external image-editing application. As this is not a plugin-related question, though, you might want to post about it in the general Maxwell Render section of the forum, to see what strategies other people use.

Regarding modifying resolution without altering perspective, the Maxwell camera works like a real one, so the way of doing what you want involves changing (reducing, in this case) the film size, without zooming the camera, physically. A tool that will help you do this is the Camera Heads-up Display, which is also found in the Camera toolbar; just try enabling that, and then reducing the film size. Another way is to use region-render to render only a portion of the total output area. Also, in case it helps, you can set the camera's focal length in SketchUp, but you need to use Ruby -- open the Ruby Console and type Sketchup.active_model.active_view.camera.focal_length=35, and hit Enter.

Sharpness, by the way, which is used to modify the anti-aliasing filter, is not the tool to be using to address any of these questions.
By JDHill
#361921
Does the filmback have any effect on the resolution of the image? I read in a couple of forums which states that the images needs to be rendered at 180 per inch (300 is preferable). And that A4 landscape should typically be rendered at 3508 x 2480 pixels, A3 landscape should be rendered at 4962 x 3508 pixels. As I normally work in metric scale.
Besides for aspect ratio, filmback (film size) does not affect resolution. Do not be confused by dpi; if you need an image 3508 x 2480, then that is what you need -- its "size" in inches or millimeters depend on how many pixels per inch/mm you print.
1. Is there a formula which you can refer me to so that the image does not look stretched and blurred when it is being printed or merged with another high resolution photograph? How do I render to a resolution of 300 pixel/inch or 150 pixel/inch on say a custom format of 800mm x 1600mm, or a typical A2(841 x 594mm)?
As above, do not get confused about dpi. If you are using 300dpi and you need an image 5" wide, you need (5 inches X 300 pixels per inch) = 1500 px output resolution.
2. What is the relationship between roll angle and target distance? I read somewhere in the manual which refers to the two as integral to establish Depth of Field as the background would be blurred. Is there a way to set that the focus to infinite? Like in real world photography, it is possible to set the focus to infinite so that everything appears clear and none of the image appear out-of-focus?
Focus of a camera is dependent upon both upon its focal point, and its fStop. A (numerically) low fStop means a large aperture; a large aperture reduces the camera's depth of field, which is the distance between the nearest and furthest points from the camera at which objects will appear in focus. The reason for this is that a wider aperture allows light to enter the camera from a wider range of angles. Infinite focus does not mean that all objects are in focus; there is no way around the physical limitations of camera lenses, and infinite focus just means that given your focal length and aperture, objects you are interesting in capturing are coming into the camera along basically parallel rays. The upshot is, to increase the depth of field, 1) shoot with a smaller aperture (higher f-number), and 2) shoot with a shorter lens. If you are trying to match an existing photo, you cannot use the second, so you need to play with the fStop. More information on depth of field can be found here.
3. I have noticed that there are a few camera presets sensors that are available. EX: Nikon, Canon, Fujifilm, Kodak, etc. Take for instance the Canon camera which incorporates a point and shoot 2x cropped factor (IXUS 990IS, Powershot G10) , 1.6 cropped factor cameras (500D), mid range cropped factor cameras (50D,60D) , as well as a full frame cameras (5D, 1D). Does the crop factor affect the Focal Length as you would get in real life? For instance, a 10mm FL used on a EOS 500D would in turn be 10mm x 1.6 = FL 16mm / FOV 107.026. Or are they just sensors which varies in terms of its sensitivity to light? It would be great if more explanation about this could be elaborated.
The camera sensor presets, which do not appear in the plugin, are simply presets for camera film sizes. They literally just set the film height and width.
4. Why does the devignetting affect cause the entire scene to become overexposed? I could troubleshoot this by increasing the shutter value or toggling around the apertures. Am I on the right track with this or is there something wrong with the way I set up my scenes?
The Devignetting feature just provides a way of artificially removing the natural vignetting that occurs as a function of camera focal length and aperture size. You should not use it unless you specifically want to get an image that lacks this aspect of the camera simulation. People generally use this when they want to render something like a 360° panorama, made up of several separately-rendered images. As you can imagine, vignetting can be a problem in such cases.
5. I am still struggling with understanding the function of (sharpness) under tone mapping? What does this value do which affects the rendering?
I realize you posted this before I posted my last answer, but just to reiterate, the Sharpness value only affects the anti-aliasing function. Some reasons for using it are given here.
6. And of course, there is the whole category under simulens which I completely have no idea what it does. Ex: Aperture Map, Obstacle Map, Diffraction, Frequency and Scattering.
Simulens is a set of post-rendering processes that simulate camera lens effects. For general info on these, please see here.

Lastly, as I briefly mentioned in my previous post, this forum is intended for questions specific to the Rhino plugin (bugs, questions on plugin-specific features, plugin-specific feature requests, etc), so it is not really the place to get answers on general Maxwell questions. This is both to keep the focus here on the Rhino plugin, and also because it is better if these types of questions and answers are out in the main forum, so that more people will be apt to find them later.
By rogersalph
#361963
Hi JD,
Thanks for the reply, I would be looking to go through your thorough explanation this weekend! I will redirect my future post to the main Maxwell Forums in the future. I might have misunderstood the different subcategories as these are one of my few posts in the forums.


Cheers,
Neo
Help with swimming pool water

Hi Andreas " I would say the above "fake[…]

render engines and Maxwell

Other rendering engines are evolving day by day, m[…]