The FASTEST Cycles Renders you can get in Blender!
Vložit
- čas přidán 15. 05. 2024
- Not using your PC? Let it work for you! Get Salad now bit.ly/Kaizen-Salad and start Chopping today for some amazing rewards and use my code UG5JA0 for a 2x earning rate bonus.
Even with the best hardware in the world, getting fast renders can be an issue if you don't know what you're doing! With the settings and explanations in this video you can make your renders up to 3000% faster. And it's super easy to do! These settings are an updated version of my old video and are optimized for Blender 4.0 and up, so start implementing them today to save HOURS of rendering.
Download the Benchmark Scene here; / 98720050
Get My Renderpreset on Blendermarket; bit.ly/kaizenrender
Get My Renderpreset on Patreon: / kaizentutorials
My Hardware setup: NVIDIA GeForce RTX 4090 FE, AMD Ryzen 9 7900X, 64GB Corsair Vengeance DDR5 RAM, MSI MAG B650 Tomahawk WiFi, Arctic Liquid Freezer II, Corsair RM1000e PSU
-----------------------------------------------------------------------------------------
🫶Support and Follow:
Join the Kaizen Discord: / discord
Instagram - / kaizentutorials
Twitter - / kaizentutorials
-----------------------------------------------------------------------------------------
📚 Blender Courses and addons you need:
Robotic Planets - bit.ly/roboticplanet
Gobos: bit.ly/Gobos
Master 3D Environments in Blender - bit.ly/MasterEnvironments
Real Water: bit.ly/realwaterb3d
Blender Launch Pad - bit.ly/BlenderLaunchPad
Physical Starlight Atmosphere: bit.ly/PSAkaizen
-----------------------------------------------------------------------------------------
Chapters:
00:00 Intro
00:21 The Benchmark
00:54 Step 1 - Your Render Device
03:17 Sponsortime!
04:26 Step 2 - Noise Threshold
07:42 Step 3 - Denoising
09:53 Step 4 - Performance
11:13 Step 5 - Light Paths
13:51 Extra Tips to try for Faster Cycles Renders!
-----------------------------------------------------------------------------------------
🎶 Music:
Epidemic Sound
share.epidemicsound.com/73h5ua
-----------------------------------------------------------------------------------------
Check out Lud and Schlatt's Musical Emporium:
@ludandschlattsmusicalempor6746
Some of the links in this description help support the channel!
I really enjoyed that salad! Want to start making money in your PC's downtime too? Get the Salad app here bit.ly/Kaizen-Salad and start Chopping today! Make sure to use my code UG5JA0 for a 2x earning rate bonus.
That salad thing sounds pretty cool... It's a shame that i have an RTX 3050ti... :D
You can always try! Not sure if it doesn’t work 🙏🏻
We xan actually do temporal denoising since 3.1 but is not exposed in interface
Antivirus sees this as a threat.
that's weird! You need to give it approval for firewall passage, but that's because it uses internet, but that's it! @@andriespieterse1584
the best rendering setting is an expensive GPU
It does go a long way lol
Uhm, that setting is greyed out in my bank balance...
@@GinnyGliderhmm.. maybe a bug did you report to blender ? Or try closing and opening blender again ..
Fr
@@GinnyGlider Yep, funny thing i tried this setting but received a defective one in the mail 😂🤣
Just one detail about decreasing the tile size: It helps you to render big scenes when you don't have much VRAM available. With a powerful GPU it's generally unnecessary make it smaller but when you can't render in 2048 tile size, decreasing it would allow you to render in exchange of more time actually. This is my experience messing around with it xD
That’s a great tip! Thanks for sharing this for the low VRAM gang!
3080ti: i always deactivate tilerendering. I often render in 4k and it's always faster than keeping it on 2048. Only when i push the gc to it's absolute limits i turn it back on.
+1, when i was rendering on 1660ti i constantly used tiles with numbers like 768-920 to speed up render. With 4090 using tiles is only increase render time
thank you! ive run out of vram very often didnt know this fix existed
3060 user here, thats the least of my problems
THANK YOU! It really works, went from 7 minute renders to 8 second renders. Every time I have long renders, I come back to Kaizen
That's so nice, thanks for sharing!
You missed something while talking about denoising. Under advanced settings in Cycles sampling panel there's animate seed option. That one needs to be disabled in order to avoid flickering between the frames. In fact, quite important. Also, I haven't seen anyone mention this but since 3.6 version, maybe even one or two versions earlier not sure, there's Turbotools add-on inside the Blender. That's accualy separate denoiser. If used properly can give even better results that Optix or OpenImageDenoiser.
That's not free though. I still use the ai denoiser from recycled. I've separated the addon from the special build and still jas way better results vs blender vanilla approach
Even better; apparently there is a temporal denoiser using Optix in Blender. But it's hidden somwhere under the developer extras. But yeah you're right on the seed for the noise threshold, good one!
@@KaizenTutorialsNever heard of it but I know EEVEE NEXT is a bit better about accumulating samples over frames which would be better for animations
@@KaizenTutorials the only place where I saw Optix temporal denoiser is inside the free Pidgeon Tools add-on. I've been playing with it for a while. Interesting results. Check that one. Cheers!
Nice! I'm not sure where it is. Southernshott has a video on it on his channel. @@_tomashh
The BVH is the bounding volume hierarchy. It allows triangle searches in path tracing to go down from linear to logarithmic time.
BVH trees have to be recalculated if the triangle world-space positions change so there's only so much data that can be persisted frame-to-frame. Edit: The wikipedia article on the subject offers a decent albeit simplified explanation of the subject.
Wow cool, thanks for explaining!
It is always a pleasure to find new ways to speed up my workflow!
A note about adaptive sampling:
It's worth trying out a scene with a lower sample count and lower adaptive sampling value in complex scenes, as it seems that the AI/algorithms were set up for images with uniform sample counts. In my experience the varying sample counts from this can cause artifacts in the denoiser. To prevent issues try increasing the min samples near to a quarter of the total samples and messing around with values as always. I usually use a value between 0.01 and 0.05.
The second image at 14:50 looks like it uses a smaller sample count or some other setting on top of disabling caustics, but it's also possible that the reduced noise from the lack of caustics is causing the adaptive sampling to give it less samples and therefore increase the artifacts in those areas.
On top of disabling the animate seed option that someone mentioned to reduce flickering, using the automatic scrambling distance option can also help reduce noise in renders by allowing nearby pixels to have more similar random values. Setting the multiplier to 0 is also a great way to visualize converging times and figure out what is a good sample count for you!
Transmission bounces are required for any transmissive materials, that includes glass that uses transmission, refraction, etc. Transparency bounces are only used for materials with alpha (i.e simulate a transparent material with no internal bouncing and only a surface) or using a transparent BDSF.
Well said, thanks for this in-depth comment!
Some tips for faster renders. This is coming from someone who renders stuff at work on an ancient PC with no GPU only an or 9th gen i7 CPU.
1. High resolution (~3000x3000) + low sample size (~32) + Noise Threshold 1. This is really good for low end PCs.
2. Small tile sizes (~128 for me) are good for CPU rendering, while larger ones a better for GPU rendering.
3. Sometimes adding textures after in Photoshop is actually faster. If you have a large wall with a lot of bump-mapping.... just do it in post.
I can pump out simple scenes under 5 mins, more complex stuff under 10-15 mins.
Thank you for sharing your tips and insights for people with lower-end PC's!
tile sizes no longer matter in cycles X iirc. It's mostly used to save RAM now
@@tabiasgeehuman Doesn't matter if you are using a GPU, CPU or both?
@@Ragmon1 yep afaik. I don't notice a difference on my rig at least (GTX 1660 Super and i7-10700K)
@@tabiasgeehuman Well I'm gonna need to try rendering on monday, without the tiles active. I'm very curious now.
Good to see Noise Threshold being properly explained. So many people still just parrot "lower your samples" when this has not been the primary control for render quality since 3.0. They even renamed it from Samples to Max Samples to show it's defintiely a different setting, but no the parrots just started saying "lower max samples". Lower max samples does one thing - reduce the samples available to the noisiest part of your image. Exactly what you don't want to be doing.
Haha yes! Finally someone who truly understands! Well said 💪💪
Which could be the standar setup for using NT, because I used it in 0.5 with 3000 samples and the results looks way noisier that a NT at 0.01 with 300 samples
(And for animation means more flickering)
Great video, great explanation, straight to the point, and really made a difference, thnx bro
Glad to hear it!
Thanks for this! I had everything set like you did minus making both CPU and GPU run at the same time. Changed that one setting and my 240 frame animation went from a 45 second render per frame, to a 5 second render per frame
Aah that's so nice! Glad this helped.
Your videos are great, informative and entertaining. Thanks bro
Glad you like them, thanks!
Thankyou this had everything I was searching for and more it really helped me
Awesome, glad to hear it!
That thumbnail🔥🔥
THANK YOU for your excellent suggestions. You saved me a MASSIVE amount of render time.
Glad I could help!
Love the tips shared and the delicious salad! Appreciate the sense of humor while learning! 🤣 U R Z Best man!!!🤩🤩 Luv it! 😍
Ah thanks a lot!
Appreciate the tutorial! Good tips in this one.
Thank you!
This video is a gem
Thanks so much
Wow, thanks for the kind words!
About using your CPU in addition to the GPU for rendering: It depends. Usually, it *should* be faster, but there also are a lot of cases/scenes where it actually slows down the rendering, in comparison to a GPU only render. Always test which is faster before rendering your animation
Yes! I agree; that goes for all these settings. Always test 1st!
I just love this video, amazing
Thanks love to hear it!
Duuuuuuuuude what an ammmazing video!! Thank you soooo much
Glad you liked it and thanks, appreciate it!
Thank you
This was a great video since I was just going to make a animation in blender, I can't wait till cycles gets even quicker or that USD file format gets better then I can render in UE5 super fast
Absolutely! Nice :-D good luck on your animation!
Why not try nvidia omniverse
@@kidgeorgegreenery Thx I'll try it
@@kidgeorgegreenery horrible to setup and also extremely hardware demanding (I only got to test the demo scene with the warehouse) and on highest settings with RT enabled features it lags and crashes on 64gb of ram and a 4090....
setting it to lower quality settings looks similar to eevee in blender exept for ambient occlusion, glass (transparency) and the builtin post processing (bloom etc.) that needs manual setup in blender.
@@kidgeorgegreenery exept if they have fixed all of my issues including the weird setup as I have to add, that I tested it roughly half a year ago
Amazing Videoo Fun to watch !!
Thanks a lot!
Biggest mistake I recently made was a total rookie move. Rendering a simple animation in cycles, but each frame was taking almost 2 minutes. Finally realized I just forgot to change the device from the default CPU to GPU. Went from about 2 minutes per frame to under 20 seconds! Sometimes the simplest setting makes the biggest difference.
We've all been there haha! Well done on figuring out the issue.
Very good explanation!
Thanks!
Thank you for this essential point, which raised existential questions! ....As persistent data!!!!!
Glad it was helpful, haha!
merci pour vos conseils avisés.
De rien.
was looking for a rendering guide today and saw your old one with a link to this one, quite helpful thanks
Glad to hear and glad the link is doing it's job haha! ;-)
Helped me a lot! you will help me save minutes/hours on my projects
Awesome!
I was told several times earlier to deactivate CPU from Optix, because it "gets in the way", though both my processor and graphic card are not as good as yours.
Also the scene is amazing!
Ah yeah it seems to depend on the hardware honestly. On my current and previous setup (3070 and ryzen 3700X) using both was faster though! And thanks, appreciate it!
@@KaizenTutorials I`ve got 3060 TI and Ryzen 5 2500x. guess I need to do some tests
This is great!
Glad to hear it, thanks!
You missed one of the crucial settings that gives 50 percent more boost 😊 just un check the tree option and see miracles 🎉
Light tree under lights ❤
Never heard of this but will check it out! Thanks!
@@KaizenTutorials thank me later 🤗
Here's my rig, it's an gaming rig tho my workstation is using Ryzen R9 5950 + RTX 3090, im sure you already know the performance in Nvidia side so here is the intel ARC GPU for you instead.
Intel i5-10400 + ARC A770 16GB
Render time :
GPU + CPU = 8m02s
GPU Only = 6m29s
note : since it's taking too long to render at 4000samples
I reduced to 2048 samples and noise threshold to 0.03
So clearly if you're CPU is slow it will actually hinder the performance.
Ah yeah so it seems, thanks a ton for sharing these numbers! As for the intel ARC gpu's; i'd hoped for them to be faster. Using NT 0.03 and 2048 samples takes my machine around 1 minute.
Dont know if anyone said it before, but if you use a bigger format, like 4k instead of 1080 and use 1/4 of the samples the denoiser is able to mantain a lot more details and sharpness. You divide the samples by 4 because 4k is 4 times the area of 1080p. Max Hay uses this technique and downscales the render to 1080p afterwards. You don't actally have to change the resolution since changing the format % to 200% should be the same thing. Hope it helps someone! P.s. Please do a full video about temporal denoising techniques and fireflies/artifact removal :P Loved the video from Southernshot, but I was asking myself if there are any other workarounds (maybe even in AfterEffects or premiere)
Thank you for this! And yes maybe I will do a video on that in the future.
This video was helpful. Thx
You're very welcome!
Yo, nice video man. Can you tell me the song that starts playing at 13:50 that song has played on multiple videos I've seen and I cant seem to find it anywhere. Thanks bro.
Yes I can! It's called Tongue Tied and you can find it on Epidemic Music here; share.epidemicsound.com/73h5ua
I was alway using OID for animation because I heard in the past that it was better for animation. I Guess it's not the case anymore, gonna try optix Denoise now.
There’s apparently also a secret temporal denoiser! Southershotty has a video on it. It also uses Optix, but is hidden behind developer extras.
Use small tile size can help reduce VRAM in my experience
So i've heard yeah!
u know i never played with the noise threshold before this . . . u have saved so much time on renders . . . from 5 min to under 1 . . . with openimage its still fairly good unless zoomed in on reflections
Awesome, glad I could help you!
what blender version you use?
Honestly I barely use the noise threshold because I like the more obvious result of just increasing the number until it's too slow. I feel like the not noisy areas don't take long to render anyway
If you have the Noise threshold off all areas will take exactly the same amount of time since Blender will render all samples for every pixel.
That Witcher and Oppenheimer soundtrack on your Spotify...🔥👌🏻
Soooo good!
Another tip: render from the command line. Some utilities online can do this, for those unfamiliar with shell scripting (like the Flamenco add-on, which I use), but this is another way of managing memory. This way, you don't have the full Blender UI running in the background while rendering - helpful for longer animations. A middle ground to this is to lock the Blender interface while rendering (meaning you can't use the UI while it's rendering) by going to the top menu -> Render -> Lock Interface.
Yeah i've heard it can shave off a couple of seconds, but I honestly couldn't get it to work myself. Then again I have 0 coding experience and understand nothing about it haha
@@KaizenTutorials that's fair. I have done a few tutorials in the past; maybe rendering from command line be my next one 😄
@@SpencerMagnusson I subscribed in case you do a tutorial in the future. Thanks for sharing your knowledge. The Blender community is so amazing.
@@studiojohnny thank you! I'm wrapping up some videos for BCON LA but I added this to my backlog to get into soon.
Dude is this true ?
“KAIZEN™ means improvement. Moreover, it means continuing improvement in personal life, home life, social life, and working life. When applied to the workplace KAIZEN™ means continuing improvement involving everyone - managers and workers alike.”
Yes! It's the very reason I chose this name for the channel and my business. :-D
@@KaizenTutorials to stay motivated its amazing mate !💗
Not me waiting for my 350 frame animation to finish rendering while watching this video 😅. Usually, I significantly decrease the sampling and resolution, so these tips are a huge help
Aah dang haha, poor timing. But for next time you now know! ;-)
AFAIK, Tile size optimizations only works if you have more GPU cores than what cycles can use with parallel processing in a single tile. If you have equal to or less, then you are best off rendering 1 large tile. If you have more, your better off rendering a number of tiles that has the largest number of cores per tile simultaneously without going over. at least, that's how I understood it. I don't know how many cores cycles is optimized for. if it's 4 cores, and you're doing CPU rendering with a threadripper with like 60+ cores, multiple small tiles will be faster because it's rendering multiple tiles at a time, but AFAIK the average consumer-grade GPU can only render a single tile at a time, and reducing tile size and increasing the number of tiles, actually increases render time, rather than reduce it, because it's only 1 tile at a time, and switching tiles takes time to load new data to the GPU. I am NOT a programmer, or really any kind of expert, I just watch a lot of youtube videos, and tried to render on my laptop and this is what my experience of it points to. simultanious tiles = total # of cores MOD max cores per render.
So from what I've been seeingin the comments tiling is mostly useful to prevent VRAM issues and for CPU renders. Modern GPU's render faster on full tiles vs. small tilesizes.
@@KaizenTutorials is there a way to get a specific, optimal size for the GPU to take advantage of tiles, or is it always full image tile for GPUs now?
If youre using GPU, deactivate Tiling, Turn on Denoiser Optix or OpenImageDenoiser, low the samples to 12 and in output in resolution activate 200% With this you will denoise a bigger image that will have more details and you will get a better result with the denoiser because this one understand better the image. And deactivating the tiling, its because GPU handle better the whole image than tiling images. Doing test the tiling always got more time than the whole image.
Great tips, thanks!
I’ve seen a few people recommend that low samples high resolution trick and I’m a bit skeptical about it. At least in the scene I’m testing that has volumetrics, the results are quite a bit worse and has very noticeable blotchy denoising artefacts in dark areas. Not to mention the slightly slower render times even at such low sample counts and you’d be using exponentially more storage space if you keep the renders at that 200% scale. Doesn’t seem worth it to me.
good tips but no offense this video could be like 1/3rd the length that it is
it could be. It could probably be 1 min long, but there wouldn't be any fun or explanations in it. And I'm a strong believer of not just showing, but also telling!
I like the talking and animations. We are to used to fast videos and forgot how to enjoy some quality content ❤
Kaizen God.
thank you for the tutorial it went from 5 hours to 5 MINUTE!!! hahahahaha.... BTW for the tilesize it actually make my render more faster not significantly faster but enough to cut the time for rendering.👌👌
I did a comparison between gpu + CPU and on a 4k x 4k simple logo scene with metallic materials it was 13 seconds faster with both activated (optix RTX 3060ti and Ryzen 5990x) vs 7 seconds slower when tested the same scene but with a 2k x 2k resolution, I think it benefits on larger and bigger resolutions at least for me, to have both CPU and gpu activated.
Ooh ok that's a good find. thanks for sharing this!
Very interesting and a good hint! So better no CPU for animations in standard 1920x1080 resolution, but CPU may be faster for single-images with larger resolution, for printing the result in a big DIN-A4-scale, or larger.
Surprised you havent mentioned scrambling distance. Its an absolute lifesaver on cityscale+ scenes
I don't even know what it is! Care to explain?
There is mistake in the description of how Light Paths works. Transparent is not responsible for bounces in the glass, it works in bounces of the alpha texture in the shader (ex. leaves in trees with alpha mask)
You're right! But that also affects glass believe; atleast it's transparent component. which you can see in the render as using these bounces did make it see-through again. But for the light refraction inside of glass you'll need other bounces!
@@KaizenTutorials You can doublecheck it with array of glass planes on front of Camera. If you add alpha weight to you glass shader, transparent bounces will work
To avoid running out of memory, I always save the project, restart the pc then render immediately after its on.
because restarting your pc will clear all ram and that's what we want.
That's a good tip too! Makes sure the system is fresh.
For tile size:
It did massively help for me on the CPU (especially in multi threaded systems), but in my experience, it's comparatively slower on the GPU
Thanks for your input!
Cheked to see the pc specs in the description only to find out that i literally have the same pc, only a different AiO cooler. Noice!
Awesome haha! 👏🏻👏🏻
Dude I've been rendering these animations for so long taking hours of my day just for a 5 second animation just to figure out I've been limiting my hardware and could've saved hundreds of hours just by clicking a button, my life is a lie
also thank you this video changed my life
damn that's rough! Happy this video helped you out tho
Thumbnail is very eye-catching, but it took me a minute to understand it. I think switching the places of the button and text would be much better!
Ok! Thanks for the input. 🙏🏻
Hey Kaizen,I used eevee for some renders in the past but sometimes I scared that it could crash my computer,is eevee a better render engine to use for less powerful computer like mine ?
EEVEE is definitely great for less powerful machines, but it's also not a path tracing engine and doesn't look as good as Cycles. If you use the tips in this video I think you can render in Cycles on any PC!
Nice video!
Remember that Salad using you GPU is not free, electricity cost money.
That's true ofcourse!
I wonder if command line rendering would improve rendering speed as you don't have the UI open at all, and everything is done in the background?
I’ve heard it does, but I couldn’t get it to work
Quick heads up about 2:53 : Render times can still be slower if you have both CPU and GPU enabled. I tested it in a scene I was working on and it was quite a bit slower (I'm using OptiX).
Ah yeah that's true. It's really dependent on hardware, scene and setup.
Dude! I didn't even know you actually had to "enable" your GPU in the settings, I thought you just click GPU Compute (even though it was greyed out) and you were good. I've noticed that 3 minutes has changed to 3 seconds (a little dramatic, more like 7 seconds)
Eyyyy that’s a big win then haha nice
I've found that Persistent Data causes some artifacts in random frames, such as lighting differences. I'd often have to go back to the problematic frames and switch off Persistent Data to render out those frames separately
It can yeah, also with animation stuff. But generally speaking it does a great job in my experience!
I have an RTX2060 and was running out of vram using Optix rendering an animation. I was advised to switch to Cuda which didn't run out of memory, any idea why?
I believe it’s just a matter of how the devices handle the process. Optix stores a lot using its RT cores, filling the VRAM. Cuda uses something else (not sure what) which apparently doesn’t fill up as fast and is probably the reason cuda renders tend to be slower.
What font do you use for your captions, it’s gorgeous!
It's called 'Caught'! :-D
Got the rendering time to 59,75sec with rx 6600, ryzen 5 3600 and 16gt.
Gpu only, no tiling, no noise threshold sample 64. OpenImagineDenoise accurate, seed 1, max bounses 5,4,4,4,0,8, Fast GI, Simplify 1 max divisions 12 & texture limit 8k. Use spatials, persistent data.
Great! what was it before optimization?
thank you for the tips! are you able to speak from outside of brand-deal youtube sponsor talk and say weather or not salad is actually safe to use?
No problem! As for Salad; I've been using it for weeks now without any issues, so I'm pretty sure it's safe. Salad also states themselves it's safe.
Hi, thank you for a extensive video on one of the big alltime problems :)
Thy for the work you put into this great video.. dont take my rambling to serious if you read this ;)
a bit of nitpicking on things where you are a bit unclear in what you say:
Optix: I absolutely agree that on most reasonable systems for most scenes CPU+GPU is probably faster. But rendering with GPU+CPU is not always faster. It only will be when your CPU can handle preparing the render overhead while also rendering. On a weak CPU fast GPU system this can increase GPU idel time, while it waits for new data... It will not only depend on the System but also on the data that has to be prepaired. So its also sceen dependent.
For samples you are talking about an image with one "color" .. needs only one sample. I would say unlit objects souch as emission shaders. The color has nothing to do with it.
Denoisers: "since optix uses an ai powerd alogrithm using the RT cores on your GPU it has a way more stable result for animation renders" Sorry - this makes no sense. Optix and Open Image Denoise are both "AI powered" and saying that the RT cores calculate better then a CPU would be like saying numbers spit out from a calculator are better then numbers from paper. I totaly agree with your advice though :) but pleas dont make up reasons.
Persistand Data ... this one im not shure how it works. But for shure it dosnt just prepaires data once beforehand and then renders on... then animations wouldnt work. Judging by the bugs it has led to in the past i would guess that it trys to update data only when nessesary.
Best
Thanks for your input and kind words!
As for your remarks;
I agree GPU + CPU is a discussion that will depend on a lot of points. Both the scene and the hardware are points to note for if this will work or not. However from what I've seen from people telling me in the comments, most people with OK hardware do seem to get a boost from using both. This might also be because a lot of more recent CPU's have integrated graphics.
I get what you mean with the unlit objects. However the goal of a sample is to determine the color of a pixel, or atleast that's how documentation on path tracing seems to describe it. Therefore I think colors have something to do with it, although if every pixel would have a solid, unlit color, but unique at the same time, 1 sample would still be enough. So you're absolutely right on that!
As for the denoisers; this is what I gathered from research. I'm honestly not sure how they work, but Optix does infact use RT cores on RTX GPU's, which is why you can't use Optix on non RTX GPU's. This unique feature could be faster. My analogy would not be calculator to paper, but simple calculator to a fancy one with additional computing power.
Persistent Data is a weird thing. Obviously it doesn't store everything, since it would indeed not work in animations. And in my previous video on this subject I recommended people NOT use it on animations, since it created artifacts with moving objects. However this doesn't seem to be the case anymore. I think the data it stores does not include relevant data for animation anymore or something?! Hard to find proper documentation on these things sadly.
Is RTX 4070 (non ti or super)a good card for blender cycles?
For sure! I’d look into VRAM though. The more VRAM the better for Blender :-)
How does the Persistent Data feature work when using a renderfarm? Good? Worse?
I think it works just the same. Persistent Data is just a setting and Renderfarms take your blend file and use it;s settings for the render or they let you set up your render settings in their tool. Either way renderfarms shouldn't functionally work any different than Blender itself.
damn this tips made my renders waaaaaaaaaaaaaaayyyyy faster
Glad to hear it!
I wonder what settings u uses for ceramic render cycles because denoise slows down a lot
Settings are always the same for me to be honest
@@KaizenTutorials i got the render problem solved but i do not no how.
Rendertime with Blender 4.0.2 for the original scene from the link in the video-description, frame 1, no changes made: 05:25:27 minutes!!! Hardware: AMD 3970X Threadripper (32 Cores); ASUS ROG-STRIX RTX-4090 OC; 128 GB RAM; 2 TB Samsung SSD 970 Pro; OptiX, Cycles CPU+GPU-rendering; But turning down the samples and the threshold will make it faster for sure.
Rendertime with changed parameters: Noise Threshold 0.05; Bounces: 8, 4, 4 ,8 ,0 , 8; Persistent Data; Max Samples 2048 equals 00:31:21 minutes!!!
you should turn off the CPU and only use the GPU, and the default sample rate of 4096 is unnecessary high if you try to render a single image. What do you mean with original scene?
@@MrDragon005That's what I never got a clear answer for at discord-channels: Is it better to use CPU AND GPU, if having a "not sooo slow one", like me, or use GPU-only? Please explain your opinion. :) "Original Scene" means, that I have not changed anything, in the scene, before hitting "F-12". :D Rendertime is 00:28:14 minutes if using GPU-only!!! WHY????
That’s awesome!
Both is better in my testing both in my current setup as my old one with a 3070 and ryzen 3700x.
Using 4096 is fine because changing the noise thresholds prevents unnecessary sample renders. So you can leave it up. As I show even setting the samples to 20K doesn’t matter because the noise threshold never let’s you get there!
does blender use automatically memory ram when is running out of vram, or do i have to render on both cpu and gpu for that to happen?
I think it does take in CPU RAM, but if you run out of VRAM your render will always crash, except if you started by only using the CPU in the first place
i am new in blender my pc specification 8 gb ram amd athlon 3000g using intrigated graphics card this is help full for me
Glad to hear it!
Ryzen 5700F + Intel Arc A770
(CPU+GPU) First time without changes without any programm on PC- 30m
(CPU+GPU) Secont time 2048 samples with browser - 17:06
(GPU) Third time 2048 smaples - 16:19
looks not good xD But 3D I use for make 3d elements for Website, and for this task intel can be in PC
Thanks for sharing! You still managed to get almost 100% increase in speed though, so that's great!
Bro in the case of the the denoiser, I found out that Openimage its way better than Optix even in animations because it reduces more the flickering, unless... Im doing it wrong with Optix (And yeah I have an RTX 3070), any ideas???
Not sure, in my experience optix is vastly better for animations. You can also use a seed for the denoise variation and set that to 0 so the seed doesn’t change. Further improves it! Oh and apparently therss a hidden temporal denoiser somewhere. Southernshotty has a video on that!
I don't wanna be a shill, but there is a addon I use to use a hell of a lot called "Super Image Denoiser" which does have a temporal denoising setting. Back when I used it it could do a pretty amazing job at denoising scenes, but nowadays I'm fairly sure it's broken as it's free version (I think) has stopped development and is now been put into a pretty nifty addon tool kit
No worries, thank for mentioning this improvement for people.
why not use optix for still images? it uses the rtx card, not the cpu. so it is faster and let me multitask when i am rendering. by not taking my cpu as hostage. unless, open image is lighter on my rtx and let the card render faster bt splitting the load to the gpu and cpu. what do you think ?
I guess you're right in the scenario of not taking your CPU hostage. But I never render and do other things simultaneously, so I only use the full power of my PC to render and then do other stuff not on my pc during the render.
Isnt it always better and faster to do the de-noising in the compositing tab?
Afaik that's the old way of doing things!
@@KaizenTutorialsOh ok, I was always told that was the better way to do it but everything goes out of date so fast now its hard to keep up
omg I need to try this persistent data. why the hell is it not on by default??!
I believe because it takes up a lot of RAM, so that makes things a bit harder on low end PC's.
I understand that for some projects with special materials or dark spaces, you really need a high sample count, but for most stuff I use 258 or 512 samples and let the Denoiser do the rest. I tested it a lot and for me, there is absolutely no difference anymore when I go higher than 512 + Denoiser.
Hmm yeah I guess it could be scene dependant, but does it really make it a lot faster if the scene doesn't require many samples anyways to set it lower vs keeping it higher and using the proper noise threshold? Effectively this does the same in my experience.
A question about the selection of GPU and CPU in the system settings: I have now activated both again and with the CPU activated, the render takes almost twice as long in most cases as when only the GPU is activated. I have an RTX 3090ti and i7 14700KF. I guess my GPU is too strong??
Hmm ok that's weird! With my setup it's always faster, but it seems to be both hardware and scene dependent!
@@KaizenTutorials Gonna test this on some heavy scenes as well. But with the CPU enabled sometimes even my PC Crashes, might have to change some tweaks in the UEFI
Looool! A GPU never can be too strong! I wish I had 2 RTX 4090 OC inside! ;) :P
Hello, I see that you are well versed in Blender, I would like to ask you a question. What settings would you recommend for rendering fur? I have a character model with millions of hair particles and even in a simple scene where only this model, it renders badly, dirty. I have a good computer, 3090 RTX graphics card.
I think in this case you're best off to render things in separate layers like I show in the end. This way you can render with separate samples and noise thresholds. Now you can take everything that's not fur and render it lower quality with denoising and the fur on high quality, but since each render focusses on it's own thing it's going to look a lot cleaner!
@@KaizenTutorials Wow! Thank you so much! I hadn't thought of that, but it sounds logical. But if the fur is rendered separately, then how do I combine that into one image or animation? I just don't understand how to render the fur separately and then join it with the rest of the background. After all, the fur is transparent...
In my experience, tile size just puts less strain on the hardware. I kept the tile size the same while I was learning blender, but kept getting crashed when trying to render multiple donuts. Now Don crash yet
yeah and then u render animation, get boiling noise and setting up 10k+ samples or installing temporal denoise addons, cuz this advices work only with still single image renders
Well temporal denoisers work! Blender apparently has a built in one. Southernshotty has a video on it!
You're using a 7900x or a 7900X3D kaizen? I'm currently on a 7800x3D. 7900x any good? Found description lol.
I'm using a 7900X as I found the price for the 3D high for what it offered extra. So far the 7900X has been serving me extremely well. I can run 10 apps at the same time, game, video edit and use blender. All simultaenously!
@@KaizenTutorials woah you reply really fast. Thank you. I'm going to upgrade my GPU first XD. Got a 3070 ti for now. Saving for the goat, maybe 4090 or 5000 series if possible. Thank you for the wonderful content. I'm learning a lot.
I havent tested this in blender but the smaller tile sizes apparently speed up render time in rhinoceros (cad software) when using cycles rendering on the CPU. They claim that the more cores you have, the smaller you can set the number such as 16, 32, 64 or 128. Smaller numbers for more cores. While GPU rendering, you generally set it as high as you need for the scene.
I tested the GPU & CPU rendering (optix in your scene) and I got worse results just as always compared to GPU only. I dont know what you did differently but I am using a 4090 with a 7950x3d and also 64gb of ram :I
I just tested roughly with the tiling set to 2048, 64 and 16 and the tiling is not parallel anymore and appears to be only for memory constraint situations now. The overhead of copying the tiles and files in and out of memory makes it slower the smaller the tiles (higher amount of tiles needing to be stitched) and the random values for the noise and denoise appear to be changed as well, resulting in slightly different lighting on first glance and across multiple frames (more flicker). I used 64 samples in cycles on the CPU to not have to test these for 20 min each xD (on 2048 samples)
It's weird how the same hardware can behave differently. My renders usually vary in time too, even if I render the same frame multiple times, the times always vary. All my tests with tiling did nothing, but then again I didn't test using CPU only, since I'd never render using only CPu anyways haha
Thanks for sharing your tests/insights!
@@KaizenTutorials Rendering with CPU only really is good for testing new materials as the compiling of the shader for cycles rendering is (supposedly) almost instantaneous while it has to compile these for cuda/optix first when you switch to rendered view/rendering with the GPU if it isnt already cached. Another benefit is the area of cloud/server rendering and you want lots of "vram" to render a big scene. Here the CPU rendering can be the way to go as well, as it is often much cheaper to rent a server with a good CPU and lots of system memory (which is used in place of vram when using CPU) than it is to find a server with a cheap GPU that also has tons of vram. I heard that (design) studios often use CPU rendering to dial in the look and switch to GPU rendering later for the animations, if they have enough VRAM.
aah ok wow, thanks for this in-depth reply!
@@DiverseGreen-Anon
Any recommendations for Apple M3 CPUs with RTX capabilities?
Not really anything besides using the settings as displayed. I think Apple benefits best from the Metal option, but not sure as I don't have a mac or have ever used mac for blender.
yahooooooooooo thanksssss
💖
11:39 The lines should bend since the ray is changing medium but thats very nitpicky
I think that's actually called refraction, not transmission. :-)
I haven't downloaded your scene so I don't know how you have setup your glass material, but as a general rule I would not recommend to set the Transmission bounces to 0. Because if you have not setup some "fake" glass material where you mix in a Transparent BSDF or other Alpha transparency, but either use only a Glass BSDF or Principled BSDF, glass is a transmissive material and will need Transmission bounces or otherwise everything seen through it will be black. By the way, the channel to make a glass material in the Principled BSDF is therefore called "Transmission", not "Transparency" 😉
The Transparent bounces (which are by the way not dependent on the Total bounces unlike the others above it) are used for Alpha transparency which is produced by the Transparent BSDF or the Alpha channel in the Principled BSDF and also controlled by the density of volumetric materials.
You're right, however I'm using a proper glass material in this setup AND I have transmission bounces to 0 and it still works!
@@KaizenTutorials As I said, I have not downloaded your scene so I do not know how you have set it up. Is there anything to see behind the glass except for the world environment? Are the windows single sided planes or do they have a real thickness for correct refraction? The thing is, transmission bounces at 0 work for a single-sided glass material if there is no object behind it (otherwise the object would be black). If your glass has a real thickness, it will no longer be transparent, because the rearward face will be black through the front. If you have a single face and nothing to look at behind the window you're good, because the environment will be visible no matter if you have 0 bounces or not. So as I said, works for you but I would not suggest this as a general rule.
ANOTHER TIPS:
u can use ai upscaler (CupCat AI upscaler) and render with half the resolution desired.
u can use ai frame interpolation (FlowFrames) and render with half the number of frames
Yep, good tips! They tend to work a lot better for projects with slow, predictable motion, but if you have one of those it will save you hours and hours!
my problem is that the Cpu is just as fast as setting it on the gpu. currently i just port my animations to unreal engine 5 and render it on that and it is as fast as watching the complete rendered animation as a video
Aah yeah ok that makes sense
I think your hardware setup is missing from the description
You're right! Just added it in, my bad!
I have an old cpu and a new gpu. I found that if I increase my samples it gives my cpu time to cool down in between its denoising tasks 😂.
A tip for my fellow budget Optiplex fiends
Haha nice tip! Thanks for sharing :-D