Skip to content

Step 6 / Ethereum & Shared Applications

Different pools use different rules.

Start with what the assets have in common and where the trades are likely to happen.

One stall exchanges two collectibles whose prices move freely. Another exchanges two claims that are each meant to be worth one dollar. The stalls face different problems.

The same is true of trading pools. We will change one choice at a time: the expected price relationship, the mix of assets, the price range, or extra program rules. A newer version number alone does not tell us which design fits.

What relationship does the pool expect?

A constant-product pool can keep quoting across a broad range of relative prices. If one reserve becomes very small, its price rises sharply. That flexibility means much of the deposited money may support price ranges where few trades happen.

Now imagine two tokens, D and E. Each can reliably be returned to its issuer for one dollar. Most exchanges between them might be expected to stay near one D for one E.

A pool designed for this near-equal relationship can offer lower price impact around it. This is the aim of stable-swap. Curve’s StableSwap blends flatter pricing near the expected ratio with a more protective curve when the reserves become uneven.

The exact quote still depends on the reserves, settings, and fees. The design expects a price relationship; it does not guarantee that the issuers can keep their promises.

Suppose E’s backing fails. Outside buyers now offer 70 cents for E, while D stays near one dollar. The two tokens no longer represent equally strong claims.

If the pool offers a favorable trade, people have a reason to put weaker E into it and take stronger D out. Providers can be left with more E. The pricing curve cannot rebuild E’s missing backing.

A weighted pool changes a different choice: how much of the pool’s starting value goes into each asset. Balancer pools can use unequal weights and more than two assets. Their balances and weights work together to set prices.

Use a fictional $1,000 starting allocation. An 80/20 split means $800 of A and $200 of B. Those are shares of value, not counts of tokens.

If A costs $10 and B costs $1, divide each dollar amount by its token price. The allocation is 80 A and 200 B. It is not eight A for every two B.

For comparison, a 50/50 split at those same prices is $500 of each: 50 A and 500 B. The chosen weights change the starting exposure to each asset. They do not freeze prices or promise returns.

A weighted pool does not need the two assets to stay near one dollar as the stable-swap example did. These designs answer different questions.

Reserve A ↑Reserve B →

Quote across a broad price range; inventory makes each additional unit dearer.

Conceptual shapes, not numerical quotes. Curves show reserve tradeoffs; the range view shows where liquidity is active.

Choose each pool design and read the problem it is meant to address. The drawings compare ideas; they do not calculate what a real trade would receive.

The sketches show ideas rather than calculated trade quotes. For each one, name the choice: a broad price range, a near-equal relationship, value weights, a chosen interval, or extra program rules.

Choose where the assets support trades.

A provider might expect most trades to happen within a small price range. Instead of spreading a position across nearly every possible price, the provider can place it within that chosen interval.

This is called concentrated liquidity. Focusing the same assets inside the interval can let a trade move the price less there. That is what “deeper trading” means: the same-sized trade has less effect on the price. The benefit depends on the position being active within its range.

To make the slider easy to read, we label the starting relative price 100. This is a price index: it compares later prices with that starting price. The number 100 is not a dollar amount or a count of tokens.

An index of 80 means the price is 20 percent below its starting level. An index of 120 means it is 20 percent above. The highlighted interval therefore runs from 20 percent below the starting price to 20 percent above it.

Follow the model’s interval from 80 to 120. As the price moves through it, traders change the mix of the two assets. When the price leaves the interval, the position has become entirely one asset and stops earning swap fees while inactive.

The position may become active again if the price returns. Or the provider may move it to another range. Moving it can require transactions, incur costs, and lock in a changed mix of holdings.

A narrow range can become inactive quickly. It can require more attention and expose the provider to a different asset mix. “More fees for the same money” leaves out these conditions.

This feature is mainly about where assets are available for trades. It is not simply a separate magic price formula. Uniswap v3 introduced it as a native feature, and v4 kept it.

80120

Inside the interval: this position can supply active liquidity and earn applicable swap fees.

Illustrative range. Exact boundary behavior uses protocol ticks and position direction. Market risk continues while inactive.

The starting relative price is labeled 100; move the slider below 80, then above 120. Notice when the position is active: at the edges this simplified model shows the one-asset boundary, while exact real boundaries depend on the protocol’s rules.

Add rules around a pool action.

Another design choice is to run extra code before or after a pool action. In Uniswap v4, optional contracts called hooks can do this around swaps or changes in supplied liquidity.

A hook might change fees as conditions change or use custom accounting rules. That makes the pool more flexible. It also means two pools using the same protocol version can behave differently.

Check the particular hook’s code and permissions. A familiar exchange name does not tell you which fees, controls, or safety assumptions apply to every pool using it.

Other AMMs use different curves, outside price inputs, or changing settings. The families in this lesson are a map of major ideas, not a complete list.

The designs can coexist. A new version does not automatically change an older pool, and it does not become the best choice for every trader. Start with the expected assets and trades, then inspect the actual rules.

The idea to keep

Ask what a pool assumes about its assets, where the deposited assets can support trades, and what happens when the assumption fails. A small quoted price impact does not answer all three questions.

Make it yours

A moment to try it.

Take your time. Explain the reason, not only the answer.

1 of 4

Match each design to the choice it changes.

Use examples only—never enter recovery words, keys, account details, or real balances. Loading saved answers…

Loading your saved progress…

Sources & a little more detail

Illustrative stories and example numbers teach the mechanism. They are not forecasts or live market quotes.

Keep the conversation going

Lesson discussion

Share a question, a useful example, or something that clicked.

Loading discussion…