Anime Encode

Appearance

Usage Guide ​

Complete guide to using the Anime FFmpeg Re-Encoder notebook.

Workflow Overview ​

mermaid
graph LR
A[Configure] --> B[Download]
B --> C[Encode]
C --> D[Upload to Drive]
D --> E[Auto Shutdown]

style A fill:#2196F3
style B fill:#4CAF50
style C fill:#FF9800
style D fill:#9C27B0
style E fill:#F44336

Basic Usage ​

1. Open and Configure ​

  1. Open the notebook in Google Colab
  2. Mount Google Drive (Cell 1)
  3. Configure settings (Cell 2)

2. Set Download URL ​

python
DOWNLOAD_URL = "https://nyaa.si/download/1234567.torrent"

Supported formats:

  • Torrent files (.torrent)
  • Magnet links (magnet:?xt=...)
  • Direct HTTP/HTTPS links
  • FTP links

3. Choose Profile ​

python
SELECTED_PROFILE = "Balanced 1080p"  # Change as needed

4. Run Workflow ​

Execute the main workflow cell and wait for completion.

Advanced Usage ​

Batch Processing ​

Process multiple files at once:

python
# Download season pack
DOWNLOAD_URL = "https://nyaa.si/download/season-pack.torrent"

# All episodes will be processed automatically

The workflow will:

  1. Download all files
  2. Find all video files in temp folder
  3. Encode each one sequentially
  4. Upload all to Drive

Manual File Placement ​

Skip download step by placing files manually:

python
# Disable download
ENABLE_DOWNLOAD = False
ENABLE_ENCODE = True

# Place your .mkv/.mp4 files in:
# /content/temp_downloads/

# Workflow will encode the latest file

Re-encode Existing Files ​

Already have files in temp folder?

python
# Skip download
ENABLE_DOWNLOAD = False

# Find and encode latest video
latest_video = find_latest_video()
if latest_video:
encoded_path = encode_video(latest_video, profile, USE_GPU)

Custom Output Location ​

Change where encoded files go:

python
DRIVE_OUTPUT_FOLDER = "/content/drive/MyDrive/Custom/Path"
os.makedirs(DRIVE_OUTPUT_FOLDER, exist_ok=True)

Workflow Control ​

Enable/Disable Steps ​

Control which steps run:

python
ENABLE_DOWNLOAD = True        # Download step
ENABLE_ENCODE = True          # Encoding step
ENABLE_MOVE_TO_DRIVE = True   # Upload step
ENABLE_AUTO_SHUTDOWN = False  # Auto shutdown

Example scenarios:

Download Only:

python
ENABLE_DOWNLOAD = True
ENABLE_ENCODE = False
ENABLE_MOVE_TO_DRIVE = False

Encode Only:

python
ENABLE_DOWNLOAD = False
ENABLE_ENCODE = True
ENABLE_MOVE_TO_DRIVE = True

Re-upload Existing:

python
ENABLE_DOWNLOAD = False
ENABLE_ENCODE = False
ENABLE_MOVE_TO_DRIVE = True

Auto Shutdown ​

Save Colab compute units:

python
ENABLE_AUTO_SHUTDOWN = True
SHUTDOWN_DELAY = 10  # Seconds before shutdown

When enabled, the runtime will:

  1. Verify file exists in Drive
  2. Display file size and location
  3. Wait for delay period
  4. Disconnect runtime

Auto Shutdown

Make sure files are in Drive before enabling! Check output logs to verify upload success.

Download Options ​

aria2 Configuration ​

Fine-tune download behavior:

python
ARIA2_CONCURRENT = 16     # Connections per server
ARIA2_SPLIT = 16          # Connections per file
ARIA2_MAX_JOBS = 5        # Parallel downloads
ARIA2_SEED_TIME = 0       # Torrent seeding time (seconds)
ARIA2_BT_STOP_TIMEOUT = 300  # BT timeout

High-speed connection:

python
ARIA2_CONCURRENT = 32
ARIA2_SPLIT = 32

Slow connection:

python
ARIA2_CONCURRENT = 8
ARIA2_SPLIT = 8

Torrent-Specific Settings ​

For torrents, you can adjust:

python
ARIA2_SEED_TIME = 3600  # Seed for 1 hour after download
ARIA2_BT_STOP_TIMEOUT = 600  # Wait longer for peers

Seeding Etiquette

Consider seeding for a while to help the community. Set ARIA2_SEED_TIME to non-zero value.

Encoding Options ​

GPU vs CPU ​

GPU Encoding (NVENC):

python
USE_GPU = True

Pros:

  • 5-10x faster
  • Low CPU usage
  • Good quality
  • Free GPU on Colab

Cons:

  • Slightly larger files
  • Fixed presets
  • GPU must be available

CPU Encoding (x265):

python
USE_GPU = False

Pros:

  • Best quality
  • Better compression
  • More control
  • Works without GPU

Cons:

  • Much slower (0.5-1.5x realtime)
  • High CPU usage
  • Takes longer

When to Use CPU ​

Use CPU encoding when:

  • Maximum quality needed
  • Encoding for archival
  • GPU not available
  • You have time to wait

When to Use GPU ​

Use GPU encoding when:

  • Speed is priority
  • Good quality is sufficient
  • Batch processing many files
  • Using Colab Free tier efficiently

Monitoring Progress ​

Download Progress ​

[#1 50MiB/350MiB(14%) CN:16 DL:15MiB ETA:20s]
  • 50MiB/350MiB: Downloaded / Total
  • 14%: Progress
  • CN:16: Active connections
  • DL:15MiB: Download speed
  • ETA:20s: Estimated time

Encoding Progress ​

frame= 5234 fps=182 q=24.0 size=145MiB time=00:03:38 bitrate=5432kbits/s speed=7.6x
  • frame=5234: Frames processed
  • fps=182: Processing speed
  • q=24.0: Quality target
  • size=145MiB: Current output size
  • time=00:03:38: Video duration processed
  • speed=7.6x: Real-time multiplier

Upload Progress ​

Successfully moved to Google Drive
-rw------- 1 root root 682M Jan 25 12:34 output.mkv

Shows file size and timestamp.

File Naming ​

Input Files ​

Supported:

  • [Group] Anime Name - 01 (1080p).mkv
  • Anime.Name.S01E01.1080p.mkv
  • anime_name_01.mp4
  • Any .mkv, .mp4, .webm

Output Files ​

Automatically adds quality suffix:

Input:  anime_name_01.mkv
Output: anime_name_01_1080p.mkv
anime_name_01_720p.mkv  (if 720p profile)

Storage Management ​

Check Available Space ​

Before encoding:

python
!df -h /content/drive
!du -sh /content/temp_downloads

Cleanup Temp Files ​

After encoding:

python
!rm -rf /content/temp_downloads/*

Drive Storage ​

Monitor Drive usage:

python
from shutil import disk_usage

total, used, free = disk_usage("/content/drive")
print(f"Drive Free: {free / (1024**3):.2f} GB")

Tips & Best Practices ​

1. Test First ​

Always test with a single short video before batch processing:

python
# Test with 5-10 minute video
# Verify settings
# Check output quality
# Then proceed with full episodes

2. Keep Runtime Alive ​

Colab may disconnect if idle:

  • Keep browser tab active
  • Check progress periodically
  • Use Colab Pro for longer runtimes

3. Optimize for Your Use ​

Match profile to use case:

  • Streaming: Balanced 1080p
  • Mobile: Mobile 480p
  • Archive: High Quality 1080p
  • Space-limited: Small 720p

4. Batch Smartly ​

Don't process too many at once:

  • Free tier: 3-5 episodes max
  • Colab Pro: 10-12 episodes
  • Monitor runtime limits

5. Verify Output ​

Always check encoded file:

python
!mediainfo /path/to/encoded.mkv

Look for:

  • Correct resolution
  • HEVC codec
  • 10-bit color
  • Reasonable file size

Common Workflows ​

Weekly Episode Encoding ​

python
# Configure once
SELECTED_PROFILE = "Balanced 1080p"
USE_GPU = True
ENABLE_AUTO_SHUTDOWN = True

# Each week:
# 1. Update DOWNLOAD_URL
# 2. Run workflow cell
# 3. Wait for completion

Season Batch Processing ​

python
# Download season pack
DOWNLOAD_URL = "season-pack.torrent"
ENABLE_AUTO_SHUTDOWN = False  # Process all before shutdown

# Process all episodes
# Monitor progress
# Manually shutdown when done

Quality Testing ​

python
# Test different profiles
for profile_name in ["Balanced 1080p", "Small 720p", "Mobile 480p"]:
SELECTED_PROFILE = profile_name
# Encode same clip
# Compare results

Next Steps ​

Released under the MIT License.

Copyright © 2025 syqrel