# Long-Dated Desk (TURBO)

> **Status: the desk is operating on mainnet.** TURBO opened on 19 August
> 2026 and is served from its own application at turbo.netnet.capital.
> Contract addresses are published on [Official Channels](/official-channels)
> and nowhere else.
> **Participation is restricted to persons 18 years of age or older.**
> Dollar figures on this page are the deployed parameters, quoted at recent
> marks where so labeled, and never a promise.

TURBO is the fund's Long-Dated Desk: a standing counter that sells
**Knock-Out Notes**, called **cards** everywhere in the product, on the
tokenized equities the Manager already holds. Cards come in both
directions. A CALL card is a fixed quantity of a stock token's upside
above a financing level **K**: its live value is **units times (mark
minus K)**, quoted from the public Chainlink mark of the tokenized
equity, and at the listing leverage of **25x** its K sits 4% under the
market price. A PUT card is the call's exact mirror, a fixed quantity
of the downside below K: its live value is **units times (K minus
mark)**, and at 25x its K sits 4% above the market price. Either way
the desk buys the card back at the printed formula at any hour of any
day, and a small move in the stock is a large move in the card. The
card **knocks out** the first time the mark touches its K, from below
for a call or from above for a put: it is worth zero there, it pays
nothing, and it never comes back. Every card is fully reserved from the
moment it is sold, a call in the underlying stock token and a put in
USDG, and every payout settles in the reserved asset.

This is not a game of chance and this page does not describe one. There
is no randomness anywhere in the product: no beacon, no draw, no dice.
The referee is the mark and the price is a formula. What the desk prints
instead of odds are the facts that decide every outcome: the price
formula, the one number the card dies at, the leverage at listing, the
statement that **a knock-out pays nothing because the card is
all-or-nothing**, the statement that **one touch of K kills the card
permanently, even if the price recovers a second later**, the fee
schedule, and the session state of the mark. A card is a leveraged
position with an expected life measured in days, and nothing in this
documentation should be read as suggesting that buying one is a way to
make money.

The desk is part of the fund's RWA thesis: tokenized real-world assets
on Robinhood Chain are working inventory, not shelf decoration. The
cover behind every call card is stock the Manager's Sleeve already
holds, the cover behind every put card is USDG in a separate Manager
pot, the prices come from the Chainlink feeds already pinned in the
fund's registry, and every payout is the reserved asset itself.

The consumer application speaks plain English by design. Its board
carries the printed facts on every card and links here. This page is the
controlling description of the mechanics.

## The card

* **A series is a fixed contract.** Each listing names one stock token,
  one financing level K, one quantity of underlying units per card, one
  card count, and one backstop expiry. Each of the four series on the
  launch board carries **100 cards**, and the units per card were set so
  that a card was worth **$5** at the mark the series listed at. All
  cards in a series are identical, and none of those terms can ever
  change once the series is open. The *price* of a card moves every
  block, because it is the position's live value; the *terms* never
  move.
* **The formula.** A call card's value is `units × max(mark − K, 0)`
  in dollars, and a put card's value is `units × max(K − mark, 0)`,
  where the mark is the Chainlink price of the tokenized equity. Buying
  pays that value plus a 1% markup plus the opening fee; cashing out
  returns that value less a 1% markdown, paid in the reserved asset at
  the same mark, the stock token for a call and USDG for a put. There
  is no order book and no counterparty to find. The desk quotes the
  formula, always.
* **Leverage at listing is 25x.** A call's K is set 4% under the market
  price at listing and a put's K is set 4% above it, so a 1% move in
  the stock is roughly a 25% move in the card. The launch board carries
  one 25x call series and one 25x put series per name and no other
  tiers.
* **Leverage decays as the mark moves away from K, and the desk prints
  that.** K is fixed, so a 25x call after a rally, or a 25x put after a
  sell-off, is no longer a 25x position even though the badge says 25X.
  Every card displays its live effective leverage and its live distance
  to K, and a series stops selling new cards once effective leverage
  falls below half the listed figure. That guard closes a misleading
  sale only; cash-out and every other exit are untouched. The Manager
  lists fresh series at current marks as the old ones drift.
* **Expiry is a backstop, not the product.** Series list with an expiry
  of about 30 days, and the four launch series all expire on 18
  September 2026, but a card's expected life is days, and the intended
  exits are cash-out at the desk or the knock-out. A card still alive at
  expiry settles at intrinsic value on the first fresh in-session mark
  at or after the expiry time, paid in the reserved asset.

## A worked example: a $5 card at 25x

The card size on every launch series is $5, so the example is written at
that size. NVDA at $225.00, listed at 25x, so **K, which is also the
bar, is $216.00**, 4% under the market. $5 of card value buys 0.5556
units, which is $125 of notional exposure, and costs about $5.30 all-in
after the $0.05 markup and the $0.25 opening fee.

| NVDA | Card value |
|---|---|
| +4% → $234.00 | $10.00 |
| +2% → $229.50 | $7.50 |
| +1% → $227.25 | $6.25 |
| flat → $225.00 | $5.00 |
| −1% → $222.75 | $3.75 |
| −2% → $220.50 | $2.50 |
| −4% → $216.00 (**K**) | **$0, knocked out, all-or-nothing** |

A 1% move in the stock is a 25% move in the card. NVDA ranges 2% to 4%
on an ordinary day, so the honest expected life of this card is a day or
two. The whole product fits on one line: **your card is worth 0.5556
times (NVDA minus 216), and it dies at 216.**

Cash-out on a call pays in NVDA tokens, not USDG. Closing the +4% row
pays $10.00 less the 1% markdown, which is $9.90, less 5% of the $4.60
of profit over the $5.30 paid, for about $9.67 in NVDA. Closing the −2%
row pays about $2.48 and no performance fee, because losing closes are
never charged.

The 25x put is the same table reflected. Listed at the same $225.00
mark, its **K is $234.00**, 4% above the market, and $5 of value buys
the same 0.5556 units. The card is worth 0.5556 times (234 minus NVDA),
it gains as the stock falls and loses as it rises, it dies the moment
NVDA touches $234.00, and its cash-out and expiry payouts arrive in
USDG rather than the stock token. The put's reservation is `units × K`,
which is $130.00 of USDG behind a $5 card. Every fee row above applies
unchanged.

## The knock-out rule

**One touch of K kills the card permanently.** If any valid Chainlink
round prints at or below a call's K, or at or above a put's K, the
series is knocked out, every card in it is dead, and a knock-out pays
nothing. The price recovering one round later changes nothing. This is
the knock-out feature the buyer is choosing when they buy the card, and
it is stated here rather than discovered later.

Nothing is withheld at the bar. The bar sits exactly at K, so the
card's value at the moment of knock-out is `units × (K − K)`, which is
zero. The card pays nothing because nothing is left, never because the
house kept something. There is no partial outcome, no residual, and no
claim of any kind against a knocked series, by anyone, ever.

Enforcement is permissionless and cannot be waived. Anyone may present
a proof of any round through K since the series opened, and the proof
kills the series whenever it arrives, so a touch that bounces between
keeper runs still counts. The desk applies the same rule to itself: buy
and cash-out both refuse to execute when a recent round has printed
through K, so the desk never sells a card it can already prove dead and
never pays out on one.

While a card is alive, the other half of the rule holds: **you can sell
it back to the desk at what it is worth, less the printed markdown and a
5% cut of any profit, at any hour**. A living card is always quotable
and cash-out can never be halted, by anyone, for any reason. There is no
third state. The card is alive and sellable at the formula, or it is
knocked out and worth nothing.

## The price contains no time value

The printed price is pure intrinsic value, and that is the card's fair
value in the arithmetic sense, not a marketing sentence. A forward
struck at K is worth the mark minus K, and the paths on which the mark
touches K contribute nothing to that value, because the position is
worth exactly zero at the touch. The barrier costs nothing because it
sits where the card is already worthless. The desk therefore sells the
card at what it is worth plus a disclosed markup, with no volatility
model and no hidden margin.

The same argument, run on a short forward struck at K, holds for the
put word for word: the put's remaining value at the moment of touching
K is also exactly zero, so intrinsic is the put's fair value at every
leverage too.

Two real-world gaps between the model and the feed both run in the
holder's favour. The feed prints in discrete rounds, so a touch of K
that recovers between rounds leaves the card alive when continuous
monitoring would have killed it. And a price that gaps through K kills
the card at zero, which is exactly what the card was worth there.

The buyer should read the same fact from the other side: the price
includes **no time value at all**. A card is not an option and is not
priced like one. Holding it costs nothing per day, and it earns nothing
per day. It is a leveraged claim on the next move of the mark, with a
death line 4% away.

## The fees

The desk charges exactly three things, and nothing else:

* **5% of the premium to open.** The fee is 5% of the card's value at
  purchase, not of the notional; at 25x that is 0.2% of the exposure.
  It splits in half, 2.5% to the Manager and 2.5% to the NetNet RWA
  Sleeve described under [Real World Bonds](/rwa-desk), the standard
  split across the fund's desks, swept permissionlessly.
* **A 1% markup on buys and a 1% markdown on cash-outs.** Buys pay 1%
  over the formula value and cash-outs receive 1% under it. Both
  accrue to the house pot
  that covers the cards, not to the fee buckets.
* **5% of net profit on a winning close.** When a cash-out or an expiry
  claim pays out more than the wallet paid the desk for its cards, the
  desk takes 5% of the profit, **never of the principal**. A flat or
  losing close pays no performance fee, and the fee can never turn a
  winning close into a losing one. It is charged in the payout asset at
  the closing mark and carries the same 2.5%/2.5% split.

The profit calculation uses what the closing wallet actually paid the
desk. A card that arrived by transfer was not bought from the desk by
that wallet, so it carries a cost basis of zero and its entire proceeds
count as profit: a transferred card always closes with the flat 5% taken
from its full value. That is the price of keeping transfers completely
free of desk accounting, and it is repeated in the transfer section
below because secondary buyers need to know it.

Every payout on the desk is in the reserved asset: the stock token on a
call card and USDG on a put card. A call holder who wants USDG sells
the token through the normal venues; the application offers a one-tap
sell as a convenience, and that convenience is outside the desk.

## Market hours and the price ladder

Robinhood's stock tokens trade 24/5 and their Chainlink feeds freeze at
the last close between sessions. The desk trades 24/7 anyway, because
equity market hours are a disclosure and never a gate, and it takes its
mark from the best source available at that hour. In plain language, the
ladder is:

1. **In session, on a fresh mark: the Chainlink mark.** Fresh feed, full
   formula, no haircut. Knock-outs are instant on any round through K,
   and this is the only rung on which an expiry ever settles. A feed
   that simply goes quiet mid-session, which the tokenized equities do
   on calm tape, no longer stops the desk trading. The quote steps down
   to the rungs below and carries their printed haircuts, exactly as it
   does overnight and at the weekend.
2. **Off-hours, where the token's own market is deep enough: the
   token's on-chain price.** The tokenized stock keeps trading on-chain
   through the weekend even while the feed sleeps, so where that market
   passes a published depth bar the desk prices cards from a 30-minute
   average of it, with a small haircut against the trader. **NVDA
   passes the bar today. AAPL does not yet**, so AAPL falls to the rung
   below until its pool deepens. On this rung a card can knock out at
   3am on a Sunday, through a deliberate two-step: the breach is
   recorded on-chain, and the kill lands only if the token's market is
   still through K ten minutes later. A single manufactured wick
   therefore cannot kill a card, and anyone, including the holder, can
   clear a recorded breach the moment the price recovers. While a
   breach is pending, buys and cash-outs on that series wait.
3. **Off-hours, otherwise: the frozen last close.** AAPL quotes here at
   launch. The mark is honest about being frozen: the quote applies a
   wider printed haircut against the trader, off-hours purchases are
   capped per session at a slice of the series' remaining capacity, and
   no card can knock out on this rung because the mark cannot move.

**Listing a new series is the one action that still requires market
hours and a fresh in-session mark.** A series' K is fixed off the mark
it lists at, so the desk refuses to open one from a stale or off-hours
price, which would hand the opening buyers a tranche the live world has
already moved past. Every other path runs on whichever rung is serving
at that hour, including buying an existing card, cashing one out,
knock-out enforcement, expiry settlement, and claims.

The haircuts are printed in the quote, not hidden in it. On the live
token-market rung the band on the mark is 0.20%, which at 25x is about
5% of a card's value at the money; on the frozen-close rung it is
0.50%, about 12.5% of value at the money. On every rung the haircut is
capped at 25% of value, so a living card always quotes above zero, no
matter how close to K it trades. Knock-outs and expiry settlements
always read the true mark with no haircut, because a settlement must
never be quoted against the holder.

**The off-hours disclosure cuts both ways, and the desk prints it
rather than pricing around it.** A frozen mark can gap at the next open
in either direction. A buyer on the frozen-close rung may be buying a
card that the live world has already killed, and will find out on the
first fresh round. A seller on the frozen-close rung may be cashing out
below what the re-open would have paid. Every quote, card, and
confirmation names the rung that priced it, a frozen mark is always
shown with its timestamp and the next session open, and a frozen mark
is never rendered as if it were live.

A feed that stops altogether is a different matter from a feed that is
merely quiet, and there the desk fails closed rather than guessing. A
series whose feed produces no valid round for 14 days of in-session
time is voided, and holders refund their net premium in the reserved
asset at the last mark. A refund is not a close and pays no fee.

## Fully reserved, and whose money it is

Every card is covered by its payout asset before it exists. Selling a
call card reserves its full unit quantity of the stock token from a
pre-funded house pot in that asset. Selling a put card reserves `units
× K` of USDG from a separate pre-funded put pot, which covers the
payout in full even at a mark of zero. A purchase the pot cannot cover
is refused at sale rather than owed later. Every payout the desk can
ever make on a card, cash-out at any mark and expiry settlement at any
mark, is arithmetically smaller than its reservation, so the desk
cannot be made insolvent by any price path. When a series knocks out,
the whole reservation returns to its pot at once. The contract's
standing invariant is that the desk's balance in each asset always
covers the pot, every reservation, and every accrued bucket.

Both pots are the Manager's own capital: Sleeve stock the Manager
already holds behind the calls, seeded at about **$15,000 per name** at
launch and quoted at recent marks, and about **$15,000 of USDG per
name** in the put pot. That is $15,000 per name per side, so about
$60,000 of Manager capital stands behind the launch board. **No
treasury money is involved on any path.**
The Treasury, the fund's reserves, and protocol-owned liquidity are
never a counterparty to a card, no code path lets them seed or backstop
either pot, and no leg of any card involves NET at all.

One cost of the put side is stated here rather than buried. Selling
puts is genuine short-downside exposure for the house: put payouts land
exactly when the Manager's own equity book is falling. That exposure is
bounded per card by the reservation, bounded per name by the put pot,
and carried by Manager capital only.

Reservation makes capacity hard, visible, and honest. A call card
reserves 25 times its value in stock and a put card reserves 26 times
its value in USDG, so at the $5 card size a call card locks about $125
of the stock token and a put card locks $130 of USDG. A $15,000 pot
therefore covers about 120 call cards or about 115 put cards per name,
which is roughly $600 of open call premium and roughly $577 of open put
premium. The launch board was listed well inside that. Each of the four
series opened with 100 cards, which is $500 of card value per series
and $2,000 across the whole board. A series' card count is fixed at
listing and does not replenish as cards close or knock out, so capacity
returns to the board when the Manager lists a fresh series rather than
when an old one empties. The board shows the cards remaining on every
series. It is small by construction and the application never pretends
otherwise. Raising it means the Manager seeding more and listing more.

## Cards are tokens, including dead ones

Cards are standard **ERC-1155** tokens, one token id per series, freely
transferable and holdable by any wallet or contract. Every card in
existence was bought from the desk at the printed price against
reserved cover; there is no free mint and never will be one. A card
describes itself on-chain, including its series terms and its state, so
any wallet can render what it is holding.

Three warnings belong to anyone touching a card outside the desk:

* **A dead card stays transferable, and it is worth nothing.** The desk
  cannot burn a knocked card out of a wallet, so worthless ids will
  circulate. **Check the series state on-chain before buying a card
  from anyone but the desk.** A knocked series pays zero on every path,
  to every holder, forever.
* **A transferred card closes with the flat 5% fee.** A card received
  by transfer has a cost basis of zero at the desk, so its full
  proceeds count as profit and the performance fee is 5% of its whole
  value on close.
* **A card is a dying position, not an inventory item.** Escrowing a
  card for a week means possibly handing back something worth zero.
  Anything that holds cards on someone's behalf must treat them as live
  positions.

## The board at turbo.netnet.capital

The TURBO BOARD is its own application at turbo.netnet.capital. The
arcade at play.netnet.capital links to it and holds no desk surface of
its own.
Each row is a live series: the ticker, the 25X badge with the live
effective leverage beside it, the live card price, **the death price K
as a literal number**, a distance-to-K meter, the cards remaining, and
the backstop date. MY CARDS shows live values with one-tap cash-out. A
knocked series renders as a spent card everywhere it can appear, with
its value struck through and its death price shown, and with no
cash-out affordance, because the interface is the main defence against
the dead-card resale problem described above. Market state is a label
on the row, never a gate: no button is disabled by the clock, and every
off-hours quote names the rung and the haircut that priced it.

NVDA and AAPL are the tickers the tokenized stocks identify themselves
by on-chain. The cards are fixed-formula contracts on the
Chainlink marks of those tokens. They are not options, warrants, or
certificates on any company's shares, they are not listed on or cleared
by any exchange, and the desk is not affiliated with, sponsored by, or
endorsed by NVIDIA Corporation, Apple Inc., or Robinhood Markets. The
tokenized stocks carry the issuer and redemption mechanics of their
tokenization, which are not the fund's to control.

## What the desk can never do

* **Touch the Treasury, protocol-owned liquidity, reserves, or
  emissions.** The house pot is Manager capital, and no leg of a card
  involves NET.
* **Sell a card it can prove is dead.** Buy and cash-out refuse to
  execute whenever a recent round has printed through K or a recorded
  off-hours breach is pending.
* **Pay out more than it reserved.** Every payout is arithmetically
  inside the card's reservation at any mark whatsoever, fresh or
  frozen.
* **Block an exit.** The Manager can halt new sales and nothing else.
  Cash-out, knock-out enforcement, expiry settlement, and claims can
  never be halted by anyone, and the knock-out and settlement paths are
  permissionless.
* **Change a live series.** K, the units, and the expiry are immutable
  once a series opens. The fee schedule, the markup and markdown, the haircut caps,
  and the ladder rules are constants of the deployed contract; changing
  them requires a successor deployment, not an in-place edit.
* **Keep a fee it did not disclose.** The three charges above are the
  only ones, and cumulative sweeps equal cumulative fees charged,
  verifiable on-chain.

## Program terms

1. **Operator.** NetNet Capital Management. The house pot is the
   Manager's own capital; reservations are contract-escrowed, and the
   operator cannot block, redirect, or re-price an exit.
2. **Eligibility.** Participants must be **18 years of age or older**.
3. **Instruments.** Cards are fixed-formula contracts on the Chainlink
   marks of tokenized equities trading on Robinhood Chain. They
   reference the token's mark, never the underlying company's shares,
   and imply no exchange listing, registration, or affiliation.
4. **Costs.** 5% of the card price to open, a 1% markup on buys and 1% markdown on cash-outs, and 5% of
   net profit on a winning close, never of principal. Call payouts are
   in the stock token and put payouts are in USDG.
5. **Outcomes.** A card is all-or-nothing. One touch of K kills it
   permanently and a knock-out pays nothing. While it is alive it can
   be cashed out at the printed formula at any hour, less the printed
   markup or markdown and haircut in force.
6. **Amendments.** Series terms are immutable and the desk's constants
   are fixed at deployment. Changes require a successor deployment.

## Parameters

| Parameter | Value at launch |
|---|---|
| Launch board | Four series: NVDA and AAPL, one 25x call series and one 25x put series per name, 100 cards each. SPCX is wired into the desk and its pool is pinned, and it is deliberately not listed until month two |
| Card size | $5 of card value at the listing mark, so about $125 of notional exposure per card; $500 of card value per series and $2,000 across the four |
| Card token | ERC-1155, one token id per series, freely transferable including after a knock-out |
| Leverage at listing | 25x, so K sits at 96% of the mark on a call and 104% on a put |
| Card value | `units × max(mark − K, 0)` on a call, `units × max(K − mark, 0)` on a put, quoted from the Chainlink mark of the tokenized equity |
| The bar | Exactly K on both sides; all-or-nothing; a knock-out pays zero and is permanent |
| Open fee | 5% of premium, split 2.5% Manager / 2.5% Sleeve, sweeps permissionless |
| Spread | 1% each way, accruing to the house pot |
| Performance fee | 5% of net profit on a winning close, never of principal; transferred cards carry zero basis and close at a flat 5% of value |
| Payout asset | The underlying stock token on calls; USDG on puts |
| Put cover | Cash-secured in full: `units × K` of USDG reserved per card from a separate Manager put pot, covering the payout even at a mark of zero |
| Backstop expiry | 30 days at listing, within bounds of 7 days to 6 months; the four launch series expire 18 September 2026; settles at intrinsic on the first fresh in-session mark at or after expiry |
| Sales guard | New sales stop when effective leverage falls below half the listed figure; exits unaffected |
| Off-hours pricing | Token's own 30-minute on-chain average where the pool passes the depth bar (NVDA at launch); frozen last close otherwise (AAPL at launch) |
| Quiet in-session feed | Trading continues on the lower rungs at their printed haircuts rather than stopping; the desk takes the Chainlink mark only while it is fresh |
| New series listing | The one action that requires an open equity session and a fresh in-session mark; buys, cash-outs, knock-outs, settlements, and claims never do |
| Off-hours haircuts | 0.20% band on a live off-hours mark, 0.50% on a frozen one, capped at 25% of card value on every rung; never applied to knock-outs or settlements |
| Off-hours knock-outs | Instant on a fresh Chainlink round in session; ten-minute recorded checkpoint on the token-market rung; impossible on a frozen mark |
| Off-hours purchase cap | 25% of a series' remaining capacity per closed session, frozen-close rung only |
| Feed outage | 14 days of in-session silence voids the series; holders refund net premium in kind at the last mark, fee-free |
| House seed | About $15,000 of Sleeve stock per name behind the calls, plus about $15,000 of USDG per name in the put pot, quoted at recent marks; that covers roughly 120 call cards and 115 put cards per name, against the 100 cards each series was listed with |
| Halt scope | New sales only; no exit can be halted by anyone |
| Owner functions on any fund contract | **None** |

## Risk factors

* **Leverage, stated without decoration.** At 25x a 1% move in the
  stock is roughly a 25% move in the card, and K sits 4% from the
  price at listing, under it for a call and above it for a put. The
  named stocks routinely move that much within a
  day or two, so **the expected life of a card is days, and the common
  outcome of holding one is a knock-out that pays nothing.** The
  maximum loss is everything paid for the card. Buying cards is not a
  way to make money, and nothing here should be read as suggesting it
  is.
* **The knock-out is permanent and total.** One touch of K, at any
  hour, kills the card. A recovery one round later, or one second
  later, does not revive it. Off-hours the token's own market can kill
  a card on a weekend, and a frozen mark can hide a knock-out that the
  first fresh round then delivers.
* **Leverage decay.** As the mark moves away from K, in the card's
  winning direction, its effective leverage falls even while its badge
  still names the listing figure. The live effective leverage and
  distance to K are printed on every card and are the numbers to read.
* **Off-hours pricing risk, in both directions.** Off-hours quotes
  carry printed haircuts, a frozen mark can gap at the open against a
  buyer or a seller, and an off-hours cash-out may pay less than the
  re-open would have. The desk prints the rung and the haircut on
  every quote rather than pricing the gap away.
* **Market risk on the payout.** Call winnings arrive as the tokenized
  stock, which keeps moving after the close of the card, including
  while a holder waits to sell it for USDG. Put winnings arrive as
  USDG.
* **Capacity and the pot are operational.** The house pot is Manager
  capital, the board is small by construction, reservations on sold
  cards are always protected, and available capacity rises and falls
  with the pot without being a commitment.
* **Dead cards circulate.** Cards are freely transferable after death,
  and a knocked card is worth nothing. Anyone buying a card outside
  the desk must check its series state first, and any transferred card
  closes with the flat 5% fee.
* **Tokenization risk.** NVDA and AAPL here are tokenized stocks and
  carry the issuer and redemption mechanics of their tokenization,
  which are not the fund's to control.
* **Smart-contract risk.** The desk is a new contract, in addition to
  the Chainlink and Uniswap infrastructure it reads.

Full fund mechanics are in [The Fund (Mechanism)](/mechanism); the
Sleeve that provides the house pot and receives half of every fee is
described under [Real World Bonds](/rwa-desk); the complete risk
section is [Risk Factors](/risks). Nothing here is investment advice.
