Return to

Learn Blender with us!



I added the Glow-egg in the guard, but couldn’t get volumetrics to look that good coming from the beam itself, so will probably have to do that in post.

I did add general volumetrics to the scene, though.


It can be finnicky if you aren’t using full pbr workflow, here’s a decent tutorial on it:

its best to think of it like laser tag in reverse: the light makes you see the “fog” instead of the fog making you see the light/


I’ll watch it. But I can’t stand that guy’s accent. It looks like he has a lot of great content but a persons voice is what makes or breaks a tutorial for me.

Is it petty? Yes. Is it helpful? No.


yeah, it’s like deaf borat

still tho, good primer on it


The setup working the best for me:

  • Use a Surface emission material for almost all Cycles interactions like diffuse, Glossy etc except for the camera!
  • Make a slighly bigger tube of volumetric emission (just plut the emission node into the volume output) and make it visible only for camera. and mabye refraction, if you see the Blade trough a glass tube or something like that.
    (Turning on and off visibilities in cycles saves you a lot of performance. I prefer to make more complicated materials like Subsurface Scattering or volumetric stuff in general only visible for the camera and make a simpler version for all other interactions to reduce render time.)
  • Crank the Volumetric Emission Strength up … like a lot. 50-100 is probably a good value depending on your scale. You will see the volume getting a lot denser.
  • Do all of the glow effects in post using either the glare filter try making an own arrangement with blur and mix nodes. Glow in general is a camera effect so you don’t want to do it in the scene, because you’ll get less realistic results while having increased render time.
    I also did a lightsaber once and there’s a breakdown showing the stuff I rendered and afterwards the stuff i did in post here
  • If you go to the render layers, you can select passes. By Checking the Emission checkbox, blender will provide all Emission materials in front of a black background as a second rendered image (don’t worry, this does not impact render time) You can use this to have the bloom effect ONLY on your lightsaber.
    -To get the lens flares simply look up a nice bokeh Image (idk something like this with a bit smaller light blobs or make one yourself and multiply it with your glare or blur effect. Play around with the bokeh saturation and multiply strength to get the results you want.
    The node setup:


TFW you realise that you can render an image with low samples at a smaller resolution to use the compositor, then re-render it at high quality and those settings will stay.

TFW I’ve been waiting for 4-8K renders to finish before compositing this whole time.


lol that’s rough


Progressive refine with 15,000 samples. That’s a lot of hours.


Finally at a point where I’m happy enough with it until my skills improve.


OK … I’m guessing you needed those 15.000 Samples for this lightsaber … to be hones this shouldt not take more, than 1.000 … I’d even say it’s probably more of a 500ish scene. you might want to use denising or some noise reduction techniques in general. There’s a good article on blendergurtus website, but I can also chat with you on my personal experience.


progressive refine just runs until you stop it and save the output, the upper limit doesn’t matter


Yeh, I stopped it around 450.


Ohhhh … I actually didn’t knew that. I kkind of like pixars system with render Man tho. You basically set a noise threshold and then each “chunk” gets rendered as long as it needs to be under this threshold. So If theres an area in your image with a lot of noise, this area will get a lot more cycles than other areas.


Didn’t use any tutorials at all for this one. Pretty happy with the results.


Modeled my apartment’s floor-plan. Final goal will be modeling my entire apartment, furniture and all.


What is everyone’s opinion on using pre-made models for your scenes? Things like lamps or tables or something like that. Or addons that have pre-modeled things like grass or trees or buildings?

Do you feel like you can use these things and still call the end result your own? Or do you model everything you use by yourself?

I’m having an internal argument on whether or not I feel comfortable using CC0 or paid assets (with commercial licensing granted) in my scenes and still calling them my own work.

Note: I’m not talking about importing a table, rendering it out with some lighting and saying it’s mine, I’m talking about a finished Scene.


definitely use them, but not at the expense of using your own modelling

if you do everything for say, architectural renders, yourself, you’ll never meet deadlines

but if you just flip assets then you aren’t learning anythign or improving.

General rule of thumb: block out the scene yourself, as muc h as you can. use objects as “filler” to hit the finish line.


Alright, I can see what you’re saying. I often have trouble finding the balance in what I do, and my mindset going in to blender was to make everything myself and model everything (except my fog scene, but I wasn’t super proud of it because I thought it was “cheating”).

But I guess you’re right. I wont always have time to model everything myself if I’m doing a huge scene, or if I’m on a deadline in the future.

Thanks for this.


I totally agreee … don’t be me and condem the use of pre made assets! I mean you can’t do anything from scratch really.
You have to get inspiration, you have to get reference and … like you’re using blender and you didn’t write the render engine yourself. It’s totally normal to have other peoples work in your own creations and trust me it’s a pretty good feeling if people see you using their content to create something, that is better, than the sum of its parts.


Thought this deserved some attention over the weekend.