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
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
Use the ActivePose tool. It's not as good as the IK tools found in other programs, but it's marginally better at keeping pinned limbs in place over the normal transform tools. Just pin the arms and legs and drag the head and torso to create slight movement.
I can change the speed and animation position in Unreal easily enough to stagger them around, reverse and add as layers in sequencer and bake too
I can do a lot of this in Carrara also and iClone but sadly not DAZ studio but its using all my VRAM anyway and never reaches 1%convergence so that is impossible even if I could
OK these are redistributable so if I am bored sometime I will convert a few them to cr2 in Poser11 and save General weight converted dufs in DAZ
https://github.com/microsoft/Microsoft-Rocketbox
they are very lowpoly, you would need to manualy rotate joints to pose as I certainly would not be adding limits to them all if I do this
no promises and if ShareCG gives me trouble uploading I will likely give up!
I used Alienator to swap out my G8M
I did one to see if anyone wants it
Licensing included in readme so you can confirm it is indeed redistributable
https://sharecg.com/v/98537/view/21/DAZ-Studio/Rocketbox-Business-man1#
Thanks Wendy! I installed character you converted into DS and he looks great. Can a bvh or aninblock be applied to the character?
no aniblock sadly unless you make them, BVH might work if retargetted to that character elsewhere
these are meant to just be seat fillers posed by rotating bones and maybe slight movements with puppeteer
can be saved as props after posing too
gets too complicated with EULAS fast if I start creating animations, I cannot redistribute many, the Rokoko plugin in Blender might work
I can animate them in iClone and Unreal but cannot redistribute those animations
Here's the lowest Res figures
Thank you! That's exactly what I wanted to know.
As has already been pointed out, the current figures at base resolution have fewer polygons than Victoria 4 etc. (roughly a quarter) and more distant figures don't need all, or beyond a certain point any, textures so map sizes are not really relevant.
G8Female has fewer polygons than V4, but what about everything that goes along with G8Female such as hair models, clothing models, shoes, etc. Are those lower resolution than the same type of items made for V4?
G8 Female can be made to wear all the clothes, hair etc that V4 can, so I don't see how this is an advantage to V4.
The main advantage of V4 over G8F is that she likely has a lot less morphs, and so will use less RAM than G8F. However if you turn the finally posed G8F into a prop (which you should before instancing) than that one V4 advantage goes away.
A while back. I did some testing just for the fun of it, took the G8F textures down to 25% of the originals.
Geometry memory consumption: 2.899 MiB (device 0), 0.000 B (host)
Texture memory consumption: 9.204 MiB for 12 bitmaps (device 0)
That's 12MB's for a fully functional G8F with clothing and hair
What I did was, took the normal Basic Genesis 8 Female, removed the lashes, all the other texture maps, but the diffuse one (used Shinteo's), set the character and the clothing to base resolution, SubD to 1 and set all the parts of the figure that are covered by clothing "Not Visible"... Forgot the mouth though, so that's still another place to reduce the load.
The base resolution G8 is already quite light on the resources, it is the SubD, the textures and texture maps on the figure and everything fitted to it that are adding the load sometimes beyond the capabilities of some systems.
Instancing can also be used to add duplicate figures without adding the load.
A trick I have done in the past that works for stage scenes...
Make an outline of a head and shoulders. Texture it all black. Place in the extreme seats; right and left edges, the mid-front.
Make an appoximation of a human profile with primitives. Texture these all black. Place these in the seats closer to the camera, but not the focus.
This leaves just the well lit and close to the camera seats to deal with. The front row, the seats directly in front of the camera.
If most of the light is coming off the stage, the outlines should appear as figures as long as you don't have strong lights coming from behind the camera.
The use of Depth of Field can also reduce the noticability of the "tricks."
Click Video
messing around in several apps, each with their own strengths and weaknesses
Wow, that's really great, Wendy!
That's incredible. What was the software you used.
mostly Twinmotion and some iClone in that one but I since rendered more with Unreal engine, posed people props mixed with posed G8F in MFD and Rocketbox men with animations
and some animated mostly Rocketbox performers playing Protozoon instruments with puppeteer animations of his poses FBX exported and made into iMotions,
I work with iClone 6 so lots of round trips no live links for me
would have done a lot more but those curtains took over 24 hours to apply and save Apex cloth to!
those Rocketbox Avatars could make the trip back to DAZ via Poser with animations but simply could not be bothered and hate to think how long it would take
Unreal renders themselves quite fast just all the work getting there
and then mostly just waiting looking at save bars rather than actual work
this also has Twinmotion shots and a bit of Carrara Octane near begining (the spinning shot)
this is my final one
click video
I'm really impressed! I struggle just to do a still picture and here you're creating a whole animated theatre.
There are several ways to do it, reducing textures is one of them. Even if you reduce maps, removing normal maps out from middle to background figures is important as their affect is lost at a certain distance.
RSSC 01 works well with forground characters where the textures are needed with a few simple presets for background characters. It also has presets for stripping all but the diffuse map out.
RSSC 02 - https://www.daz3d.com/mmx-resource-saver-shaders-collection-2-for-iray - is best suited for middle ground to background characters and hair. Since they would all be using the same textures for skin and hair, minus the transmaps, it would greatly reduce resource use.
Its also important to note that you do NOT have to use Genesis figures at high resolution. The base res only uses about 1MB of vram for the geometry for any genesis figure.
It is that final figure that always makes me baffled why people think they need to use older figures (ie low poly) for background and medium level figures in the scene. 1 MB really is nothing, so using a figure like Loretta will reduce that to maybe 300K. Unless you want 100s of such (non-instanced) figures in the scene, it will not add up to a significant difference in total resource use. For a typically audience like scene, anywhere from 10-40 different figures are needed, depending on how blurred via DOF, or dark via lighting they will be seen. These 10-40 will then be instanced over all the chairs to create larger audiences. That means the VRAM usage for just the geometry is from 10-40 MB for a Genesis figure, or 3-12 MB for a 5K figure like loretta, which is simply not relevant. What really matters is optimising the figures textures, clothing and hair.
Whichever figure is chosen, the final figure should be reduced to a prop once it is dressed and posed. This will drastically reduce the amount of RAM needed.
Havos, the problem I think with instancing some many figures for an audience is how to animate them and not having the all do the the same motions, at the same time. At least that is my issue as an animator.