1 Reply by pituz

(2,671 replies, posted in Using SVP)

flowreen91 wrote:

Fixed in trt-rtx-svp-2026-10-04.zip in the same Drive folder. Coming from the 09-24 archive, only SVP 4\rife\vstrt_rtx.dll needs replacing (close SVP and the player first).

If it still crashes, tell me your player and the resolutions of the files.

Hello! Vapoursynth has been updated to version 81, and unfortunately, your plugin is no longer working. Could you please update it to support the new API? Thank you!

Go to forum: Using SVP

2 Reply by pituz

(2,671 replies, posted in Using SVP)

Flowreen91, thank you! It's working properly now — engine building and seeking are significantly faster with TRT-RTX on RTX 4070.

Go to forum: Using SVP

3 Reply by pituz

(2,671 replies, posted in Using SVP)

flowreen91 wrote:

Fix (in the updated archive): constant-fold that part of the graph for the
padded resolution before building the engine. After that, TensorRT-RTX and
TensorRT output is bit identical. Folding takes 0.3 s, once per model and
resolution. The engine is already cached per resolution, so the folded model
sits next to it.

Thanks for your response. I really appreciate this. But after seeking errors still occur.
https://i.ibb.co/CKctVrWF/1.png

Go to forum: Using SVP

4 Reply by pituz

(2,671 replies, posted in Using SVP)

flowreen91 wrote:

If you want to apply TensorRT-RTX, do this:
1. Close SVP.
2. Make a backup of your "SVP 4" folder.
3. Copy the "SVP 4" folder from the archive on top of yours, overwrite all.
4. Start SVP and use RIFE as usual.

Thanks for response! But I've got error after replacing files.
https://i.ibb.co/Jj2KgBRM/1.png

Go to forum: Using SVP

5 Reply by pituz

(2,671 replies, posted in Using SVP)

flowreen91 wrote:

SVP's script generator already has the rife_rtx hook that loads vstrt_rtx.dll, it just needs the updated rife package (vstrt_rtx.dll + tensorrt_rtx DLLs + newer vsmlrt.py).

How can I properly use rife_rtx?
Here is what I have done so far:
Placed tensorrt_onnxparser_rtx_1_6.dll, tensorrt_rtx_1_6.dll, and tensorrt_rtx.exe into SVP 4\rife\vsmlrt-cuda.
Placed vstrt_rtx.dll and the updated vsmlrt.py into SVP 4\rife.
However, it doesn't seem to work: SVP still uses the regular TensorRT backend (resulting in slow engine compilation) instead of TensorRT-RTX.
Could someone explain how to correctly trigger or activate rife_rtx in SVP 4? Thanks!

Go to forum: Using SVP