Hello. So i'm new to Blender and i need some help. For one of my first projects I'm trying to make simple interior image. Everything is fine until it comes to rendering part. I can't make clear image without that light noice crap. I get only good results at 10000 samples but it takes 4+ hours to render. Any solutions?
>>520962Post an example of your render so we can see what you're talking about
>>512461Look before you post fucktrard
>>520962http://www.blenderguru.com/tutorials/using-portals-accelerate-render-times/Also, if you have an nvidia graphics Card you can render images with the GPU instead of CPU. Go to preferences and select rendering with CUDA cores and after select your GPU in the render properties tab.I hope this solves the issues
>>520970GPU is first word your computer has use for rendershas very modest pc - refuses render much samples
>>520962try this for interiors https://www.youtube.com/watch?v=oveSskhIEAcBasically set up an area light, set it as a portal on the window and viola
>>520962Look up how rendering works. More samples means less noise/cleaner images but longer renders. Also most of the time you will always have noise in your scene. Which can be fixed with a denoising filter in post.>>520993>>520970Andrew Price please leave and take your crappy, no MLT renderer with you.
>>520970>nvidia graphics Card Amd works, too, retard, even faster.
>not using luxrender0/10 effort
I find that changing the clamp direct and clamp indirect values to something around 1.0 can help reduce some of the white speckle noise.
>>520962I have a similar set up here.Yes there is noise in the shadows but I only used 100 passes (about 15mins) I have reduced AO down to .33 and used a simple emission plane as a light source. - bear in mind, I can't use GPU . ..I think 10,000 passes are a little excessive.Notice I have reduced bounces on light path and tile size for CPU render (it's faster).
>>520962Further to my reply here >>521049I'm wondering if your light source isn't the problem?Looking at your pic. you only seem to have a small sphere, try a larger plane. I used a white set at a strength of 15.Hope this helps.
>>521041Clamping can help but can also introduce inaccuracies.