DXswap Launch Parameters - seeking community feedback

In May, work began on forking Uniswap v2, turning on the governance capabilities and putting control of the new protocol under DXdao. DXswap development has been spearheaded by @AugustoL and is now in testing and auditing phase. He has presented several updates to the community on development, with the latest on Kovan deployment.

Given that DXdao will govern the protocol, there are several parameters that need to be decided by the community before launch. Over the last few weeks, there has been an weekly open call devoted to DXswap launch and strategy. Meeting notes and forum discussion from the last couple of weeks.

Below, I want to lay out the current strategy these discussions have yielded and open them up to the broader community for comment. Feedback and consensus are crucial to move forward with a formal proposal for launch.

There are three main areas for the community to consider:

  • Fees
  • DXdao provided liquidity from the treasury
  • DXD liquidity mining rewards

Fees
Recent liquidity mining schemes show how quickly liquidity can move, but what makes Uniswap an attractive product to fork are the fees it generates. Over the last 30 days, it has seen total trading volume of over $9bn, paying out $27m to liquidity providers (LPs). Uniswap has no protocol fee, but even just a 0.01% protocol fee would yield $30,000 a day of on-chain revenue.

Of course, traders hate fees. DXswap will need to be competitive to attract volume (more on that below), but the current plan is to have a small protocol fee and demonstrate DXdao’s ability to generate on-chain revenue, the proceeds of which will be sent to the curve and support the price of DXD. The current design allows for the protocol fee to be charged as a fraction of the LP fee. Consensus has landed on setting the protocol fee at 1/10 of the LP fee.

Example: Trader pays a “Swap Fee” of 0.3%. LPs would effectively receive 0.27% of the swap fee, while DXdao would get 0.03% in the form of the LP token sent to DXdao treasury.

One of the distinguishing features of DXswap will the ability to adjust fees. All Uniswap pools have the same 0.3% and even Balancer has a set-it-and-forget fees. This is a key component of the strategy to make certain pairs more attractive on DXswap to aggregators and more sophisticated strategies. Providing and incentivizing liquidity (more below) will help buttress the lost fees for LP’s.

To be decided by the community:

  • DXswap protocol fee - current consensus on 1/10th of LP fee
  • Fees for targeted pools at launch - what should fees be for pools with DXdao sponsored liquidity or pools that have DXD rewards. More on this below.
  • How can pools adjust fees? Currently, pool fee adjustment must be approved by DXdao on alchemy. Might a Snapshot like program allow LPs to signal their fee preference?

DXdao supplied liquidity
One of DXswap’s advantages over Uniswap (or SushiSwap) is the ability to deposit some of DXdao’s treasury into DXswap. This does not come without risk (including impermanent loss) and the community needs to decide on how much, what assets and how (technically) to deploy these.

As of writing, the treasury holds $8.5m in ETH, $3.9m in DMG and $2.9m DXD with an additional 2,700 DXD ($1m) released to the treasury every month.

I think it would be best to have two different liquidity injections, Stage 1 of $750k at launch and then Stage 2 of $1.25m of liquidity some time after launch (4-6 weeks?). I am proposing the following allocation to DXdao sponsored pools:

DXdao Sponsored pools Stage 1 allocation: $750,000
Pair $ % Pool fee
DXD/ETH $105,000 14.0% 0.10%
DXD/DAI $56,250 7.5% 0.10%
DXD/USDC $56,250 7.5% 0.10%
DXD/USDT $56,250 7.5% 0.10%
DMG/ETH $105,000 14.0% 0.10%
Dai/USDC $18,750 2.5% 0.30%
Dai/USDT $18,750 2.5% 0.30%
USDC/USDT $18,750 2.5% 0.30%
ETH/Dai $105,000 14.0% 0.10%
ETH/USDC $105,000 14.0% 0.10%
ETH/USDT $105,000 14.0% 0.10%

DXD and DMG would come from the treasury, while other assets would be converted from ETH. There remains some technical challenges and potential opportunity (Liquidity.eth) to getting ETH into DXswap pools. Stage 2 liquidity would presumably start with the same allocation with adjustments based upon trading and liquidity trends after launch.

To be decided by the community:

  • How much of treasury to allocate to DXswap in total and to what pools
  • How to get treasury funds into DXswap

DXD Liquidity Incentives
Yield farming has been….popular to say the least. APYs are increasing almost as fast as gas costs. Still, the past few months have proved that liquidity responds quickly to on-chain incentives. The challenge for DXswap will be to make that liquidity stick. In that regard, liquidity mining campaigns can be seen as achieving two objections, marketing for launch and long-term liquidity incentives.

I propose DXswap employ two different, concurrent liquidity mining strategies to achieve these objectives.

  1. Launch Rewards - DXD incentives for the first two weeks of launch. Awarded purely on deposit size in targeted pools, rewards can be claimed immediately.
  2. Ongoing Incentive Rewards - Weekly awards to liquidity providers based on trade volume in targeted pools. Vested to avoid farm & sell (3 months?)

Launch Rewards would produce a catchy APY, but ongoing incentive rewards should attract the long-term liquidity providers interested in contributing to the DXswap (and DXdao) ecosystem.

In terms of size, I landed on 1300 DXD ($500k) for Launch Rewards to be distributed over two weeks, with 300 DXD distributed on the first day, 250 DXD on day 2, 200 on day 3, etc. And for Ongoing Incentive Rewards, 50 DXD ($19,000) awarded daily, distributed weekly and vested.

Liquidity incentives should only be for specific pools to prevent gaming and abuse, but the reward program should be extended to projects and tokens close to DXdao community. At launch - in addition to the pools listed above for DXdao sponsored liquidity, GNO-ETH, PNK-ETH, GEN-ETH and LRC-ETH should be eligible for DXD rewards. Additional pools should be considered by the community for strategic or partnership reasons.

To be decided by the community:

  • How much DXD to distribute?
  • Vesting or immediately claimable?
  • How to calculate rewards?
  • Which pools to incentivize?

I used this spreadsheet model, where you can see some of the underlying projections and calculations for the whole excercise. Projecting liquidity and volume is difficult as is how the liquidity mining campaign will affect the price of DXD, especially with a bonding curve available.
++++++++++++++++++

DXswap’s launch is dependent on the community coming to consensus on the parameters laid out above and approving a formal proposal.

Please provide feedback below. We’ll be discussing these items on tomorrow’s DXswap strategy call, the Biz dev call on Monday and Thursday’s Weekly call too (and maybe on the dev and governance call??).

8 Likes

When is the date for launch, or approx date range?

also, where does the revenue go to ? does it go to the treasury ?

1 Like

Hey @CLCL!

When is the date for launch, or approx date range?

Once we get the DXswap launch parameter ratified by the DXdao we are ready! (This month hopefully)
There are thoughts about letting the DXdao itself deploy the contracts which means everyone can observe the proposal about the deployment on Alchemy.

also, where does the revenue go to ? does it go to the treasury ?

My first guess is that it will first go to the DXdao avatar address because dxtrust (bonding-curve) can only natively handle Ether and DXD. DXswap will generate fees by all kind of token pairs which mean we will actually collect all kind of tokens. Having a monthly “burn” event where the DXdao will exchange all ERC-20 tokens to ETH which will then be send to the bonding-curve (and of course the collective native Ether).

Great questions, we will discuss the plan today:

1 Like

I personally care much more about growth then revenue so I’m squarely in favour of launching with no protocol fee (which all things considered would put us at a competitive disadvantage to Uniswap on day one, while trying to take market share). We don’t have 9bn in volume, so earnings are likely to be disappointing in the beginning, if anything. This can always be changed if/when Uniswap enable their contractual right to 0.05% protocol fee - thereby we also live up to our mission of keeping projects/products real in the face of VC money etc. I pool a lot on Uniswap and would move my liquidity if LPs made the same, not if I was at a disadvantage on a new platform with less volume.

I am in favour of the proposal to seed a limited amount of liquidity. Not sure why we would have two stages though, surely we want to try to make the pools liquid and appealing on day one. We get one shot in my opinion, nobody is going to care 4-6 weeks later. So maybe 1-1.5m on day one instead? Instead of spreading it on so many pairs I am also in favour of specialising. Due to Aave you can now lend against your Uniswap tokens for some stable coins, so we are at a disadvantage there and I don’t see us take any sizeable market share on the big pools (USDC etc.) on day one. Focus is key I think. After all we compete with the LPs by doing that too (we take some of the fees). So I’d say we back our own pairs and make trading of these better on DxSwap - and then focus on partnerships using our competitive advantage (that projects can build preferential pools on DxSwap as opposed to Uniswap). Same with the rewards, let’s focus on key partners that may in turn also want to push us.

“At launch - in addition to the pools listed above for DXdao sponsored liquidity, GNO-ETH, PNK-ETH, GEN-ETH and LRC-ETH should be eligible for DXD rewards. Additional pools should be considered by the community for strategic or partnership reasons.”… this basically sums up my position.

1 Like

This focus you mention, first, on “our own pairs” and, second, bringing partnerships to DXswap, I agree that these are two strategies to go after. With both of these strategies, I don’t think the lower swap fee or the small protocol fee will affect either one of these much.

It will be great to experiment with different ideas. However, I don’t think that DXswap’s success is make or break in the first 4-6 weeks. The DAO can make changes along the way as we learn.

1 Like

I suggest we focus mostly on DXD pairs to avoid the “farm and dump” we see in other yield farming tokens like Sushi and Kimtchi. If you need to DXD to stake, buying pressure will over-compensate the farmers selling pressure.
We are in a unique position to execute a Vampire attack, since our LPs rewards are in a token that gets revenue shares across multiple products at equal volume DXD is hugely valuable than a token like $SUSHI.
Given the points above, this is my proposal:

  • We keep USDT/ETH as the only reward pool without DXD
  • We increase rewards on other pools up to $1M - $1.5M total rewards on day one.
  • We reach out to other projects and offer some kind of partnership for a pool with their pair / ETH
3 Likes

I dont know about those rewards numbers but I’m down for reaching out to other projects, especially mstable and dmm

What do you think the market’s reaction will be when DXSwap is launched? Forks are generally seen negatively. We need to make sure it does not tarnish dxdao name.

1 Like

I think the key is to focus on a few core aspects in the launch comm’s;

  • The differentiation - e.g fees can be set for each pool so we can partner with projects directly.
  • Uniswap just raised 11m from VC’s. We’re here to keep them real and competitive.
  • We’re also not a blatant vegetable/food incentivized money grab, so there’s that.

I’d love to keep building on the differentiation, but point 2/3 is valid enough reason to go ahead.

3 Likes

Great points above + discussion on DXswap call on Friday. Just wanted to lay out a couple items from the discussion above others offline:

  • “Launch Parameters” and “Iteration Roadmap” - features that need to be decided on for launch vs. items to add to DXswap ecosystem. As we move towards audits and deployment, we’ll need to make sure that we’ve put these items into separate buckets

  • Focus on pairs we can win - echoing some of @gamaranto comments above, but any launch or rewards strategy is to try and corner the market on a pair. Lower fees and better liquidity are how to get there. It will be difficult to do this for large pairs, easy for DXD pairs (with DXdao liquidity) and possible for smaller/partner pairs with liquidity incentives

  • Sushiswap - Rise, fall, rise again…the saga continues and like it or not, DXswap will be placed into the same bucket as Sushiswap or “just another Uniswap fork”. I agree with @madmax and @balotelli45, that it’s important for DXswap to distinguish itself.

  • Can DXswap compete with the farm of the week? - Considering other projects are also distributing a token, it will be tough for DXswap to compete in the “APY” game. I think this means less rewards to Launch Rewards and more geared towards Long-term Liquidity Incentives.

  • Project partners - DXswap is a great opportunity to engage with other projects in the DXdao ecosystem. There could even be formal proposals that set out DXD rewards for certain liquidity pools. Project level cooperation will be crucial to building liquidity in pairs that will stick.

Still thinking these through and I’ll send through another note before Friday’s meeting.

I do think it’s helpful for us to start thinking of things as “Launch Parameters” and things for the “Iteration Roadmap” for ongoing initiatives.

2 Likes

Aluna Social/ALN plans to put up an ALN/ETH pool on DXswap, with plans to issue ALN rewards to those LPS. I think it would be great to sponsor that pool as well. They should be proposing a Mesa listing this week.

1 Like

Hi all, I wanted to seek community feedback on Launch Parameters and Iteration Roadmap. Launch Parameters are needed for deployment and will be baked into the protocol. Most importantly is finalization of Launch Parameters.

Launch Parameters

  • Fees

    • Protocol fee: 1/10th of swap fee
    • Default pool fee: 0.15%
    • Max pool fee: 10%
  • DXdao sponsored pools - $1m total liquidity

    • DXD/ETH - $300k & 0.10% pool fee
    • DMG/ETH - $200k & 0.10% pool fee
    • PNK/ETH - $50k & 0.10% pool fee
    • ETH/Dai - $150k & 0.10% pool fee
    • ETH/USDC - $150k & 0.10% pool fee
    • ETH/USDT - $150k & 0.10% pool fee
  • Deployment and Fund allocation

  • Next steps:

    • Seek final community feedback on forum
    • Submit signal proposal this week
    • Augusto & Nico begin work on technical deployment and fund allocation

Please provide feedback on inputs above before Wednesday

Below are items that are still in active feedback stage.

Iteration Roadmap

  • DXD Rewards (DXD/ETH, DMG/ETH, ETH/Dai, ETH/USDC, ETH/USDT, LRC/ETH GNO/ETH, PNK/ETH, GEN/ETH, MTA/ETH, ALN/ETH, Dai/USDT)
  • Pool Governance
    • Process & flow for applying to DXdao for fee change (short-term)
    • On-chain mechanism for pool governance and voting
  • DXswap.eth.link
    • Token Lists
    • Rewards page
    • Integrating Uniswap pools
  • 3rd party integrations (DXswap.info)
  • Partnerships + Mesa

Next steps:

  • Prepare DXswap Launch rewards plan for feedback to community this week. Prepare a signal proposal for DXswap Launch Rewards for next week.
  • Research long-term incentive rewards calculation, time decay and withdrawal fee
  • Pool governance: create guide for fee change application prior to launch, build technical solution for on-chain signalling
  • DXswap.eth.link - a DXswap token list is needed for Day 1. It’s expensive to add to Mesa list, so should take all of Uniswap list and then add to it going forward.

Outstanding Questions

  • DXswap revenue to the curve
  • Long-term incentives - time decay and pool specific?
  • DXdao DXD rewards - what to do with them?
  • Integrations, new features and more iteration

Please provide any feedback on Launch Parameters. Be on the lookout for a signal proposal on Wednesday and then another post on DXD rewards by the end of the week.

2 Likes

Jumping in late to this. I think Launch Parameters are fine like this. I would not set protocol fee to 0 because that’s what differs DXswap from other swaps and people need to be aware of this and see this.

Why stable coins pools in the first post have such a high pool fee 0.03%? Fees in swaps are here to offset impermanent loss. For stable coin pools there is almost no IL, so fees should be lowered, not set higher.

About Launch Rewards:

“… rewards can be claimed immediately?” Would this not lead to wash migrate? Or does immediately mean after the initial two weeks?

I’m not a fan of strong non-linear rewards. Informed people like us will rush in and get the most out of it, and the late followers once again only get little incentive to join.

Ongoing Incentive Rewards

Vesting is key, best would be only pay the vested amount if you still have your liquidity in the pool. If you move on, you just lose it to the others or it goes back to the treasury for the ongoing incentive rewards and help to make this phase longer. This would make liquidity more sticky without change in the pool logic. (What also could be done is stream the reward to the LP, if you withdraw, this stream just stops)

“Focus on pairs we can win”, - very good!

Happy to see more swaps!

3 Likes

What if we fostered adoption and migration to the xdai chain by launching dxswap on the xdai chain exclusively.

Uniswap is the reason gas prices are so ridiculously high right now. If we can provide a cheaper, still decentralized alternative that’s functionally the same, we could grow both the dao and the xdai chain at the same time. The way the dao is going, with Mesa and omen migrating, why not just skip the migration step and launch in xdai.

We hold fair amounts of coins in quality projects, and could theoretically add a decent amount of liquidity at launch. The token bridge is ready, just needs to get the dashboard built. This would give us a competitive edge over uniswap by being the first mover for a swapping protocol on xdai with good liquidity.

The projects we would add to dxswap would be excited that their community would have a trustless way to trade their coins in a cost effective way, and would show support on social media. If we can show liquidity providers that we can handle more swaps a day, we can offer them more fees to profit off of. If we can show traders they can save 10s of dollars a swap by switching over to the xdai chain and using our swapping protocol, they’re going to come over and trade with us.

This is just my thought, especially today with the release of the UNI token and how badly it’s running up the gas price on the network.

Cheers.

1 Like

I think this is an interesting idea and starts to address some worries that I have been having (that are exacerbated by the UNI release) about the product market fit of DXswap. Namely that it is an now an undifferentiated product in a crowded space with a clear big dog (Uniswap). Right now DXswap is like Uniswap in that it is a general purpose protocol, and as of last night cannot rest on governance as a differentiating feature. DXswap risks being a smaller version of uniswap that doesn’t have the name recognition or liquidity to compete. To use a bad analogy: If we think of Uniswap as a beautiful, sharp kitchen knife, maybe instead of trying to create another knife just like it, instead we should be trying to create a scalpel. Similar product but very different use case.

So what does that mean? one option is creating a product that can be optimized for specific use cases (I wrote about one idea I had here to use Mesa and DXswap for limited run collector items. Integrations for DXswap with Mesa and Omen are key, and if done right could create nice symbiotic relationships that. Users may not even realize they are using dxswap.

The other option is to expand into sectors that are under served by other AMMs. Considering the plan for the whole suite of Gnosis and dxDAO products to migrate to xDAI, this is a no-brainer! It seems likely that uniswap will move to a rollup (probably Optimism), so there is a good chance that the AMM market on on xdai will be ripe for the taking. Establishing dxswap as the go to AMM on xdai would help build the brand and hopefully help establish it before lots of bridges start appearing and jumping to another sidechain/rollup becomes super easy.

1 Like

Why not both? It’s a general purpose swapping protocol.

It costs $30 to swap on uniswap today. That’s a lot of lost profit forfeited to miners. That’s less profit for traders, less profit for liquidity providers. If we could put out a v1 Use at your own risk, this project is in beta AMM on xdai, we’d be able to get so much volume by lowering trading costs.

V2 can have built into the dashboard token bridge functionality, potential governance token functionality, limited-edition hype beast bonding-curve nft promotions(really good marketing idea),(maybe just bake rails into the dxswap interface zkrollups on layer 2) etc…

But, if I could take coin A and turn it into coin B for liquidity pool fees and a few cents on the xdai chain, I’d be soooo happy. Someone is going to do it, might as well be dxDAO.

Cost effective layer 2 Uniswap. That’s the competitive edge, that’s the selling point. There is a huge demand for a scaling solution. Trading velocity would go through the roof if the cost of trading dropped 90%. That would make liquidity pools more profitable, thus earning the dao money by providing liquidity to our new dex.

It’s my personal opinion that launching our AMM on xdai before anyone else is able to do it would put us in a good position for growth.

I don’t want to be a debbie downer because I like these ideas, but I’m skeptical that only having a cheaper uniswap on xdai is a huge advantage.

The first reason is liquidity. There is a chicken/egg problem here in that without liquidity, trading is subpar, and without trading, liquidity stays away. Yield farming can help with this, but only to a point. As has been discussed above, the focus with liquidity mining would be on pairs that the dxdao can support and win on. I don’t think that will be enough to bring much general purpose trading without some other carrot. Furthermore, if the consensus is that xdai is less secure (or perceived as less secure), lots of liquidity may be discouraged.

Second: Gas costs. I’m not sure what the gas costs for the bridge will be, but my guess is it won’t be negligable. This is fine for traders who send large batches of funds over, but many uniswap users just buy intermittently. So if you want to say, trade ETH for SNX and then go back and stake SNX, your gas costs are: bridge in, trade, bridge out which may not be a saving over that same process without the bridging but with a more expensive trade. I don’t expect serious traders to come to xdai when they can use other L2s that are optimized for trading like loopring or DeversiFI.

Third: Uniswap is going to move to a L2 sooner rather than later (see unipig). If that happens and their bridge is just as good as the xdai bridge, why would you come to dxswap if you could spend the same amount to go to uniswap’s L2?

So with that said, dxswap makes a lot of sense on xdai if people are moving funds to use on xdai and while they are there, they can easily swap with dxswap. This changes the framing of dxswap from the product itself to a key piece of infrastructure that products can use within the xdai ecosystem. I see uniswap evolving this way too.

Uniswap V3 is also something to be aware of. Who knows what it will entail, but Hayden has been talking it up recently. If they add a bunch of new features and push to a rollup, no matter how fast dxdao pushes an implimentation to xdai, we will still be playing catch up.

xDai will be a great sandbox to test out all of the dxDao products without having to worry about gas costs, but I would caution against making xdai a selling point that products are focused on.

^ some good points,

I think derivatives market is a good target to consider too, it is a huge sector, leverage is useful, and leverage also helps generate more volume and as a result more revenue through fees.

im not a technical person, but is there a way Rails can come into play here ?

Good points from both @rossgalloway and @MasterSpoon. We should humble and clear-eyed about where we can & cannot compete with Uniswap. It’s a behemoth, but even taking a small sliver of the market share would be meaningful revenue for DXdao. Also, as @CLCL points out derivatives market could be huge and we don’t really have a play there. This piece from Multicoin on creating the decentralized Bitmex might be helpful for coming up with that strategy. DXdao needs to compete in that market.

Two additional DXswap thoughts :

  1. Pool Governance is an area that we can compete with Uniswap now. Augusto/Nico’s fork added fee adjustability. This needs to be incorporated into the UI in a Pool Governance section. Uniswap is also not upgradeable. So it’s really the on-chain treasury to encourage usage of v2 and entice migration to v3.
  2. DXswap.eth (the front-end website) can be much more expansive than this first fork of Uniswap v2. @JohnKelleher has postulated that you could actually redirect all of the flow from DXswap.eth to Uniswap if the price was better than DXswap. More importantly, any DEX can plug into DXswap.eth, almost as an aggregator of sorts. Hopefully, DXdao can iterate and build the best AMM protocols, be it Balancer-style adjustable pools or whatever Uniswap v3 comes up with. This is a long-term game and I expect the AMM space to be completely different in 6-9 months. DXswap is just getting started. That’s why it’s good to only focus on what we can compete on now, but prepare for more.

And then layer2/xDai/Rails - this is all worth exploring but I don’t think there is a silver bullet. There’s a ‘swap’ type option for Rails that is mostly a UI trick, but Loopring just announced that you should be able to build AMMs in their zkrollup (it’s not fully EVM compatible). I think it will be tough for xDai to soak up liquidity for an AMM to offer competitive pricing (so many token pairs on Uniswap), but Gnosis just announced its relaunching Gnosis Protocol on xDai, so we will already be able to test around, a “sandbox” as @rossgalloway said. Layer 2 market will continue to change, so I think DXdao should explore all of the options.

2 Likes

This is an interesting thread on how you could split revenues in an AMM between LPs and Token holders. Swaps pay LPs like usual, but LPs pay a tax to leave. It’s an elegant concept that could spread the fee over both users and LPs.

1 Like