0.2.0

[maohgad-web]

v0.2.0 - MGPU Bridge, engine depth and motion vectors

Second packaged build. Drop-in: no compilation needed.

Upgrading from 0.1.0 breaks unless you do two things

Move nvngx_dlssnr.dll into a folder called mgpu beside the add-on, and delete the copy next to the game executable. Some titles load anything named nvngx_*.dll sitting next to their own executable, and then the neural stage binds to the wrong GPU. The panel says INSTALL PROBLEM if it finds one there.

Put mgpu_depth_tap.fx in ReShade's Shaders folder. New, required, and in this download. Without it ReShade never binds depth and the bridge never arms. The add-on enables the technique itself.

What changed

• The game's own depth and motion vectors now reach DLSS 5 on the second card. It derives motion from colour on its own, which is what 0.1.0 ran on. Real depth and velocity give it ground truth instead of an estimate, and the gain is image stability: less jitter and sizzle on faces and fine geometry in motion. A frame that cannot supply them falls back to the derived motion.
• Both cards can have their own DLSS Super Resolution pass, so the second card can run neural at a lower resolution and scale back up - a way for a weaker card to keep up. Ships off; experimental, in the add-on's panel.
• General optimization, and lower latency at higher resolutions. The two cards balance load between them, in either direction, so a speed mismatch no longer builds a backlog. That is what paid for depth and vectors: they still fit inside 0.1.0's one frame window. The latency improvement requires Reflex.

What's in the zip

file	what it is
nvngx.dll_mgpu_bridge.addon64	the add-on
mgpu.ini	its settings, read from beside the add-on
gpu1.ini	empty ReShade preset for the bridge runtime
ReShade2.ini	config for the bridge's own ReShade runtime
reshade-shaders\Shaders\mgpu_depth_tap.fx	keeps ReShade's depth bound. Required
README.txt	install steps and every known limitation
LICENSE	MIT

Where it goes

Unpack into the folder containing the game's .exe; the zip keeps its own structure. In Steam: right-click the game, Manage, Browse local files, then keep going down until you see the .exe - on Unreal titles usually ...\<Game>\Binaries\Win64\.

<the folder with the game .exe>\
    Game.exe                          the game (already there)
    dxgi.dll                          ReShade, ADD-ON-ENABLED build. Yours.
    ReShade.ini                       yours, ReShade makes it on first run
    ReShadePreset.ini                 yours, ReShade makes it on first run
 
    nvngx.dll_mgpu_bridge.addon64     <- from this download
    mgpu.ini                          <- from this download
    gpu1.ini                          <- from this download
    ReShade2.ini                      <- from this download
 
    mgpu\
        nvngx_dlssnr.dll              <- NOT from this download. Yours.
                                         NOT beside the .exe.
 
    reshade-shaders\Shaders\
        mgpu_depth_tap.fx             <- from this download

The mgpu folder is the one step you do by hand. Nothing in this download creates it and the name has to be exactly mgpu.

Do not rename the add-on. The filename must contain the literal substring nvngx.dll - the DLSS-NR snippet checks the file path of whichever module called it. Rename it and the add-on loads, logs normally, and produces nothing.

Nothing from NVIDIA is included in this download, and none of it may be. _nvngx.dll is the driver core and is already resident on a current driver. nvngx_dlssnr.dll is the DLSS-NR snippet, now loaded from the mgpu subfolder - on the development machines it was already in the game folder. This project does not ship it, cannot ship it, and does not document how to obtain it. Do not redistribute NVIDIA binaries.

What you need

• Two RTX 50-series GPUs, with a monitor on each, extended. The neural output is displayed by the card that produced it. Headless works and is meaningfully slower. A single display is not supported: the bridge presents through its own swapchain in its own window. 40-series has been reported working by a contributor on an RTX 4080 SUPER with an RTX 5060 Ti.
• ReShade 6.8.0 or newer, installed with add-on support. The effects-only build never loads .addon64 files and says nothing about it. No effect packages needed.
• A DirectX 12 game. D3D11 and Vulkan do nothing; the add-on stands down and says so in the panel.
• No shader pack and no other add-on.

Before you run it

The bridge can take a while to arm, and that is normal. It waits for a depth buffer and a velocity buffer, and neither exists in a menu. Across the five titles below the wait ran from under two seconds to about forty-five. If it never clears, [R63] and [R78] in ReShade.log say which is missing, and TAP = ABSENT means the shader is not installed.

Do not change resolution, DLSS mode or graphics presets while armed - the stream is armed once against the swapchain as it stands, and rebuilding it can freeze the session. Frame generation is untested. Washed or flat colour? Take tone down in the panel. AutoArm=1 ships; set it to 0 to arm by hand with CTRL+ALT+F10, and before measuring anything.

Titles this build was run on

Each completed a bounded run, arming and exiting cleanly, with engine depth valid on every frame and zero contract mismatches. Logs in docs/0.2.0.

Title	Resolution
Resonance - A Plague Tale Legacy	2560x1440
CONTROL Ultimate Edition (DX12 executable)	2560x1440
DragonSword Awakening	2560x1411
Cyberpunk 2077	3840x2160
The Blood of Dawnwalker	2560x1440

A launch and transport check, not a benchmark. The performance figures in the README were measured on 0.1.0 and have not been re-measured.

Status

Research code, not a product. It creates a second D3D12 device and allocates cross-adapter shared heaps on your hardware - the source is all here and reading it first is a reasonable thing to do.

Windows SmartScreen and some antivirus will flag an unsigned DLL, particularly one named after nvngx.dll. It is unsigned because signing certificates cost money, not because there is anything to hide; build it yourself from source if that matters to you.

Provenance

Built by GitHub Actions, not on a desktop. Everything in this zip came out of a public CI run against public source:

run:     https://github.com/maohgad-web/Neural-coprocessor/actions/runs/34751346468
SHA-256: b53b84b83f393261d974b7abdc438034f0e63f4b6b228a43a727f251b2649ef4

That hash is the digest GitHub recorded for the CI artifact, and this release attaches that artifact unmodified. Verify your download with:

Get-FileHash .\mgpu_bridge.zip -Algorithm SHA256 b53b84b83f393261d974b7abdc438034f0e63f4b6b228a43a727f251b2649ef4

<hr /><em>This discussion was created from the release <a href='https://github.com/maohgad-web/Neural-coprocessor/releases/tag/0.2.0'>0.2.0</a>.</em>

[AdilMachad0]

Hello! I followed the step-by-step guide, but mGPU isn't arming. It shows a "NOT-ARMED" message at the top, and even when I press "CTRL+ALT+F10", the log recognizes the command, but absolutely nothing happens... I believe my dlssnsr.dll file might be the wrong one, or maybe I missed some other detail. Do you know what might be causing this or if anyone has had a similar issue?
​Right at the beginning of the log, it throws a Hook error message, but it doesn't crash or hang on any of the codes mentioned in the Read.me...
​I've tested this with 4 RTX cards (5060, 5060 Ti, 5050, and 5070 Ti).

[maohgad-web]

Hello @AdilMachad0 ! I would love to help you solve this problem. First, could you check if you have an integrated GPU (iGPU) enabled? Your situation sounds very similar to what others with active iGPUs have reported and disabling it fixed it.. Also, another user mentioned that a dlssnsr.dll file obtained from an OptiScaler fork didn't work for them, so that could definitely be the culprit. as for the step-by-step guide is it the video here or the fork one?: https://youtu.be/_K1H3mgcHy4.

If neither of those fixes it, please share your ReShade log! I can debug it specifically to rule out any driver issues.

As a last resort, we can look into your display configuration. The proven setups are either two monitors (each connected to a different GPU in extended mode) or simply leaving only one display connected to the primary GPU and using special K to reroute inputs with a pad.

Thank you so much for your interest and for testing it across all those GPUs!