Unreal Engine is capable of some pretty amazing things.
It's mind-blowing that that bottom video wasn't even ray traced. This was before raytracing came out for UE. all the textures are from the quixel photoscanned database. It says how important it is to use photo scanned textures instead of photo SHOPPED textures. In my opinion photogrammetry is the single most important factor of photorealism that is lacking in projects.
I am watching this thread with interest. I don’t know much about UE but it is something I have been recently eyeing as a way to bring some of my Daz scenes to life in VR.
If you are confident that your scenes are perfect then leaving them to render overnight is fine. For myself, I am constantly noticing things that could look better so I render, tweak, render again and so on. The next scene is always a follow-on (story) from the previous one so I can't do batch mode for the very reason I stated above - I render and tweak each scene. That is also why render speed is so desirable and why these examples of renders coming out of game engines are encouraging. Shame that DAZ doesn't see it that way but that's just my opinion.
By the way, someone will advise me to use the IRay viewport. I do sometimes but it crashes a lot - especially with my extensive tweaking (shaders, especially). Also, the viewport is just so slow in IRay that I prefer to work in OpenGL. A good compromise would be something like Eevee.
?
I spot render all the time to check lighting and whether things are actually touching etc, there's no need for those to even get close to complete so they take a minute or so. I go refill my water or something. I frequently redo finished scenes when I notice an issue. There is literally no need for real time rendering just for that. I'll also note that tweaking Unreal engine scenes can be every bit as time consuming as doing it in DS, if not worse. Redoing lighting is time consumptive.
Most DS users want to simply create high quality realistic still images using content from the DAZ store. And for that IRay works well. But I can see limiting rendering to one brute force pathtracer coming to a way's end at some point.
Once NVDIA RTX becomes fully developed and practical
it may make its way into Daz studio at some point......
I dont see any reason for Daz to change their render engine.....
IRay is the best choice for still,single portrait/Illustration makers
.........................
No matter to me the native engine in DS, as I personally will not use Daz studio to render delivery finals for the same reasons I dont render finals from iclone pro..........................
Neither program has professional ,large complex scene managment or camera tools
(see the new outliner in Blender 2.81)
nor do they have useful particle and other VFX options
or professional matte compositing rendering options for the types of complex post effects work I need to do in After Effects or Davinci fusion. ..... however YMMV
I have to admit Iray is just too slow to render animations even with the quality turned down, it's just not viable. I wish it was quicker as I think the stills are great but not the animation. I still can't work it out if it's my setting or the implementation of Iray into DS. Maybe it's time to look at other engines for animation !
I'm completely uneducated when it comes to game engines but I get the impression that people stick with IRay in DAZ Studio because it can render all the nice features that seem difficult to export. Such as HD morphs and geografts. If anyone can show me a G8 complete with those features rendered in a game engine, I'll be more than willing to take on a new learning curve.
This is my very personal opinion...people stick with IRay because the store is full of one click solutions and ready to render stuff. And they might have invested in some pretty expensive hardware just to be able to use it.
I'm completely uneducated when it comes to game engines but I get the impression that people stick with IRay in DAZ Studio because it can render all the nice features that seem difficult to export. Such as HD morphs and geografts. If anyone can show me a G8 complete with those features rendered in a game engine, I'll be more than willing to take on a new learning curve.
This is my very personal opinion...people stick with IRay because the store is full of one click solutions and ready to render stuff. And they might have invested in some pretty expensive hardware just to be able to use it.
Of course people stick with what works for them. It is only when it becomes a PITA to use the built-in solution that alternatives are sought. Such is the case with IRay at the moment, with all the problems of CPU fallback and driver issues. I'd probably put up with the slow renders if IRay was stable but it isn't. I'm struggling to fit two characters and a couple of props into 8GB VRAM without it dropping to CPU. Add these problems to the long render times and those alternatives become more and more attractive.
I find Iray very stable, I've rendered 14 characters and props without dropping to CPU.
What is being discussed are tools; tools are used to create a product, which relies on the tool-users having the pre-requisite skills
It isn't the case of one being better, it's a case of the user being more or less skilled with that medium.
With comparable skills available, either solution can provide what you need in terms of visual acuity, realism and and whatever else is deemed appropriate.
What is being discussed are tools; tools are used to create a product, which relies on the tool-users having the pre-requisite skills
It isn't the case of one being better, it's a case of the user being more or less skilled with that medium.
With comparable skills available, either solution can provide what you need in terms of visual acuity, realism and and whatever else is deemed appropriate.
... Both mediums have their strengths.
IMO, play to the tools strengths.
Agreed. users can show quality examples of what game engines can achieve, but it's not plug and play examples. there are usually skilled users doing most of the work outside of the lighting engine used. The majorirty of users in this community can't, or don't want to put in the effort and work to achieve great results outside of DS and Iray which is why DS and Iray are so popular, a user can simply click a few buttons and hit render and get decent results
I am watching this thread with interest. I don’t know much about UE but it is something I have been recently eyeing as a way to bring some of my Daz scenes to life in VR.
You and me both! I have been playing with daz assets in Unity and VR and so far have had mixed results
As others pointed out, game engines still fake how lighting works. They usually go half the way it takes, and fake what's left, while Daz/Iray is more comparable to doing renders in your typical 3D app.
Personally, having been in the games industry for almost 20 years as a lead artist... I'm enjoying Iray simply because natural lighting is easier to achieve if you just feed it proper PBR materials whereas it requires more work on a game engine to achieve the same results (depending on the complexity of the desired lighting of course).
Also of course, with the wrong PBR and light values, anything will look bad in Iray or Unreal & Co.
That aside, I think the real question is whether the typical Daz user really is someone with the latest GPU who cares about physically correct lighting perfection, or whether they're people with normal budgets who just aim to have fun and mess around without aiming for perfection, in which case, Daz would be almost better off partnering with Epic or Unity to implement their technology and at the same time offer a bridge between Daz and Unreal/Unity, giving their store content access to a new market and perhaps even attracting some game engine users into Daz, as Daz is a more focused package and probably easier to use/learn for certain purposes.
I think some of the issue I have with the original post is that we also don't get any info on the process leading to those renders. Working an hour or two and rendering for 30 minutes is quite different from working for a couple weeks and rendering in one second. I'd really like to see a comparison that involves the total worktimes and not just the render time.
Aside from the other excellent points made here I would just add the obvious point that rendering engines are tools but they are not all the same tool -- meaning you could consider Iray a hammer and Unreal a screwdriver... they are not really doing the same job, or working in the same way. Any craftsman will tell you if you want to get a particular task done you should to use the right tool for the job. Iray is definitely a better tool for photorealistic stills, but it is worth noting there are many other render engines that outperform Iray at photorealistic stills... though most of them are not free, and you will also most likely need to do a fair amount of manual work getting proper conversions for the your DAZ products. In Iray that work is all done for you, you just have to click render... so the advantages and disadvantages even out, and IMO opinion Iray makes the most sense for most users.
That said, IMO the version of Iray used in DAZ studio is overkill for animation, especially if you intend to do anything more than a few seconds. And I don't see that changing anytime soon (if ever). That said, even with the recent improvements DAZ is not a great tool for creating animations either IMO -- so it is hard to imagine any serious animator who is going to choose DAZ Studio as the place where they do their work. At least not without a huge amount of improvements by DAZ in the animations workflow/toolset.
Aside from the other excellent points made here I would just add the obvious point that rendering engines are tools but they are not all the same tool -- meaning you could consider Iray a hammer and Unreal a screwdriver... they are not really doing the same job, or working in the same way. Any craftsman will tell you if you want to get a particular task done you should to use the right tool for the job. Iray is definitely a better tool for photorealistic stills, but it is worth noting there are many other render engines that outperform Iray at photorealistic stills... though most of them are not free, and you will also most likely need to do a fair amount of manual work getting proper conversions for the your DAZ products. In Iray that work is all done for you, you just have to click render... so the advantages and disadvantages even out, and IMO opinion Iray makes the most sense for most users.
That said, IMO the version of Iray used in DAZ studio is overkill for animation, especially if you intend to do anything more than a few seconds. And I don't see that changing anytime soon (if ever). That said, even with the recent improvements DAZ is not a great tool for creating animations either IMO -- so it is hard to imagine any serious animator who is going to choose DAZ Studio as the place where they do their work. At least not without a huge amount of improvements by DAZ in the animations workflow/toolset.
As others pointed out, game engines still fake how lighting works. They usually go half the way it takes, and fake what's left, while Daz/Iray is more comparable to doing renders in your typical 3D app.
Personally, having been in the games industry for almost 20 years as a lead artist... I'm enjoying Iray simply because natural lighting is easier to achieve if you just feed it proper PBR materials whereas it requires more work on a game engine to achieve the same results (depending on the complexity of the desired lighting of course).
Also of course, with the wrong PBR and light values, anything will look bad in Iray or Unreal & Co.
That aside, I think the real question is whether the typical Daz user really is someone with the latest GPU who cares about physically correct lighting perfection, or whether they're people with normal budgets who just aim to have fun and mess around without aiming for perfection, in which case, Daz would be almost better off partnering with Epic or Unity to implement their technology and at the same time offer a bridge between Daz and Unreal/Unity, giving their store content access to a new market and perhaps even attracting some game engine users into Daz, as Daz is a more focused package and probably easier to use/learn for certain purposes.
It doesn't matter if they fake lighting.
What matters is the result one gets, not how that was arrived at; it is, perfectly possible, to get realistic results with fake lighting, although I sucked at it.
That aside, I think the real question is whether the typical Daz user really is someone with the latest GPU who cares about physically correct lighting perfection, or whether they're people with normal budgets who just aim to have fun and mess around without aiming for perfection, in which case, Daz would be almost better off partnering with Epic or Unity to implement their technology and at the same time offer a bridge between Daz and Unreal/Unity, giving their store content access to a new market and perhaps even attracting some game engine users into Daz, as Daz is a more focused package and probably easier to use/learn for certain purposes.
I like this argument. I'm definitely in the low-budget, having fun category. I love the results I get from IRay and would not look elsewhere if IRay didn't keep raising the bar on the amount of VRAM required to prevent a scene from falling back to CPU. A real attraction for an alternative engine would be fast - if not real time - rendering. I don't have to be a professional artist to desire that and it should not be beyond the capabilities DAZ but who knows what agreements they have with the likes of NVidia?.
i alsou would love to see a partnership from daz with unreal, it could help a lot in the future, specially in case of "competition" because instead of just competing against each other both could wins with a partnership and daz become more friendly toward game enginers.
and indeed even if a "iray render" with a lot of time can look "more realistic" in many cases, you can still achieve a monster level of realism with unreal, if you proper know how to settup the light, the textures and the prosprocessing, if you really know what you are doing you can achieve really amazing realistic things on par with daz with easy if you have the knowledge and skill to do.
I think everyone who is equally into modern gaming and Daz Studio has encountered that moment when you stare at a game scene and think "man, that beats most Iray renders".
Usually even without raytracing, at 60 frames per second, maybe in 4k. Doesn`t have to be UE, there are plenty of examples.
There is no discussion which one is technically more accurate, but at times games come scary close.
And they can balance out insufficiencies with specific features.
If Daz Studio is supposed to target hobbyists, adding a real time application is the way to go imo.
Tbh. right now I am not sure who it really targets.
I think gamers are a possible market. Steam has probably more screenshots uploaded than Daz has images in the gallery.
Many games tap into the creativity of players (Minecraft, Sims, Conan Exiles, etc.).
Imo a possible "Real Time Daz Studio" should kinda identify itself more as a game than as 3D software.
As something for creative people to have fun with, with as little technical barriers as possible.
Where you can click on a character and a sofa and pick "walk there and sit" as animation.
Of course what we have now has to stay, just saying that branching out might be a good idea in the long run.
They could try go the opposite direction as well, but I am not sure if there is enough room left on that market.
It boggles the mind when people confuse game rendering with 3D graphics rendering. They are not the same in any way shape or form. Do some reading on the subject before posting on this site about how game engines render faster than Iray (or any PBR 3D graphics render engine).
Look at those videos on something more substacial than a 1080p gaming monitor. I have a Dell 8K 32" and they look like dog droppings to me, even at 4K. Low resolution, and bad bad shaders. No SSS, and the lighting resolution ... well ... unatural is the best way to explain it. I know ... you're gonna say, "well most people have or don't have XYZ". That's a bad excuse to produce a low quality image ... really fast. An excellent high-resolution image will look awesome to everyone; a low quality image will look acceptable to most, horrid to a few, and outright unusable to anyone in the pro field.
To say that DAZ is a "hobbyist" version of 3D graphics authoring software does not say that it is not as complex in rendering as "pro-software", and the fact that hobbyists can use it doesn't change the technical aspects of Iray as a render engine. If you want stupid fast (and low quality) rendering use Unreal or any of the other game engines. If you want high resolution images, buy the hardware needed to make Iray render faster. Don't complain about things when you are comparing apples to pimentos. (That analogy is about right).
I think everyone who is equally into modern gaming and Daz Studio has encountered that moment when you stare at a game scene and think "man, that beats most Iray renders".
Usually even without raytracing, at 60 frames per second, maybe in 4k. Doesn`t have to be UE, there are plenty of examples.
There is no discussion which one is technically more accurate, but at times games come scary close.
And they can balance out insufficiencies with specific features.
If Daz Studio is supposed to target hobbyists, adding a real time application is the way to go imo.
Tbh. right now I am not sure who it really targets.
I think gamers are a possible market. Steam has probably more screenshots uploaded than Daz has images in the gallery.
Many games tap into the creativity of players (Minecraft, Sims, Conan Exiles, etc.).
Imo a possible "Real Time Daz Studio" should kinda identify itself more as a game than as 3D software.
As something for creative people to have fun with, with as little technical barriers as possible.
Where you can click on a character and a sofa and pick "walk there and sit" as animation.
Of course what we have now has to stay, just saying that branching out might be a good idea in the long run.
They could try go the opposite direction as well, but I am not sure if there is enough room left on that market.
I'd like to see DAZ do something like that. So you as a script writer for an animation are going to use your keyboard & mouse to direct the characters & write the dialogue. You see a rough draft of it as you create it. When you finish you tell DAZ to say your script and to render the play, including synthesized voices and such. However, at the current time Unity & UE4 are much closer to being able to do that comprehensively then DAZ Studio.
Another thing worth noting is alot of game engine renders incorporate Post FX, which makes the results look much better. Iray has fairly weak Post FX by comparison and they seem to be largely unused by the community, perhaps due to the better possibilities afforded by dedicated photo editors and filters. Of course if you want to go down that rabbit hole then you get rapidly get exposed to the fact that the full Iray capabilities are rarely used by the vast majority of the user base... so the argument is easily made that the quality gap is not so much an issue with the specific render engine but more often the skill of the user who is utilizing it. There is an old saying to this effect: "a poor craftsman blames his tools". That said, I still think the version of Iray in DAZ Studio is not generally well suited to the needs of animation/motion... and the reason this is true is motion covers a multitude of visual sins/defects that a still image would clearly expose. So you can afford to render at much lower quality if you intend to see the image in motion. Even so, it is worth remembering the people who are responsible for those games are highly skilled professionals... they know all the tricks to make something look good and use them shamelessly.
This is the reason I am so invested in education. A proper understanding of the tools can usually lead you to a better final product or, conversely, it can lead to a clear understanding of the need to simply change tools.
As far as I have noticed, Daz scenes are not really optimized. That OPs 30 minute render in the hands of Iray expert ( not me, I believe in brute force like 2080ti ) probably would drop to 15 minutes, and it would looks just as good. Scenes in games on the other hand are very well optimized, since well, they have to run in real time. Of course 15 minutes vs. 1 second is still a huge difference, but it's the same against pretty much any normal render engine. Maybe with Octane or Cycles you can get 13 minutes vs 1 second, but it's still a huge difference. Don't get me wrong, I love game engines, and I'd love to have 1 sec renders with Iray/Octane/Cycles too, but I'm afraid that's never going to happen.
Also this whole conversation is spinning around archviz. Even I can make decent Unity scenes with just environments, but when I bring in characters, then difference to Iray is like a night and day. Maybe there is some special magic shaders for characters I haven't found yet, but difference in quality in human skin shaders is still really big in my opinion. Just like Eevee for Blender. Great engine, but I just can't get as good results as with Cycles or Iray. I've seen videos of UE digital humans, and reallusion is also advertising with their new skin shaders, but I really haven't have time to test those out yet. Maybe those are really good, and close to Iray quality, but my experiments haven't been very successful yet.
There is a real-time version of Iray -- it is just that DAZ Studio does not support it.
Realtime rendering mode was deprecated and subsequently removed entirely from Iray years ago. And it's current equivalent - Iray Interactive - is actually fully supported by current versions of Daz Studio. It's just that very few people are aware of how to activate it. And fewer still are fans of actually using it (since the visual differences between it and Photoreal can be drastic.)
It boggles the mind when people confuse game rendering with 3D graphics rendering. They are not the same in any way shape or form. Do some reading on the subject before posting on this site about how game engines render faster than Iray (or any PBR 3D graphics render engine).
Look at those videos on something more substacial than a 1080p gaming monitor. I have a Dell 8K 32" and they look like dog droppings to me, even at 4K. Low resolution, and bad bad shaders. No SSS, and the lighting resolution ... well ... unatural is the best way to explain it. I know ... you're gonna say, "well most people have or don't have XYZ". That's a bad excuse to produce a low quality image ... really fast. An excellent high-resolution image will look awesome to everyone; a low quality image will look acceptable to most, horrid to a few, and outright unusable to anyone in the pro field.
To say that DAZ is a "hobbyist" version of 3D graphics authoring software does not say that it is not as complex in rendering as "pro-software", and the fact that hobbyists can use it doesn't change the technical aspects of Iray as a render engine. If you want stupid fast (and low quality) rendering use Unreal or any of the other game engines. If you want high resolution images, buy the hardware needed to make Iray render faster. Don't complain about things when you are comparing apples to pimentos. (That analogy is about right).
There is a real-time version of Iray -- it is just that DAZ Studio does not support it.
Realtime rendering mode was deprecated and subsequently removed entirely from Iray years ago. And it's current equivalent - Iray Interactive - is actually fully supported by current versions of Daz Studio. It's just that very few people are aware of how to activate it. And fewer still are fans of actually using it (since the visual differences between it and Photoreal can be drastic.)
It wasn't ever made available in DS due to stability issues - which may indicate why it has now been dropped from Iray entirely.
Comments
I am watching this thread with interest. I don’t know much about UE but it is something I have been recently eyeing as a way to bring some of my Daz scenes to life in VR.
?
I spot render all the time to check lighting and whether things are actually touching etc, there's no need for those to even get close to complete so they take a minute or so. I go refill my water or something. I frequently redo finished scenes when I notice an issue. There is literally no need for real time rendering just for that. I'll also note that tweaking Unreal engine scenes can be every bit as time consuming as doing it in DS, if not worse. Redoing lighting is time consumptive.
I thought that guy was modeled and seriously impressed at first but he's real.
She looks like Björk sort of.
I might consider iRay renders in Unreal Engine someday down the road, if I decide to create 3D renders instead of toon renders.
I have to admit Iray is just too slow to render animations even with the quality turned down, it's just not viable. I wish it was quicker as I think the stills are great but not the animation. I still can't work it out if it's my setting or the implementation of Iray into DS. Maybe it's time to look at other engines for animation !
I render iray animations too but I set the path length to maximum and the iterations low and use the denoiser
well using 4.11
I find Iray very stable, I've rendered 14 characters and props without dropping to CPU.
I did another test of a few engines
my exports don't always match though I got the lace scale wrong in Twinmotion and
the picture frames were all over the place in UE4 and some other props so I plonked them wherever I could without checking the original
I'm always interested in these discussions.
They do have value, but so often miss the point.
What is being discussed are tools; tools are used to create a product, which relies on the tool-users having the pre-requisite skills
It isn't the case of one being better, it's a case of the user being more or less skilled with that medium.
With comparable skills available, either solution can provide what you need in terms of visual acuity, realism and and whatever else is deemed appropriate.
... Both mediums have their strengths.
IMO, play to the tools strengths.
Agreed. users can show quality examples of what game engines can achieve, but it's not plug and play examples. there are usually skilled users doing most of the work outside of the lighting engine used. The majorirty of users in this community can't, or don't want to put in the effort and work to achieve great results outside of DS and Iray which is why DS and Iray are so popular, a user can simply click a few buttons and hit render and get decent results
You and me both! I have been playing with daz assets in Unity and VR and so far have had mixed results
As others pointed out, game engines still fake how lighting works. They usually go half the way it takes, and fake what's left, while Daz/Iray is more comparable to doing renders in your typical 3D app.
Personally, having been in the games industry for almost 20 years as a lead artist... I'm enjoying Iray simply because natural lighting is easier to achieve if you just feed it proper PBR materials whereas it requires more work on a game engine to achieve the same results (depending on the complexity of the desired lighting of course).
Also of course, with the wrong PBR and light values, anything will look bad in Iray or Unreal & Co.
That aside, I think the real question is whether the typical Daz user really is someone with the latest GPU who cares about physically correct lighting perfection, or whether they're people with normal budgets who just aim to have fun and mess around without aiming for perfection, in which case, Daz would be almost better off partnering with Epic or Unity to implement their technology and at the same time offer a bridge between Daz and Unreal/Unity, giving their store content access to a new market and perhaps even attracting some game engine users into Daz, as Daz is a more focused package and probably easier to use/learn for certain purposes.
I think some of the issue I have with the original post is that we also don't get any info on the process leading to those renders. Working an hour or two and rendering for 30 minutes is quite different from working for a couple weeks and rendering in one second. I'd really like to see a comparison that involves the total worktimes and not just the render time.
Aside from the other excellent points made here I would just add the obvious point that rendering engines are tools but they are not all the same tool -- meaning you could consider Iray a hammer and Unreal a screwdriver... they are not really doing the same job, or working in the same way. Any craftsman will tell you if you want to get a particular task done you should to use the right tool for the job. Iray is definitely a better tool for photorealistic stills, but it is worth noting there are many other render engines that outperform Iray at photorealistic stills... though most of them are not free, and you will also most likely need to do a fair amount of manual work getting proper conversions for the your DAZ products. In Iray that work is all done for you, you just have to click render... so the advantages and disadvantages even out, and IMO opinion Iray makes the most sense for most users.
That said, IMO the version of Iray used in DAZ studio is overkill for animation, especially if you intend to do anything more than a few seconds. And I don't see that changing anytime soon (if ever). That said, even with the recent improvements DAZ is not a great tool for creating animations either IMO -- so it is hard to imagine any serious animator who is going to choose DAZ Studio as the place where they do their work. At least not without a huge amount of improvements by DAZ in the animations workflow/toolset.
hey, it's good enough for hollywood: Disney uses Epic's Unreal Engine to render real-time sets in The Mandalorian
It doesn't matter if they fake lighting.
What matters is the result one gets, not how that was arrived at; it is, perfectly possible, to get realistic results with fake lighting, although I sucked at it.
I like this argument. I'm definitely in the low-budget, having fun category. I love the results I get from IRay and would not look elsewhere if IRay didn't keep raising the bar on the amount of VRAM required to prevent a scene from falling back to CPU. A real attraction for an alternative engine would be fast - if not real time - rendering. I don't have to be a professional artist to desire that and it should not be beyond the capabilities DAZ but who knows what agreements they have with the likes of NVidia?.
i alsou would love to see a partnership from daz with unreal, it could help a lot in the future, specially in case of "competition" because instead of just competing against each other both could wins with a partnership and daz become more friendly toward game enginers.
and indeed even if a "iray render" with a lot of time can look "more realistic" in many cases, you can still achieve a monster level of realism with unreal, if you proper know how to settup the light, the textures and the prosprocessing, if you really know what you are doing you can achieve really amazing realistic things on par with daz with easy if you have the knowledge and skill to do.
I think everyone who is equally into modern gaming and Daz Studio has encountered that moment when you stare at a game scene and think "man, that beats most Iray renders".
Usually even without raytracing, at 60 frames per second, maybe in 4k. Doesn`t have to be UE, there are plenty of examples.
There is no discussion which one is technically more accurate, but at times games come scary close.
And they can balance out insufficiencies with specific features.
If Daz Studio is supposed to target hobbyists, adding a real time application is the way to go imo.
Tbh. right now I am not sure who it really targets.
I think gamers are a possible market. Steam has probably more screenshots uploaded than Daz has images in the gallery.
Many games tap into the creativity of players (Minecraft, Sims, Conan Exiles, etc.).
Imo a possible "Real Time Daz Studio" should kinda identify itself more as a game than as 3D software.
As something for creative people to have fun with, with as little technical barriers as possible.
Where you can click on a character and a sofa and pick "walk there and sit" as animation.
Of course what we have now has to stay, just saying that branching out might be a good idea in the long run.
They could try go the opposite direction as well, but I am not sure if there is enough room left on that market.
It boggles the mind when people confuse game rendering with 3D graphics rendering. They are not the same in any way shape or form. Do some reading on the subject before posting on this site about how game engines render faster than Iray (or any PBR 3D graphics render engine).
Look at those videos on something more substacial than a 1080p gaming monitor. I have a Dell 8K 32" and they look like dog droppings to me, even at 4K. Low resolution, and bad bad shaders. No SSS, and the lighting resolution ... well ... unatural is the best way to explain it. I know ... you're gonna say, "well most people have or don't have XYZ". That's a bad excuse to produce a low quality image ... really fast. An excellent high-resolution image will look awesome to everyone; a low quality image will look acceptable to most, horrid to a few, and outright unusable to anyone in the pro field.
To say that DAZ is a "hobbyist" version of 3D graphics authoring software does not say that it is not as complex in rendering as "pro-software", and the fact that hobbyists can use it doesn't change the technical aspects of Iray as a render engine. If you want stupid fast (and low quality) rendering use Unreal or any of the other game engines. If you want high resolution images, buy the hardware needed to make Iray render faster. Don't complain about things when you are comparing apples to pimentos. (That analogy is about right).
I'd like to see DAZ do something like that. So you as a script writer for an animation are going to use your keyboard & mouse to direct the characters & write the dialogue. You see a rough draft of it as you create it. When you finish you tell DAZ to say your script and to render the play, including synthesized voices and such. However, at the current time Unity & UE4 are much closer to being able to do that comprehensively then DAZ Studio.
There is a real-time version of Iray -- it is just that DAZ Studio does not support it.
Another thing worth noting is alot of game engine renders incorporate Post FX, which makes the results look much better. Iray has fairly weak Post FX by comparison and they seem to be largely unused by the community, perhaps due to the better possibilities afforded by dedicated photo editors and filters. Of course if you want to go down that rabbit hole then you get rapidly get exposed to the fact that the full Iray capabilities are rarely used by the vast majority of the user base... so the argument is easily made that the quality gap is not so much an issue with the specific render engine but more often the skill of the user who is utilizing it. There is an old saying to this effect: "a poor craftsman blames his tools". That said, I still think the version of Iray in DAZ Studio is not generally well suited to the needs of animation/motion... and the reason this is true is motion covers a multitude of visual sins/defects that a still image would clearly expose. So you can afford to render at much lower quality if you intend to see the image in motion. Even so, it is worth remembering the people who are responsible for those games are highly skilled professionals... they know all the tricks to make something look good and use them shamelessly.
This is the reason I am so invested in education. A proper understanding of the tools can usually lead you to a better final product or, conversely, it can lead to a clear understanding of the need to simply change tools.
As far as I have noticed, Daz scenes are not really optimized. That OPs 30 minute render in the hands of Iray expert ( not me, I believe in brute force like 2080ti ) probably would drop to 15 minutes, and it would looks just as good. Scenes in games on the other hand are very well optimized, since well, they have to run in real time. Of course 15 minutes vs. 1 second is still a huge difference, but it's the same against pretty much any normal render engine. Maybe with Octane or Cycles you can get 13 minutes vs 1 second, but it's still a huge difference. Don't get me wrong, I love game engines, and I'd love to have 1 sec renders with Iray/Octane/Cycles too, but I'm afraid that's never going to happen.
Also this whole conversation is spinning around archviz. Even I can make decent Unity scenes with just environments, but when I bring in characters, then difference to Iray is like a night and day. Maybe there is some special magic shaders for characters I haven't found yet, but difference in quality in human skin shaders is still really big in my opinion. Just like Eevee for Blender. Great engine, but I just can't get as good results as with Cycles or Iray. I've seen videos of UE digital humans, and reallusion is also advertising with their new skin shaders, but I really haven't have time to test those out yet. Maybe those are really good, and close to Iray quality, but my experiments haven't been very successful yet.
Realtime rendering mode was deprecated and subsequently removed entirely from Iray years ago. And it's current equivalent - Iray Interactive - is actually fully supported by current versions of Daz Studio. It's just that very few people are aware of how to activate it. And fewer still are fans of actually using it (since the visual differences between it and Photoreal can be drastic.)
In general, a good Iray rig will likely be a good gaming rig.
Exactly which videos are talking about? Show me.
It wasn't ever made available in DS due to stability issues - which may indicate why it has now been dropped from Iray entirely.