Blender Cycle vs Daz iRay

Deleted member 1121028

Well-Known Member
Dec 28, 2018
1,716
3,307
For those interested,

Here's an update on Z-Cycles progress - a custom Blender build that integrates Iray Uber Shader Nodes for Cycles rendering.

I have done a WIP render comparison between Victoria 8 in Iray vs Victoria 8 in Z-Cycles.

The eyes and eyelashes need work. But the rest is nearly there.
Nice (y)
Specular seems a bit stronger. Is red tints caused by how SSS is handled between Iray and Cycles?
Great job regardless o/
 
Last edited:

recreation

pure evil!
Respected User
Game Developer
Jun 10, 2018
6,326
22,757
For those interested,

Here's an update on Z-Cycles progress - a custom Blender build that integrates Iray Uber Shader Nodes for Cycles rendering.

I have done a WIP render comparison between Victoria 8 in Iray vs Victoria 8 in Z-Cycles.

The eyes and eyelashes need work. But the rest is nearly there.
really nice, I hope you finish this project, would be awesome to have a fully functioning iray sss!
 
  • Like
Reactions: jclave

jclave

New Member
Aug 24, 2019
8
14
Nice (y)
Specular seems a bit stronger. Is red tints caused by how SSS is handled between Iray and Cycles?
Great job regardless o/
Thanks!

The specular difference is because of bump map that was applied in Iray but not in Cycles, as shown in the attached image. Will be fixed before the public release.

I haven't yet figured out the red tint difference in Iray.

The SSS Shader consists of 3 output variables.

I can only say Iray and Z-Cycles use the exact same formula to produce those 3 variables.

But what's not clear are :
* Volumetric algorithm that use the 3 variables between Iray vs Cycles
* Gamma correction between Iray vs Cycles
* etc..
 
Last edited:

Deleted member 1121028

Well-Known Member
Dec 28, 2018
1,716
3,307
Ah shit here we go again lol. Just ten months.

nn7jo.jpg

Hi everyone, hope you enjoy the update. I'm copying a post from Patreon in anticipation of comments/questions regarding the time it took for this update as well as a quick recap of the last 10 months.

TLDR:

"Will the next update take another 10 months!?"

Although this update took 10 months, it's not 10 months worth of content. The overwhelming majority of this time was spent on not only learning the software but figuring out how to implement some of the same functions that already exist in Daz. I discovered that there are many things in Daz I took for granted.

I expect the next and likely final update to be completed by the end of the year.



--------------------------------------------------------------------


Finally, after 10 months we have an update. I'd like to do a quick recap of the journey so far and why it took so long. (I'm lying, it's going to be a lengthy read)



December 2019, I was planning on releasing a Christmas special. After a week of frustrating bugs and errors in Daz, I was fuming and thought that there's got to be a better way. Poor Sophie would randomly look like a shark attack victim because Daz would bug out on simple tasks in the animation timeline. Seriously, how the hell does this happen?





Daz is an excellent program when it comes to doing the core things it was designed to do, which is being able to quickly pose and render beautiful still images with ready to go assets. Animation, simulations, modeling and a plethora of other tools are non-existent or very primitive. Frustrated and annoyed, I reached out to asking if he'd like to explore the option of leaving Daz with me to another software. It's always easier learning new things with someone else. We spent a lot of time learning, bouncing ideas off each other and troubleshooting things together. Thank you, you grumpy cuddly teddy bear <3

First up was Cinema 4D, a fun but pricey software. After nearly a month of learning it, watching tutorials and trying to figure out how to bring our Daz characters in, we realized it's not the best option for what we're trying to do. At least I got to make this cute little submarine from a tutorial, so... worth?





Next up was Blender. Turns out this free software has everything I need to make a great game.

- Sculpting (Make quick adjustments and customization on anything)

- Modeling (Can make anything from scratch or quickly edit and customize existing assets)

- Simulations (cloth, soft body, fluid, fire/explosions, collisions, hair)

- Advanced Character Rigs (No longer using premade pose libraries, everything is posed from scratch to fit the scene)

- Animation (Robust animation tools that make it a fun and enjoyable experience. No more shark attacks!)

- Performance (I can preview my animations in real time rather than guessing what its going to look like and having to render it every time for an accurate preview like in Daz. Projects open in 10 seconds not 5 minutes.)

- Data Exchange (Other software that do certain things better can easily be incorporated into Blender)

Most of these perks of Blender, although nice, are not needed for Personal Trainer. Other than the animation tools, I can fake the rest using Photoshop. However, everything listed is an absolute requirement for my next game, which will not be possible to make in Daz.

I've mentioned this several times already but Personal Trainer was always intended to be a learning project. It's a guinea pig for learning the technical side of things. This is why it has renders from three different engines and has so many inconsistencies in visual quality, writing and doesn't take itself seriously whatsoever. My goal was to learn as much as I can while also hopefully giving you guys an enjoyable experience and a laugh or two.

Please keep in mind that just because Blender is capable of all these things doesn't mean that I'm completely capable just yet. It's a learning process which is reflected in this update. Some animations look just as janky as my old ones and some look pretty damn good. Some renders look like a regression and some look great. I'm still learning and improving. The good news is that the biggest obstacle now is my own knowledge/capabilities rather than fighting the software itself like poor Sophie's Jaws attack I showed earlier. I wanted to get through the huge learning curve now rather than bringing these sorts of inconsistencies into my next game.



Finally, I'd like to get ahead of a question I know many will have.

"Will the next update take another 10 months!?"

Although this update took 10 months, it's not 10 months worth of content. The overwhelming majority of this time was spent on not only learning the software but figuring out how to implement some of the same functions that already exist in Daz. I discovered that there are many things in Daz I took for granted.

I expect the next and likely final update to be completed by the end of the year.



Thank you all for your continued patience and support. I hope you enjoy this update.
 
Sep 12, 2019
85
779
Daz might be better for some things. But Blender does let you do more, and that for free. Diffeomorphic makes the process of importing characters and environments pretty painless. I'm not sure why anyone would fault you for taking time to learn a better tool, especially if it results in better quality games with more animations, more assets (like from Quixel, for example, fluid/smoke simulations, etc.
 

Deleted member 1121028

Well-Known Member
Dec 28, 2018
1,716
3,307
I don't get it, what's your problem?
uh nothing? It's more tongue-in-cheek comment with the topic subject.

Daz might be better for some things. But Blender does let you do more, and that for free. Diffeomorphic makes the process of importing characters and environments pretty painless. I'm not sure why anyone would fault you for taking time to learn a better tool, especially if it results in better quality games with more animations, more assets (like from Quixel, for example, fluid/smoke simulations, etc.
Yeah, I read those very same comment about what? Something like 2 years now.
I guess it's all fine and dandy until you adress the elephant in the room : mostly all those VN failed to deliver on a regular basis, or at all.
 

jclave

New Member
Aug 24, 2019
8
14
mister jclave, we are october now :D
Hah sorry. Still not done.

Had personal issues between August and September which meant no progress could be made.

Also, some of the technical challenges are turning out to be more complex than anticipated.
For example, bump mapping algorithm works very differently between Iray and Cycles. So making bump mapping to look the same has been a very painful experience.

Postponing the release date indefinitely so that people do not get hung up on waiting.

Would like people to think of it as a surprise gift that just appears out of nowhere one day, if it gets completed that is.
 

Deleted member 1121028

Well-Known Member
Dec 28, 2018
1,716
3,307
Hah sorry. Still not done.

Had personal issues between August and September which meant no progress could be made.

Also, some of the technical challenges are turning out to be more complex than anticipated.
For example, bump mapping algorithm works very differently between Iray and Cycles. So making bump mapping to look the same has been a very painful experience.

Postponing the release date indefinitely so that people do not get hung up on waiting.

Would like people to think of it as a surprise gift that just appears out of nowhere one day, if it gets completed that is.
Ay, take your time. Health>*

If I understand things correctly it could be game-changer for what ppl doing here. I tried to spy a bit on Discord but it's not my thing tho. Anyway thanks for the update.
 
  • Like
Reactions: jclave