This is a minimalist, client side simulator UI for stress testing layered Bitcoin payment designs. It keeps all the original features and removes every preset except one named VISA?. Use it to show whether a VISA scale scenario can be supported by a Lightning centered, multi-layer stack.

• Sliders for population, activity, LN touch, graph envelope, and enforcement windows
• Live derived metrics
  • HTLC demand per second
  • Total HTLC slots and effective capacity per second
  • Slot utilization percent
  • L1 backstop load per day
  • Unsafe region flags for LN CSV and key layer CLTV windows
• Two charts drawn with canvas
  • HTLC demand vs capacity
  • L1 backlog vs enforcement window
• Shareable URL. State is embedded in the hash as Base64 JSON
• Export and Copy of the preset JSON for use in your simulator
• Only the VISA? preset is present

Everything runs in the browser. No build step, no dependencies.

Open the page locally or publish on GitHub Pages.

1. Download the repo ZIP and unzip it.
2. Open index.html in your browser.

1. Create a new repository on GitHub, for example visa-pages-site.
2. Upload index.html and .nojekyll from this folder.
3. Go to Settings -> Pages.
4. Under Build and deployment set Source to Deploy from a branch.
5. Choose Branch main and folder / (root). Save.
6. Wait for deploy. Your site will be at https://<your-username>.github.io/<repo-name>/.

Optional: add a CNAME file for a custom domain.

The single preset is called VISA?. Its values initialize the UI. You can tweak them at runtime.

{
  "id": "visa",
  "name": "VISA?",
  "values": {
    "users_millions": 100,
    "share_self_ln_pct": 33,
    "share_key_layer_pct": 25,
    "share_custodial_pct": 42,
    "payments_per_active_user_per_day": 1.5,
    "active_users_daily_pct": 60,
    "self_ln_touch_pct": 100,
    "self_ln_aggregation_ratio": 1,
    "key_layer_touch_ln_pct": 80,
    "key_layer_user_to_ln_aggregation_ratio": 8,
    "custodial_touch_ln_pct": 80,
    "custodial_aggregation_ratio": 200,
    "avg_path_hops": 4,
    "mpp_splits_per_payment": 6,
    "avg_inflight_seconds": 60,
    "network_routers_count": 300,
    "channels_per_router_avg": 150,
    "htlc_slots_per_channel": 211,
    "usable_slot_fraction_pct": 50,
    "min_htlc_sats": 1000,
    "self_ln_forced_closure_pct": 5,
    "key_layer_backstop_users_per_tx": 200,
    "key_layer_backstop_users_pct": 5,
    "l1_capacity_tx_per_day": 150000,
    "avg_tx_per_ln_channel_closure": 4,
    "user_channels_per_self_ln_user": 1,
    "self_ln_enforcement_window_days": 14,
    "key_layer_backstop_window_days": 30,
    "jamming_target_slot_pct": 50
  }
}

• Population and Activity
  • Users (millions), daily active percent, payments per active user
• LN demand mapping
  • Fractions that actually touch LN
  • Aggregation ratios that compress many end user payments into one HTLC
• LN graph envelope
  • Hops, MPP splits, average in flight seconds
  • Routers, channels per router, HTLC slots per channel
  • Usable slot fraction for safety headroom
• Enforcement and shock
  • Forced closure rate, key layer backstop users per transaction
  • L1 capacity per day and transaction cost per closure
• Enforcement windows
  • LN CSV window in days
  • Key layer CLTV backstop window in days
  • Jamming target percent if you want to simulate load on slots

The math is intentionally lightweight so it is easy to verify and port to your simulator.

• Map user activity to total payments per day
• Select the fractions that touch LN for self custodial, key layer, and custodial cohorts
• Apply aggregation ratios to convert payments into HTLC attempts
• HTLC demand per second is HTLC per day divided by 86400
• Capacity per second is effective slots divided by average in flight seconds
  Effective slots = routers * channels per router * htlc slots per channel * usable fraction
• Utilization is demand per second divided by capacity per second
• L1 backstop load per day = key layer backstop users divided by users per tx + forced LN closures multiplied by transactions per closure
• Backlog and window plots compare L1 load per day vs assumed L1 throughput per day and simulate accumulation over the window

These formulas give order of magnitude intuition and are not a protocol implementation.

• Shareable URL: click Copy Shareable Link. The current state is in the URL hash as Base64 JSON.
• Export JSON: click Export JSON, then Copy JSON. Paste into your simulator as a preset payload.

Everything is one file, index.html.

• Change default values by editing the presetVisa.values object
• Change slider ranges by editing the schema array
• To add more presets later, add more objects and a selector. The current build intentionally keeps a single preset

• If copy to clipboard fails, the page falls back to a hidden textarea and a user gesture copy
• If charts do not draw, ensure your browser allows canvas in a local file context
• If GitHub Pages does not serve the index, make sure .nojekyll is present and Pages is set to use the root of the main branch

MIT. Use it, fork it, change it.

Authored for simulation and communication about LN and layered payment scaling limits.