Carbon DEX Twitter Spaces #1: Top Takeaways
Following the announcement last week of Carbon, a Twitter Spaces was conducted to introduce the new protocol. In this post, we’ll cover the top takeaways from the call.
If you want to learn more about Carbon, see the:
Before discussing takeaways from the Twitter Spaces, let’s briefly summarize what makes Carbon unique. In short, all AMMs today are implemented using a single “bonding curve” that defines the buying and selling of an asset in both directions. In recent months, we uncovered a new way to implement on-chain liquidity where individual liquidity positions can be simultaneously governed by two distinct bonding curves, one for buying and one for selling. Each curve executes orders irreversibly at a fixed price or within a given price range, and can be updated in a highly gas-efficient manner. This “asymmetric” and adjustable concentrated liquidity structure gives users an unprecedented level of precision to personalize on-chain trading and market-making strategies. It also fundamentally changes the status quo in existing AMMs: There is no impermanent loss in Carbon, in the sense that each strategy is not a buy-and-hold liquidity position, but rather the expression of a particular trading view.
Now onto the takeaways raised in the call. Below we’re generally paraphrasing quotes from the call.
1. Carbon aims to bring CEX capabilities to a DEX, plus “something more” (7:55)
There is still this demand to have an order book-like structure in DeFi. It’s something that a lot of sophisticated players have been asking for for a while. People have an expectation to be able to do in DeFi, on decentralized exchanges, what they do on centralized exchanges; but that’s not enough. You can’t just say fine, here’s your order book and be happy with it. You can’t just offer what competition is offering. If we did come out with just a normal on-chain order book, and all it did was what centralized exchanges were already doing, then that would be nothing to write home about. You have to do something extra. You have to give the market something more than what they already have. And I think this is why Carbon had to exist.
2. Carbon’s on-chain strategies are “continuous” and “self-renewing” (10:36)
Carbon’s main feature is what we’re calling continuous on-chain strategies which is not something that’s offered by not just any other DEX in DeFi but also not offered by any centralized exchange either. With Carbon, you’re not just setting a price for a particular asset that you want to sell and wait for that thing to be executed upon. What’s also happening is that you’re also nominating a price for the reverse trade. This is one of the things that delineates Carbon so significantly from everything else. When you’re setting up that position, where you say you’re willing to sell an asset to the market in exchange for something else, at the very same time you can nominate to buy that asset back at a different price. That’s a really important realization, to determine the spread, the difference between what you bought something for and what you’re going to sell it for. It’s that idea of having a continuously self renewing cycle where you’re constantly accumulating something at a lower price and then selling it at a higher price so that you can afford to buy more of that same thing again at lower prices in an infinite cycle until you choose to stop.
3. Carbon enables novel “range trading” strategies (22:05)
Another distinct difference between what Carbon does and how centralized limit order books work is that you don’t have to necessarily choose a single price. You can if you want to. Carbon fully supports these types of strategy executions at infinitely thin price ranges. If you want to sell all of your WBTC the moment that someone is willing to buy it for exactly $60k USDC then you can set that up. But for many traders, calling the top of something or calling the bottom of something is impossible. Rather than call the specific top, exactly some number being reached, in general it makes more sense to spread out the sales of tokens around a certain target range. What these very disciplined traders will do is layer up a very large number of limit orders in the direction they’re hoping to take profits in. These could be dozens if not hundreds of different limit orders all at specific price points. This all sits on the order book waiting for a taker to take that price from them. Whereas on Carbon you can do it continuously. Rather than having to set up individual prices and committing a certain amount of tokens to those prices, you can set up a range and have that order be executed continuously across that range.
4. Liquidity in Carbon automatically moves between different price ranges (27:27)
On a centralized limit-order system I have never seen the option to automatically use the funds that it generated from the sale of something, like one of my sales being taken, to immediately feed another limit order. That’s just not how it works. In general, tokens are sold, tokens appear back in your account, or the dollars appear back in your account, and then you need to take those dollars and manually put them back into the limit order book to buy something on the way back down.
In Carbon you can just announce ahead of time what it is you’re trying to do. Rather than the protocol enforce its own particular market-making strategies on you, you instead give the protocol instructions on what to do with the tokens that you’re providing. It’s a dramatically different paradigm. Not just for AMMs but for people that are doing these kinds of activities, like channel trading. It should be very natural. It should feel very very similar to the kinds of things that they’re already doing, but it should also feel sort of purpose built, a lot less clunky, a lot more streamlined to support that kind of activity.
5. DeFi needs more day traders to address its liquidity needs
I still think one of the biggest problems facing the industry is illiquidity. This has become something of a cliche, and I think this is one of the things we absolutely need to address if our industry is to succeed. Day traders can be the lifeblood of systems like this. Giving people ways to very quickly turn tokens around and flip them; it is healthy. It’s something we should be excited to see more of. And taking it off of centralized exchanges and bringing it back to DeFi, which is why we’re all here in the first place anyway, is nothing but positive. I want to empower users to do the things they’re already doing but in a DeFi way, not in a centralized exchange way.
6. Carbon allows for individual liquidity positions to be governed by multiple bonding curves (30:56)
Today when you are providing two tokens in a pair, every AMM protocol uses a single bonding curve to determine the exchange rate between the tokens. This is the status quo up, where any bonding curve is defined by two token balances, and the price between those two tokens is some function of their ratio. In Carbon, the bonding curves will quote a price based on just one token balance. If you’re providing two tokens to a pair because you have two token balances, each token balance can have its own distinct bonding curve.
This is what we’re calling “Asymmetric Liquidity”. Existing AMMs tell you that you have to sell something for exactly what you bought it for. You can take a little bit, but otherwise it has to be the same price. In Carbon we’re doing the opposite. We’re saying you get to choose. You can choose whatever price you want, and there’s nothing the protocol can do to stop you.
7. Carbon can also be used for simple on-chain limit orders (37:46)
Carbon doesn’t force you to both buy and sell a given asset. If you’re looking for an exit, you can set your sale for a specific price. If you don’t want to buy it back, you don’t have to. It will still technically go into another order, but that other order will be disabled, inactive. You can send those things to sit idle, and it won’t be rebuying tokens with them if that’s what you wanted to do.
8. Carbon is an evolution of “concentrated liquidity” AMMs with key differences from existing models (55:44)
People that are long term veterans of Uniswap v3 are already acquainted with the idea of breaking up the constant product bonding curve into little segments and then choosing just one of those segments where you want to provide liquidity. Where it’s going to feel a little bit different, and I hope a little bit more powerful, is there are no tic ranges on Carbon. You can choose literally any price. You’re no longer choosing specific buckets to put your liquidity into, but you have full agency, full autonomy to choose any price. The system is just not organized into tic ranges the same way Uniswap v3 is.
Another difference is the irreversibility. With Uniswap v3, even though you can choose these ranges, once something is trading inside that range it can still trade back and forth when markets reverse. If you’re lucky you’ll remove your liquidity before it has a chance to trade back in the other direction. If you’re unlucky, it will have started to trade back the other direction before you can start pulling it out. Effectively, even though your limit order was once filled, it gets unfilled as the market recoils after some move. This is the opposite of how you would expect a limit order to behave. On Carbon, orders are irreversible once executed.
Another difference is IL. On Uniswap, it’s really about the volume that you can support without the price moving. It’s a very different bet. If you’re supporting a very large amount of trade volume with the small amount of liquidity you’re providing, then you can earn a lot of fees. The problem is that if the market moves away from the current price then the fees that you’re earning might be very small compared to the opportunity cost that you’ve paid. This is the causal factor behind IL. On Carbon that just doesn’t happen. You’re not making the gambit that the price isn’t going to move anywhere. Quite the opposite. You’re instead trying to profit from the price going somewhere. And that’s something that’s kind of new in an AMM context.
Lastly, on Uniswap v3, if you want to update your price range, you must destroy and recreate your entire liquidity position. In Carbon, adjustments to an order can be made “on the fly” — i.e., without a user needing to close and recreate their underlying liquidity position. Parameters are adjustable via low-cost transactions involving minimal computational overhead, making strategy management significantly more gas-efficient.
9. Carbon is one of a number of new products to be launched under the Bancor umbrella (1:12:53)
The Bancor brand is not being abandoned. Bancor is the umbrella brand under which a bunch of different products are developed and marketed to specific audiences. While Carbon will utilize the BNT token, it is its own product, and I think it does make sense to give it its own specific, laser-focused targeted marketing campaign. Spinning it out with its own name just makes that process easier. It also means that as we develop other things, which of course we are, I fully expect there will be many more specific laser focused targeted products with their own names, and they will all be under the Bancor umbrella.
10. Carbon could be used to support a token launchpad (1:15:20)
When I first presented some of the ideas behind Carbon, it was actually under the context of a token launcher. There are things that you can do with these flexible bonding curves that I think may ameliorate some of the problems that we’ve seen in how token launches have gone thus far. Maybe building into a token launch something that offers essentially new investors some sort of level of assurance; something that distributes funds back to the developers at a constant rate or something so that they can be more confident that the project that they are investing in is legitimate. But yes, I fully expect that a token launcher platform will be built on Carbon. It’s been an ambition of the Bancor community for a while.
11. Carbon will be relatively easy to deploy on multiple chains (1:24:32)
I would like to see Carbon on as many Layer 1s and Layer 2s or alternative Layer 1s as we see fit. A lot of the security risks that were sort of burdening us with previous iterations of Bancor have gone away with Carbon. Moreso than anything we’ve done before, we shouldn’t feel any reluctance as a community or any doubt to deploy Carbon anywhere we want to. It can, and it should happen.
12. Coding on Carbon is well under way (1:29:02)
Coding on Carbon has been under way for months. We weren’t just working on whitepapers and stuff. The smart contract work, UI work, SDK, and everything else have all been in development. This has been happening since July. This isn’t just talking about some cool idea we have. This is talking about something that is almost ready for its debut. Compared to previous versions of Bancor, development and deployment of Carbon will happen way faster.
