I need a few urban and environmental textures (concrete, stone, bark, dirt, ...) and I plan to create them myself. After I make the textures how can I place them on large objects in Blender to preserve high resolution but still reduce the repetitive effect in the resulting renders?

  • $\begingroup$ Actually this is not about Blender. To be short, I'd say: 1. I think checking adaptivesamples.com, among google, might help you. 3. Read about tiling textures in photo editors (for example, GIMP Map > Make seamless filter, as the easiest). 4. Don't agree, cloudy day without sun will give you shadows absence, that you'd really like. $\endgroup$
    – Mr Zak
    Commented Jun 11, 2015 at 12:49
  • $\begingroup$ This should be on photo.stackexchange.com And generally, asking more than one thing on one question is a bad idea if they are not closely related. $\endgroup$ Commented Jun 11, 2015 at 13:01
  • $\begingroup$ A tripod is always better (however I more often then not do not use one), if not using a tripod take many pictures, more then you think will be a little blurry. For focal length, as far away as you can get. When you are up close to any subject you will get barrel distortion. a longer focal length will reduce it greatly. $\endgroup$
    – David
    Commented Jun 11, 2015 at 13:01
  • 6
    $\begingroup$ I'm voting to close this question as off-topic because it is not about blender, but rather photographic techniques. $\endgroup$
    – David
    Commented Jun 11, 2015 at 13:03
  • $\begingroup$ I'm a bit disappointed that most of you think this is off-topic. It's a question about one specific step of 3D-animation with blender. Sure, it's not about the software itself, but I think this question can be answered best by someone who has either shot textures pictures himself or has experience working with image textures and knows which characteristics distinguish good textures from bad ones. @MrZak Thanks, I will check that out - good point about the clouds! @ David Thanks as well! So I should get away as far as possible and then zoom in to increase focal length? $\endgroup$
    – MoritzLost
    Commented Jun 11, 2015 at 13:26

3 Answers 3


It's in the creation and size of the texture that determines the quality of "tiling" over large objects. There are many ways to create a texture. You seem to be interested in photographic textures. The key to success here is to have flat even lighting and a camera that is perpendicular to the subject. If you want to optimize the result the camera should be stationary (IE a tripod). High resolution is also important in the original capture.

In the texture itself, for 3D rendering, the “Power of Two” is important. One potential reference http://www.katsbits.com/tutorials/textures/make-better-textures-correct-size-and-power-of-two.php there are many others. The Blender default is 1024 x 1024. I use 2048 x 2048. I believe larger file sizes (in the current environment) offer little advantage and can even slow the system down. There can always be exceptions.

*Depending on the pattern of the original image capture/texture some modification most likely will be necessary to facilitate tiling. Every image/ texture is different but here are some bullet points to follow in creating tillable textures (assuming the above).

  • Remove from the original image capture any “stand out” features (Like a large blemish on a slab of concrete or maybe a knot in tree bark).

  • Duplicate the image and scale the height -100% and the erase/delete part of the top or bottom. Flatten.

  • Repeat the previous for the width.

  • Use the new texture and make adjustments if necessary.

*The use/knowledge of image editors like Photoshop/Gimp is key here. But this is not the forum for me go into that level of detail. Hope this was of some help. Maybe I should do a tutorial?


I actually love this question, as it is leans more toward the artistic challenge of getting more from less. I'm adding this as an answer, because I personally have used, and still do use both "Dontwalk's" & "brasshat's" answers. However there are many times that I also add another technique to the mix in terms of additional textures.

In both the Cycles render engine & Post in the compositor, I tend to mix my material textures that I need to break up with either a hand painted crack pattern, a noise/musgrave/etc... texture or just some other high contrast image that helps break up what may tend to get repetitive in nature, and/or just add another image texture value based on the mix fac.

The best one that can give the variation though is actually using a displace modifier on a high res plane and setting the coordinate type to object, and taking an evenly lit render of it.

Then you can save off the variations and re-use them as color textures.


This is an issue in my own workflow, solved by doing two things. First, create more geometry, resulting in smaller faces in your mesh, that is, the same surface of the mesh will have more faces. With more faces in the same amount of UV, you can maintain the same level of detail in the texture on a given set of faces from a smaller part of the texture image.

If you're using a large image as a texture, you can map different parts of your mesh to different parts of the texture. Let's suppose that the texture is a granite surface, and that the hue across the image is not uniform due to the lighting when the image was made. Consider this image of a piece of black granite. You can unwrap part of your mesh into the darker portion of the image to the left, and another part of the mesh to the slightly lighter portion in the center.

When I make textures for my own use in Blender, I try to use natural light, not too bright, and avoiding strong light sources which would make the texture have too much variance from dark to light. I set the camera so that I can get a good image handheld.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .