by Draw Flesh Games

My Star Trek freebies

I'm guessing I'm not the only one to run into this....

I only mess around with DAZ occasionally for my own entertainment and I'm not a expert in any way, shape or form.

I get ideas for scenes and usually I'm able to find assets that will work for what I need.

But sometimes, like the last couple ideas I've had, I end up spending hours digging trying to find anything that could work and come up empty.

For my most recent idea, I was able to find all the smaller props no problem, but the main prop... the actual set... or location if you will.... a grand total of 4 options, and none are usable for what I actually want.🙁

As fun as it is sometimes, it's times like this where I'm happy I only do things for fun and not for profit.

in parameters, if i wanted to put all existing groups under a new tree with new groups above them that i can move morphs into so they wont reorder or reset after i reload daz how would i  do that?

TL;DR: Accord GPU is a free, open beta coordination layer for Windows that stops GPU-heavy creative apps from fighting each other for VRAM. If DAZ Studio, Blender, ComfyUI, or Ollama have ever crashed or OOM'd because something else on the machine grabbed the GPU first, this is built to prevent that.

The primary user base is intended to be creative professionals who frequently run multiple tools on the same system and have to micromanage which apps and jobs are allowed to run on the GPU. A proper system-wide queue for access to the GPU unlocks dramatically higher productivity and keeping the GPU running much more often.

The problem

GPU renderers, image/video generators, and local inference servers all assume they have exclusive ownership of the GPU. Run two of them at once, or even back to back before VRAM actually clears, and you get CUDA OOM crashes, driver resets, or corrupted output. There's no coordination layer between separate applications today, so people end up manually babysitting which app gets to touch the GPU and when.

How it works

A central hub app (Console) runs in the system tray and every plugin talks to it over a local named pipe.

When a component starts GPU work, it writes a small JSON "ticket" to a shared local folder. Every component joins a shared queue and pauses its jobs until the GPU is free.

GPU telemetry (VRAM, utilization, temps) is shown on a convenient appbar you can dock to any edge of any monitor. The queue of jobs appears here so you can see which app is active and what's coming up next.

Coordination is automatic once installed if you use the Accord controls. Kick off a DAZ or Blender render, queue a ComfyUI job or any inference through Ollama, and it waits its turn instead of fighting for the card immediately.

What's available right now (all free during open beta)

• Accord GPU Console - the hub, always installed, manages the queue and GPU telemetry
• Accord for DAZ Studio - a native DAZ plugin that adds a pane you can place wherever you like in your DAZ window setup
• Accord for Blender - a native Blender addon that adds a tab to the N-panel, plus a settings tab
• Accord for ComfyUI - an extension that replaces the Comfy queue with Accord-enabled queueing, and its own settings tab
• Accord for Ollama - a transparent proxy that sits in front of Ollama's API to intercept traffic and help govern behavior like sending inferences into Ollama and when to evict a loaded model
• Accord PRO - additional features that add job history including run time per job, priority rules to put different apps, renders or custom fields first in the queue, Windows power action controls when the queue runs dry, multi-GPU support and more

Roadmap

Accord GPU gains value as it covers more of the applications that heavily utilize your GPU. Which of those ships first is decided by community vote, not internal guesswork - if there's an app you want covered, go add your vote: https://accord-gpu.com/roadmap

Try it

Everything is free and no accounts are required while the open beta runs, PRO features and plugins included. Installer's here: https://accord-gpu.com/download

Accord For DAZ Studio features

• Spawn new instance - DAZ enforces a singleton by default. Accord adds a one-click button to launch a second instance - no manual profile copying.
• Queue across instances - Iray renders on multiple DAZ Studio instances are automatically queued and protected. If another app holds the GPU, DAZ waits its turn. No manual coordination needed.
• Auto-saving - When anything from a single frame to a multi-hour video render completes, save it instantly using your chosen name and codec. Your GPU is free to move on to the next queued item.
• GPU telemetry panel - VRAM usage, temperature, and utilization shown directly in the DAZ Studio panel before and during every render. Preflight protection can warn you if GPU usage is unsafe or the temperature is too high before you start.
• Multiple GPU uses - DAZ Studio calls on the GPU for both Iray renders and dForce simulations. Accord wraps both activities safely.
• Render type scope - Control exactly which render types participate in the queue. Iray is always protected. Viewport and OpenGL renders can opt in when you need them to.
• Render Queue capable - For DAZ Premier members, Render Queue job files can be loaded to the Accord queue. Each item is treated like a single job unit, allowing you to merge your existing workflows into a wider world of GPU coordination.
• Crash recovery - If your application crashes with Render Queue work left, Accord saves the work, restarts the application, and moves the queued items to the new process so you can continue where you left off.
• Render notifications - Windows toast notification and configurable sounds on render complete. Know the moment it finishes even from another room.