Skip to content
This repository was archived by the owner on Feb 21, 2026. It is now read-only.

Commit 7cd1d4f

Browse files
committed
Merge branch 'main' of github.com:Line-fr/Vship
2 parents fdfe02c + db8de3d commit 7cd1d4f

File tree

1 file changed

+33
-1
lines changed

1 file changed

+33
-1
lines changed

README.md

Lines changed: 33 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,8 @@
33
A high-performance VapourSynth plugin for GPU-accelerated visual fidelity
44
metrics, focusing on SSIMULACRA2 & Butteraugli.
55

6+
**New : CLI Tool without vapoursynth named FFVship**
7+
68
## Overview
79

810
`vship` provides hardware-accelerated implementations of:
@@ -38,22 +40,37 @@ The steps to build `vship` from source are provided below.
3840

3941
### Prerequisites
4042

43+
For All usecases
4144
- `make`
4245
- `hipcc` (AMD) or `nvcc` (NVIDIA)
46+
47+
Vship Vapoursynth Plugin:
48+
4349
- VapourSynth
4450

51+
FFVship CLI tool:
52+
53+
- ffms2
54+
- zimg
55+
- pkg-config
56+
4557
### Build Instructions
4658

4759
0. Ensure you have all of the dependencies properly installed.
4860

4961
1. Use the appropriate `make` command to build on your GPU
5062

5163
```bash
64+
#Vship Vapoursynth Plugin Build
5265
make buildcuda # for NVIDIA GPUs
5366
make build # for AMD GPUs
67+
68+
#FFVship CLI linux tool build
69+
make buildFFVSHIPcuda #for NVIDIA GPUs
70+
make buildFFVSHIP #for AMD GPUs
5471
```
5572

56-
2. Install the `vship` library:
73+
2. Install the `vship` vapoursynth plugin:
5774

5875
```bash
5976
make install
@@ -64,6 +81,21 @@ make install
6481
6582
## Library Usage
6683

84+
### FFVship
85+
86+
```
87+
usage: ./FFVship [-h] [--source SOURCE] [--encoded ENCODED]
88+
[-m {SSIMULACRA2, Butteraugli}]
89+
[--start start] [--end end] [-e --every every]
90+
[-t THREADS] [-g gpuThreads] [--gpu-id gpu_id]
91+
[--json OUTPUT]
92+
[--list-gpu]
93+
Specific to Butteraugli:
94+
[--intensity-target Intensity(nits)]
95+
```
96+
97+
### Vapoursynth
98+
6799
### Streams
68100

69101
In order to control the performance-to-VRAM trade-off, you may set

0 commit comments

Comments
 (0)