BJ Cycles - Cycles Renderer in Daz Studio
bluejaunte
Posts: 2,156
Posting this to gauge interest. It's a major project that I won't be able to develop in the regular "finish then sell" way. At least not without going bankrupt. Or treating it like a side project that will take forever. It's gonna be a grind, also because the idea is not another "need to manually convert a thousand things", it should simply take our known shaders like Iray Uber and PBRSkin and work with them as much as possible. At least that's the dream. And later when we've squeezed everything out of those we can expand to a proper Principled BDSF shader and/or other Cycles shaders but with everything pre-filled in from the Iray shaders. So more of a "get the last 20% out of Cycles" rather than start over. Hopefully.
I was thinking of putting up a Patreon for this, which is frequently used to finance development of such projects while providing access to early builds in various tiers and probably a Discord. Let me know if you'd be interested, and any thoughts you might have. Is this a dumb idea? Do we even want Cycles in Daz Studio?
Figured I can't do this without showing anything, right?
Cycles Preview Pane
A preview pane exists. Click "Start" to render progressively. Not yet live sync, as in you cannot move around the viewport and get live updates. But it shows progress of the render clearing up until samples are reached or time limit is exceeded. Option to denoise the render at the end. And oh boy, what an awesome denoiser it is!
Render Settings
Hooked into proper Daz Studio render settings. Choose BJ Cycles as the renderer in the drop down. Cycles specific settings under Advanced tab for now.
Regular Rendering (Ctrl-R)
When Cycles is selected in Render Settings drop down, render starts in the Daz Studio render window you know from Iray. Progress is shown just like Iray. When Denoiser is enabled, it denoises the final render. Cancel works too. Save the image as always.
Iray Uber Shader Support
Support meaning "technically working", not fine tuned in any way. I suspect this is gonna be slow and tedious work. At current level of support, this scene does at least render somewhat plausibly already. It's not a complete eyesore.
Current material support is roughly:
Base Surface
- Base/Diffuse color
- Base/Diffuse color map
- Opacity / Cutout Opacity / Base Opacity
- Opacity map
- Roughness / Glossy Roughness / Refraction Roughness
- Roughness map
- Metallic / Metallic Weight
- Metallic map
- Normal map + normal strength
- Bump map + bump strength
Emission
- Emission Color
- Emission Color map
- Luminance
- Luminance Units
- Emission Temperature K
- Opacity-aware emission, so cutout/emissive surfaces can be transparent where needed
Glass / Refraction
- Refraction Weight / Transmission Weight / Glass Weight
- Refraction Index / IOR
- Refraction Roughness
- Glossy Layered Weight / Glossy Weight
- Refraction Color / Glossy Color / Share Glossy Inputs
- Thin Walled
- Transmitted Color
- Transmitted Measurement Distance
- Glasses look alright here, bottle is quite wrong. Too reflective, should be darker green and more transparent.
Shader Behavior
- Ordinary materials become Cycles Principled BSDF
- Fully transparent materials become Transparent BSDF
- Emissive materials become Emission shaders, optionally mixed with transparency
- Lots of stuff wrong here I'm sure, eyeballing chaos
Instancing
Hell yeah, got that working too. Grass here is Ultrascatter instances, properly translated to Cycles instances.
Subdivision/Faceted/Angle
Subdivision surfaces supported. Regular meshes rendering correclty faceted when the angle settings on the surface is set low enough as with this cone.
Supported Devices/API
- NVIDIA GPU through OptiX (fastest by far in my tests)
- NVIDIA GPU through CUDA
- CPU
- AMD GPU through HIP (Experimental, don't have AMD hardware so needs testing)
- NVIDIA 50 series in DS4? Untested but don't see why not
Lights
Just rudimentary spotlights currently. So Daz Studio spotlilght becomes a cycles spotlight. Should be quick to add more light types though. Mesh lights also working because Emmission works (see above). In this scene there is only one spotlight (and the emmissive surfaces, they add almost no light though so this scene is lit 95% by the one spotlight).
Denoiser
From this low samples noisy mess that rendered in a few seconds to this pretty passable thing? Crazy. That glass cone is barely even visible through the noise.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2026 Daz Productions Inc. All Rights Reserved.
Comments
It is interesting, but I see it will be too expensive for me. Good luck with the further development.
Cycles is only so powerful bacaise of all the amazing shader nodes and other features coming with blender.
I personally would prefer to have DAZ as a plugin inside Blender (and also having the features added natively in Blender that is Blender missing compared to DAZ - shapes that can affact amartures, HD shapekeys and shape keys assets).
Maybe DAZ is intrested in buying your current Cycles plugin and than finishing it? I think thats more realistic than that we get ever an official native DAZ in Blender integration
I would love this. I really like Cycles as a render engine, I've always found it faster than Iray (and it played better with an older computer I had too), and it looks like you've already got things converting beautifully. I'm definitely interested!
I'd definitely be interested in this, I'd like to try the Cycles engine within Daz! I've noticed it doesn't seem to dislike dark scenes as much as Iray does
This looks interesting and the render results look amazing.
Sounds like an interesting project (and definitely like a lot of work
).
Having a render engine not dependent on Nvidia hardware but which can use Iray materials would definitely be a welcomd addition to DS.
In principle this is a fantastic project. It opens up Daz Studio and its ecosphere to anyone with a discrete GPU and at this point I have low confidence that Nvidia will continue to produce consumer-grade GPUs in quantity in the future. The A-series professional cards might persist for a little while but these are not really what I would call an affordable option for most people. At some point 6.x is going to be become the default option, probably with the release of the next generation of Genesis figures, and a lot of folks who can't afford to buy Nvidia GPUs with ever-increasing prices and ever-decreasing availability will find themselves locked out. This isn't my pessimistic take on things, it's what Nvidia themselves have announced. Who knows how things will actually pan out but having more options is always a good thing.
Thoughts that occured in a random order - please forgive the stream-of-consciousness style here:
1) Daz might not be onboard with this due to licensing agreements with Nvidia. I obviously don't know but I'm wondering why they have never pursued this option themselves. It's a no-brainer for increasing the customer base but they've never done it or at least never announced that they were pursuing it. Cycles licensing allows for closed-source commercial products (see K-Cycles) so why not? That this thread is still here suggests there is no opposition to your project though.
2) The Diffeomorphic folks have been at this for years, albeit from a Daz -> Blender conversion perspective. There are still bugs in it and still some results that aren't that great and there probably always will be. It's an awesome project with huge scope and there's no way one person can lock everything down in any sort of timescale: Skin still looks a bit artificial, card hair is nowhere near as good as in Studio itself but the thing is... this is no big deal. Because, as @Hurdy3D pointed out, we can open up the Shader Editor in Blender itself and sort out those problems, or use add-ons that bypass the parts of the conversion that are less than stellar.
I transferred over Stonemason's recent Main Street USA set and there was a honking great UV seam right down the middle of the roads that isn't there in Studio/iRay. Easy to fix in Blender (took 20 seconds), but you would have to get this sort of thing nearly 100% correct with yours because there are no similar tools in Studio. It's a big ask. Custom shaders like OOT's hair? Volumetrics? Makeup? Omni? And those are the ones that just popped into my head right now. Diffeo has a lot of this nailed so if your Python is up to scratch...
Still, I think yours is one project where the maxim 'Don't make perfect the enemy of good' would be very appropriate.
3) It's probably too soon to ask but what are your longer-term plans for this? I would be happy to contribute to funding the development on an ongoing basis, and this is going to be seriously long-term if you plan to do this solo, but if it eventually ended up behind a Premier paywall at some point I would be a little bit miffed, to say the least. I think you're going to have to address what your commercial plans are at some point in the near future if you want to get people to put their hands in their pockets.
As has already been suggested, I too would prefer to see Studio's character creation tools ported to Blender, but I'd be more than happy to support your project on an ongoing basis on Patreon.
Edit:
The other possibility is just to start selling your plugin now on an Early Access basis, as game studios regularly do. People would have something tangible (and obviously incomplete) for their investment and you increase the price as the feature-set and the conversion performance improves. It's worked for many products (Minecraft and Baldur's Gate III are the obvious ones amongst many), and you might find people more amenable to a one-off purchase with more people getting onboard as it improves, albeit at a higher price than the early adopters. You could, as board game makers like to do, pre-sell your product on Kickstarter. 3000 sales of your plugin at $40 each, for example, would certainly fund a fair bit of development but the pressure is then on to deliver.
We've chatted mostly on the Linux thread but I am a games developer with a small company and staff - it's tough out there at the moment for us, and I can't imagine how difficult it is for freelance professional modellers, but there are always options.
This is awesome - make it happen :)
Thanks for the feedback. Yeah too soon to ask about long term plans. What happens from here on out is pretty much entirely based on interest and how much money a Patreon could generate. Of course one has fantasies that there will be hundreds of people interested and I could make some kind of decent income while developing this with an enthusiastic group of backers and also the sky will be pink and there will be fluffy cloud elephants spurring me on every day and I'll be so happy for the rest of my life.
Surely I'd make sure that backers are first class citizens and not lose access in the end. That would be cruel and backwards. Certain tiers could guarantee the final version, or a certain treshhold of minimum months subscribed. Or hell, even anyone who was ever subscribed could get the final version for free, who knows. If you're thinking of Daz buying me out and then putting it behind a subscription like Mesh Grabber then I hope that I would negotiate a deal that prevents backers getting screwed over. This is so extremely hypothetical though, not even worth contemplating.
Daz has no say in what I develop. At most they may not want it in the store, but they have no way of preventing development. And don't see why they would object anyway. If anyone could do that it would be the owners of Cycles, but as you said the license is in the green so that's not gonna happen. Apache License v2 was specifically chosen so that Cycles would get adapted to other software including commercial.
Why did Daz choose Iray? No idea. Maybe it was a more solid choice at the time. Remember that anything to do with Blender was considered utter garbage by the industry not that long ago. Don't know what state Cycles was in at that time. And why not since? Too much work, too much of a headache? Iray caused a rewrite of Daz Studio to Qt6 and not even that did it. Maybe they do have some deal with NVidia that prevents them from jumping ship but that doesn't concern me. I suspect it's mostly "we have all this Iray stuff now, we need to continue with Iray" and not pull another 3Delight.
This is really amazing. Have you tried rendering any charcters ?
I don't know how many people are using Daz Studio with only a Mac, but it could be a group of people interested in having a rendered capable of taking advantage of Apple GPU (I think Cycle can take advantage of them, but I never tried a render in Blender).
I really like the idea. We could load native Daz stuff into Blender and use it there without transferring it with a bridge.
yah count me in! Especially regarding this:
I actually made a thread on this idea a few days ago. I think with a lot of people using DS4 because of older plug-ins, many are going to want to be able to use DS4 for the forseable future while still being able to use newer GPUs. So another render option for DS4 that can take advantage of new GPUs will be really important I think. Switching back and forth between DS4 and DS6 while possible is not ideal. So I hope you get enough interest to make this profitable.
I don't think DAZ would object, they already allow the Octane render plugin. I think if DAZ did this themselves then users would expect DAZ to resolve any issues with material conversion, if a Cycles renderer is done by a third party then it's not DAZ's problem to convert things to be compatible with it. I would be interested to see this work, and I hope support won't drop off, like it always seems to with other third party render plugins for DAZ.
Might make me switch from PC to Mac.
The Cycles engine in Blender already supports Apple Silicon natively, so one would hope that's somewhere on the drawing board. Though I'm not using DAZ Studio as much as I used to, I'd purchase a Mac-native rendering engine in a heartbeat.
For giggles sure, looks terrible of course.
Yeah that'd be great. Sadly I don't have the Mac that would be needed to get the plugin compiled. So that sounds like more fluffy pink clouds at the moment.
Very interesting, I take it the adoption will be similar to Octane.
Is this a dumb idea? Definitely not! Do we even want Cycles in Daz Studio? Yes please!
This is exactly what I was hoping for in DS6. Iray is so inferior to Cycles.
If you set up Patreon for this, I'd join it without hesitation.
All the best and I hope it's very successful.
I'm eager to use it. :)
Edit: This is what I just posted before seeing this.
"I hate to say it but I'm surprised DS6 still has Iray as it's main render engine. It's really old now and it definitely shows. I quite recently started using Blender 5.1.2 and Cycles is awesome in comparison. It's being developed constantly and Iray is at least 8 years out of date. Is there not a way that DAZ Studio could incorporate Cycles for example? Now that would be great. Being able to do modern rendering within DS6 without having to faff about with bridges, which don't work very well anyway."
So yeah. :)
DS next-major-version, 6 as it happens, was always going to use a new version of Qt since the old version used in DS 4 was no longer suported and was blocking numerous things Daz wanted to do.
Tugpsx,
I don't know if we have numbers to guage the adoption of Octane. I remember trying the Daz-Octane product years ago. It struck me as halfway to what Blender could offer, granted that I exported the scene wholesale out of Daz Studio. But staying within Daz's orbit, Octane was pretty decent. I know a PA, valzheimer, put it to good use. His (or Her?) Artstation page has a few renders: here, here, and here. There's more there, tagged with "Daz Studio" and "Octane Render", dating a few years back.
As to Daz and Cycles, I think that's a great idea, particularly if it can work with AMD gpus. But right now the momentum is driven by the soaring price of computer hardware. Semiconductors, I've heard it said often, is a cyclic industry. Eventually, after manufacturers invest in capex, the shortage will ease and prices will fall. Should Daz continue to improve its iRay tools, will there still be strong reasons for a Cycles alternative by then?
In any case, today, 2026, the pressure is building and a Cycles path looks like a good release valve. Best of luck, bluejaunte! It sounds like a tool a lot of people will benefit from.
Cheers!
Hah, cool thanks
I have it working in DS4 and DS6, don't think I even mentioned that.
Yay! :) Get on Patreon please, because I wanna get on BJ Cycles alpha in DS6 lol.
I'm curious, since I'm not a native Blender user. What would a DAZ plugin inside Blender look like? What is workflow are you thinking of?
Bascially, just using DAZ assets (like Genesis Figures) directly in Blender, without the need of a bridge.
But this is technically not feasable, because Blender is lacking some features which DAZ has and I don't see them coming anytime soon.
However, this will anyway never happens. Even if Blender would have the technical foundation to mimick all DAZ features, DAZ would never migrate to Blender, because they would loose control and stuff like HD morphs and dForce Hair would be accesable to non PAs. It's against their business modell.
Well, to take this step is a bit scary. Once I take people's money there will expectations. If only a handful of people subscribe, I'll still be expected to develop this thing. Let's say 5 people pay 10 bucks a month, how am I really supposed to develop much of anything with that right. Bah... money, such a tired subject. Wish we could all just exist and create.
If you're active in other communities you can maybe help spread the word.
It's not so unusal in the 3D. There are some cases, where this is reality.
However, will anyway not happen, because Blender is lacking important features and DAZ would loose their control about exclusive DAZ features ($$$).
Anyway, a offical Cycles support for DAZ would also be a big step in the right dirction. That would make the materials sync from DAZ to Blender much easier.
Launching pretty much anything is "a bit scary" so you'd have to spend a little time figuring whether 'the pressure' is worth what you might make. And, if make is what you're aiming for, you'd also need to be chatting to folk at 'other places' to see how the wider market might look... Cos you probably already have some idea of sales volumes from other things you do (?)
The Kickstarter model is worth some thought because if you don't make $x initially, then you don't start developing / You've already identified the dev-issue with Patreon ($10 a month from 10 people isn't gonna do you any favours)
Depending on 'tier pricing' then I'll register interest here... but think you should do some serious scoping - timeframes and 'budget' because it's heartening to read you'd be concerned about letting down early backers if patreon-subs dwindle.
Best wishes and keep us 'in the loop'