I have no words. I am so thankful that you decided to share your knowledge with us. Hopefully your sub-count can soar, cause you definitely deserve it! When tímes get better i will probably start donating to your patreon.
Amazing video man, thanks for all your explanation, and detailed process. I have a question, when i use this texture for a cylinder for example, the top and bottom part wont have the "brushed" / "stretched scale" look. How can i fix that?
Your formula makes the best brushed metal texture I've seen but for some reason I can't seem to render it without noise. I also tried other node formulas I learned on RU-vid but all have so much noise. When I use Open Image Denoise it ruins the texture, making it quite blurry and fake looking. Any suggestions on how to get this to render well? Thanks.
Thank you for all of these great material tutorials! I've learned so much about material nodes from your excellent explanations! This one didn't turn out for me right using Blender 4.0 unfortunately, so I made a couple adjustments. The brushed scratches were a little too thick so ended up with a scale of 14.7 on the Noise Texture to thin them out a bit. I also added the brushed noise to the normal with a another bump node before your original with a Strength of 0.023 in an attempt to give those scratches a little depth. Finally, instead of plugging the top color ramp into the Base Color, I plugged it into the Specular Tint and changed the Anisotropic to 0.768 and the Anisotropic Rotation to 0.262 to give it that brushed metal effect to lights and adjusted the Base Color manually to get the color I wanted. I also changed the Roughness texture's color ramp greys to be almost the same just to make it a little more smooth for my taste. But without all of the learning I did with your prior tutorials, I don't think I would have ever been able to make these adjustments. So thanks again for these!
Opposed to looking like brushed metal, mine looks like a swirly mess of marbleized stone, still nb though. Edit: Adding it on a smaller object it looks spot on. Cheers
i was trying this with a cube instead of a ball. the texture worked much better when i connected the UV output of the texture coordinate node to the vector input of the mapping node, because with the object output, one side had a svirly pattern instead of line pattern
A tutorial request, please, Ryan. I have a file that looks like several objects but there is only one. I want to select parts of the file and append them to a new file. I can never see the result, hence the need for a tutorial. How to append/copy to a new file?
Thanks for another material tutorial, enjoy working on these, nice little time fillers and good to have them in storage for when I need them. I was wondering, I have a few of these done now and I have been wanting to add them to my artstation page to bulk out my portfolio a bit (providing that is ok). Would it be better to one big post with all materials, think there are about 60 so far that I have done, or to have them as individual posts or groups of materials. Any advice would be great.
i was just looking through your videos as im starting sculpting, and i just realised we have the same name which i find kinda cool, hopefully i share some of your skill as well lol
Well the thing is, every surface has at least two angles. Like on a sphere, you could go up and down, side to side, and back and forth. And on a flat plane, you have two ways you could go. Side to side, or up and down. So I think you have to tell the texture what angle you want the texture to go.
Great video as usual., I have a question if you don’t mind. If we were to use the anisotropic values, whether it’s from the principled or the anisotropic bsdf itself, how would we implement it in this context to add a hint of extra realism, if any?
Copying this from the comment I made a second ago, but this might help you! I am using Blender 4.0 for this. The brushed scratches were a little too thick so ended up with a scale of 14.7 on the Noise Texture to thin them out a bit. I also added the brushed noise to the normal with a another bump node before your original with a Strength of 0.023 in an attempt to give those scratches a little depth. Finally, instead of plugging the top color ramp into the Base Color, I plugged it into the Specular Tint and changed the Anisotropic to 0.768 and the Anisotropic Rotation to 0.262 to give it that brushed metal effect to lights and adjusted the Base Color manually to get the color I wanted. I also changed the Roughness texture's color ramp greys to be almost the same just to make it a little more smooth for my taste. EDIT: you may need to play with the Anisotropic Rotation value to get the direction you want. That is just what worked for my test scene.
Is not distance a Blender Unit distance? That is, sure you have 0.1 strength, but the "simulated" height is still a full meter or millimeter making it not look correct. Don't you really want 0.0001 distance and 1 strength?
Oh, this is a bad one, sorry. Consider deleting this and re-do it, and obtain some real world parts to observe. Check out videos how to brush (polish) a square tube mitered corner, or the craft of creating a brushed or satin surface appearance in general, or the complex anisotropic patterns left by multi axis mill machining. You're creating an isotropic flat colored metal with noisy stripes in it. This isn't achieving a brushed appearance at all. Unless for very coarse machining marks, like the underside on a pot or a vinyl, you should never use bump node, which is used wrongly as well btw (adjust distance, leave strength at 1). Also, making a simple sphere "brushed" doesn't show any of the challenges with setting up an actual part ready for a *real* brushed (or satin) appearance. 1) Always use anisotropic shading for facing angles, but you can mix in very low roughness glossy shading at extreme grazing angles (although I don't bother). You can't just modulate anisotropic parameters for this one, as both will be visible at the same time. Again, observe real world behavior. 2) Adobe has some references (Adobe standard material, worth mentioning that their anisotropy is better than Blenders in that it can handle anisotropy without a roughness requirement) if you want F0 face and F82 edge tint data for various metals. Add for that extra realism, even if some choose to ignore it. I'll ignore if I don't have the data. 3) Anisotropic "color depth" varies with treatment; machined may not have any, brushed have a lot of depth, and for satin the depth is hardly visible except in extreme closeups. 4) Try a more complex part, like a European style door handle (not a knob), then make a decision; what parts of the surface to use linear anisotropy, and what (if any) to use radial anisotropy. 5) If you want to affect anisotripic spread (with no bumps), you can control rotation using voronoi color -> subtract 0.5 - multiply [0.002,0.01] or so. 0.25 is 90 deg rotation, 0.5 is 180 degree. 6) UV based tangents works fine for directional shading, but falls apart somewhat for radial shading. Meaning you have to make a conscious decision on how to approach modelling (separating parts) and if to use multiple materials to do one job, or utilizing "UV space locations" to divide (i.e. UV.u < 0, then use this part of the single shader). 7) If the part have bent tubular geometry (like the handle/snout of the Utah teapot), you need to figure out how to create a seamless voronoi patters around the seam. Erindale has a recent video on that, just need to reduce it by one dimension to remove the seam only in one dimension which is what you need in this case. No pattern seam if U space is completely filled. 8) If the part have singularities (like a Y-tube intersection with individually brushed tubes), you need to figure out how to blend the tangencies properly. I haven't been able yet, I just blend in some glossy shading (fully glossy at singularity). This may be possible to achieve properly with Blender now with blurred attribute geometry node, but check out Entagma's "New In Houdini 19.5 Pt.1" and "New In Houdini 19.5 Pt.2" for how to deal with tangent fields.