- This topic has 93 replies, 38 voices, and was last updated 2 years, 3 months ago by
.
-
Posts
-
What I really want to see in Genesis 9... better hand posing options. I'm tired of having to use Mesh grabber to move an assets thumb in a pose because of how terrible the posing options for the hands are.
IF they could release a much better render engine or some other way to improve times, well that would be just great! And motion-blur for rendering animations...
@Gts6 a new render engine would probably mean Iray shaders no longer working or needing conversion, but if it was much faster there would be many who wouldn't care much about that if they could render at twice or more the speed.
Having once being married I realise that "something big" can have various interpretations and what Daz thinks as big may well be meh to others. But as a long-time and chronic DS user I have to admit I'm a little of excited...
G9 looks like the last call for the DAZ Studio ecosystem.
My guess - G9 be another incremental improvement of the G3 mesh. It also appears they're going back to having a unified mesh instead of male/female variants like the first generation of Genesis so double the already insanely slow load times.The place that needed to improve is on the software side. It's years out of date and far too resource intensive for what it does. The still image render quality of former game engines (I guess we should call them general purpose 3D engines now) have almost caught up at much lower performance penalty. Cramming more texture and mesh resolution is only giving diminishing returns when the software engine itself can't use 8K textures efficiently. The 3D engine itself is rather a mess it can't deform bone joints or do collision in a realistic way which is kinda sad when DAZ's entire purpose was posing and rendering 3D human shapes.
On the other hand why bother? At this point improving DAZ Studio incrementally is a lost cause because they've let it's core go completely out of date by years. They need to scrap it and start over from scratch. They'll go back to being a store front selling figures for other render software.
Your last paragraph pretty much sums up my feelings about where DS finds itself today @777s.
By holding on to the old prog and ensuring that legacy content remains working they have little to no prospect of improvement.I'd like to see more (or even something) put into the horrible IK and joint pinning issues that persist. I continue to try (in vain) to create or correct more natural looking animations in DS, and that one is a frustration beyond any comparison to other deficiencies. Literally every other tool in the field has some form of actually effective IK and foot/hand/other pinning. Even Poser has more effective tools than DS. It's bizarre that it continues in new iterations. Having to use addons or other tools to do such a basic thing is a peeve.
And, not to hijack the Genesis 9 thread into a DS5 one, but when are the bugs ever going to stop? I don't mean simple, "my files and chars take forever to load or save", or "my renders take forever and inconsistently so", or "everything kicks to CPU even when I've scaled down all the bloat in my scene". I mean: Crashing on standard usage, hanging on simple processes, lights and MATs and morphs and poses and characters and plugins and and and - not working, at ALL, after an "update/upgrade" to the latest version?
I mean: Come on.I don't know anyone in the "industry", as a user base, who puts up with that. Certainly not at that kind of level, anyway. No one. Except DAZ users. Patient, don't disturb the developer-gods, don't insult the forum mods, DAZ users. I have absolutely refused to update from 4.15 because of it, and all the horror stories that have come along, and continue to crop up even with the vendors themselves doling out patches.
Yeah yeah, I know, "It's free. Whadd'you want? Whadd'you expect? You should be more grateful." I guess I "want", I "expect" for there to be real change and improvement to happen? Faster? Or in some cases, at all? And certainly not an INCREASED dysfunctionality? How selfish?
And one* other question, to anyone who might know: What does it take to create and distribute a Maya or Unreal level render engine? Like, at the code-core? (Watered down to layman's terms if you don't mind, of course - for those of us not versed in development-ease.)
*It begets other questions, of course: Why is it so difficult to get Maya/Other-level realism from DAZ Iray? Is there a way to improve on DAZ Iray at all anymore as it has over time, and if so why hasn't it happened yet? Is it solely because it's a free tool and there's not enough development support? Or some other reason? Is it political? (For example, are other, competing developers paying someone off to keep DAZ Iray inferior? Or other weird consp-theory such stuff?) Or is it more simple, and a tool like DAZ is built on a crap foundation which can only be expected to fall in on itself no matter the scrambling to prevent it? (The Windows Effect) And, since it's free, it will likely subsequently never be "re-made" from scratch, since it would require actual investment to do so? ( @littlepleasures, @777s, & @frank22 Couldn't agree more with your comments on the subject. Think the way you've explained it would/could explain a lot about DAZ motivations, and thereby what might be to come. Probably prescient if not comforting.)
I, too, would be scared of the idea of a new render engine undermining functionality, or having to re-learn a new shader system - one that made all my previous scenes obsolete or unusable without some crazy converter. But maybe I'd sacrifice a lot to get to big-boy realism, too. I dunno.
Anyway, very interested in, and appreciative of, this community's thoughts - especially since DAZ "bashing" (translated: honest criticism) isn't much allowed in the Other Forum.
As for what I'd like to see in a "better G9 release":
Skeletons.
Yeah, I said skeletons. Or muscle/tissue "light-stoppers". Or whatever you want to call them. SSS in DAZ Iray sucks without an underlayer. It needs one.
(Sorry, realize probably a nutjob kind of comment. Stick with me here.) You can try to compensate with Top Coat and depth and scatter tools all day long, artificial "light routers" and dials get you a good distance most of the time, in fact. But (at least with my experiments) if you want to get to true and realistic SSS with DAZ, I've thought for some time that you need to incorporate a sub-skin layer - a real one - to stop/slow/diffuse light as it would (does) with the real human body with muscles, organs and bones. (Edit: I've experimented with shrunken geoshells, and it works.) Otherwise all you're doing is cranking light through a glowy, light-bulb shell of a model. (And, too often, it shows.)Anyway, know it sounds crazy, but if G9 had a sub surface light blocker/diffuser - a real one, or came with capabilities for one, I think you'd start seeing some very interesting renders. Prob never happen, or would be third party at best. But if I knew the first thing about modeling or producing, it would be the first thing I was working on.
(Edit: But I'd settle for an editable geoshell.)@littlepleasures wrote: "If they update significantly to take advantage of new technologies then they risk losing backwards compatibility and alienating “loyal” customers who have spent a lot of money"
If Blender can have backwards compatibility with DAX figures, one would think DAZ could make their own product have it as well.
Did I hear someone snicker "iClone" back there?
🙂
And what's with MetaHuman? When are they gonna get off the pot? That s#1t is taking too long.
Daz/Poser are still king of the hill... even if by "king of the hill" we mean "co-regents of Mediocrity".
As many around here I am not happy at all with the state DAZ is in, But honestly I haven't found an alternative for my needs: Single frame renders of people in environments, with figure posing included and a way to use pre built figures and assets.
If there is one worth looking into I'd really appreciate a hint.Thanks for the link. At first I'd mistaken it for one I'd already bookmarked, but this is a different resource altogether. Looks like good read/resource to compare and contrast.
I do have an approach in some of the tests which has diverged a great deal from standard practices, to be sure. But with it has come a result that has also differed at least as much, and is interesting and entertaining at the least. While "reality" is often in the subjective eye, I can't deny there is a distinct difference, and I continue to experiment with it. Often feel I've reached the ceiling with DAZ, and don't deny would be better served exporting to real tools and learning. (Some of that has begun with Maya, but not yet at productive level, and the Amateur continues to mess with new DAZ projects all over again - mostly from content.) Certainly not alone in that, of course. But, really, for all its flaws, I've gotten some pretty surprising results from DS, in spite of itself.
Suppose that's what tinkering with the dollar-store toys is all about. No matter the frustrations, sometimes shapes the user as much as the result.Now where's that plastic shovel I was using?
Edit: Had to report. In the first 3 pages of the linked guide, can already confirm two known issues I'd found solutions for, one only recently ago - both through arduous trial and error.
transmitted color should always be set to 254 or 0.99 even one digit lower or higher will
ruin the entire sss, this is very important make sure its not set to 255 otherwise your
character will golw and will look transparent, setting should look like image belowI KNEW it!
I've been dinking with 255 and 0 for years, and have just "learned" to stay away. This explains it. It's an actual barrier. I'm not nuts.
(mostly)and dont give it any color or map (unless its a thickness map which changes the sss settings),
giving it color will put blue lines where two surfacess join when rendering with spectral rendering
ON, like belowJust discovered this weeks ago, via forum hint and more trial and error.
This will be an interesting read, indeed.
Daz Studio is just very basic entertainment software. It’s not even remotely close to being anything like industry standard 3D/VFX software. The professional 3D/VFX industry are absolutely not using Daz Studio. Even serious enthusiasts who want to get into animation and render high quality scenes typically “graduate” from DS to far more robust, versatile, and stable software. Daz Studio is the dollar store “paint-by-number” kit of the 3D world. It’s a quick, easy, and relatively inexpensive way to render scenes without having to go through the icky process of actually learning something or investing much effort.
To be fair, with the system builds required to use DAZ beyond a stick figure level, it's not basic entertainment software. The problem it has is there is almost no market for its output - "realistic" rendered still images of clothed human figures - hence the lack of professional users. There never was a market. You can do stop frame animation at the expense of your sanity but it's nothing in the grand scheme of things. I can think of only one area where DAZ is regularly used - visual novel type indie games. I've only ever seen it used professionally in a scant few commercial projects like music album cover art, instructional art and packaging art.
There is something to be said for pre-fab scene and rendering solutions like DAZ for people wanting not to sink a lot of their time into learning the ins and outs of a professional 3D toolkit and concentrate on the creative aspects of their projects. It's too bad that DAZ just isn't capable of scratching that itch for anybody now and largely because they failed to evolve with the creative 3D scene.
The rest of your comment is spot on. The DAZ Studio ecosystem has devolved into a 3D doll house collection for adults with the money to spend on one off items. I suspect most users simply render stuff for their own pleasure and nothing else. Eventually all their 3D content creators will move over to making for other ecosystems that have larger professional user bases and DAZ as a company will go back to doing that too.
@777s - "I suspect most users simply render stuff for their own pleasure and nothing else." You're probably spot on with that Jaybird.
If you look at massive amount of Daz content churned out compared to what ends up in the Daz gallery or elsewhere or even here in our Art Showcase, the ratio is stark, probably 1000 to 1.
It's most likely due to people not feeling confident that their work is good enough, or they're collectors rather than producers, or they're doing naughty pictures with the figures that has few outlets to publish, or all of the above. Not that there's anything inherently wrong with making pictures just for oneself. I do that a lot.
So the issue of how good DS is or isn't is kind of overshadowed by the fact that very very few renders ever see the light of day, at least publicly.So the issue of how good DS is or isn’t is kind of overshadowed by the fact that very very few renders ever see the light of day, at least publicly.
@777s @Frank22 That's me for sure. I do partial renders (or viewport tests) all the time. I rarely finish any. I'm not shy about my work and most of my images are SFW. I just do it for myself because it's a fun side hobby. I enjoy little single-image moments.As others have said, Daz is a niche market. Non-animated renders have limited utility. The only consistent solo-artist market I've seen is visual novels. But the VN market seems to be mostly fetish based--it'd be difficult to stand out as a normie VN dev. Even then, VNs are clearly not mainstream.
I want Daz to innovate as much as anyone, but they have a relatively unique and appealing specialization. Daz is the only software that does exactly what I want. And to my great shame, I've given them plenty of money to keep it up.
New in the store -
Genesis 9 Body Shapes Merchant Resource -30%
$9999.95$6999.97Genesis 9 Head Shapes Merchant Resource -30%
$9999.95$6999.97... and, they're gone -
so back to a Genesis 1 style unisex mesh?
Genesis 9 aye?
IMO they've not even maximized 8.1 yet, but hey... anything to get the masses buying a new library. Always saw Daz3D studio like "Rendering with training wheels", for those who can't be bothered to fully dive headfirst into the likes of Blender, Maya, or ZBrush (like myself). Not that it's a bad thing, just models such as https://www.turbosquid.com/FullPreview/1334834 are out of reach for people like myself.I don't think I'll be dipping into G9 unless they drastically update the micronormals texturing to the detail of something from Flipped Normals, dforce/weight for muscle and breast physics baked into the models, better hair, and so on... which is probably generations away from what they're willing to do. Soooo... maybe I'll have to make the leap to ZBrush much sooner 🙁Edit: I read LivingPixels previous post about an overhaul and DS5 is needed, and I can absolutely see that. Soft body physics is much needed but DS4 just can't handle it properly. I'm curious as to what they'll announce, but probably best to keep expectations in check.
- You must be logged in to reply to this topic.
