[OLD] GEAR Liquidity Raise and Token Transferability

**EDITS Nov 9th

  • Noted that liquidity will be wETH and reasoning for not doing stETH or aWETH.
  • Removed the whitelist.
  • Added clarification around community round 2 funds.
  • Clarified ops around setting up the liquidity pool.

0. Intro

This is an updated version of the gear token launch proposal. The purpose of this proposal is to:

  1. Officially target a timeline for making the token transferable
  2. Set parameters for the gnosis auction token sale
  3. Determine a home for protocol owned liquidity and the parameters for LPing:

Gearbox V1 was the minimum viable product - Gearbox V2 is the product that is ready to onboard users and start earning meaningful revenue. GEAR is the token that governs all of this. Therefore, this is the perfect time to bring back discussion around if, why, when, and how to enable the GEAR token to be transferable.

You can view this document in google docs form, as well as previous versions, here

1. Skip the reasoning, gimme the specs

A vote “FOR” this proposal would signal approval for the following token launch plan:

  1. Target enabling token transferability ~5 weeks after V2 launches (exact timeframe depends on dev, multisig, and marketing availability).
  2. ~4 weeks after V2 launches (1 week before transferability, Approx Nov 1st), host a liquidity fundraise token sale. Sell 1.5% of the token supply. Utilize a gnosis batch auction executed thru the gnosis multisig auction application, with the following parameters:
Parameter Value Unit
Auctioning Token 0xBa3335588D9403515223F109EdC4eB7269a9Ab5D GEAR contract
Bidding Token 0xC02aaA39b223FE8D0A0e5C4F27eAD9083C756Cc2 wETH Contract
Sell Amount 150000000 GEAR tokens
Minimum Buy Amount $0.015 x 150m GEAR USDC
Minimum Funding Threshold 0 wETH
Order Cancelation End Time Start Date/Time plus 12 hours Month, Day, Year, Time
Auction End Time Start Date/Time plus 72 hours Month, Day, Year, Time
Minimal Bidding-Amount Per Order 10000 GEAR tokens
Atomic Closure Allowance No (Unchecked) Checkbox
Whitelisting Process No (Unchecked) Checkbox
Miner Role Gnosis Auction Contract Role bestowed by multisig
Whitelist Contract none n/a
  1. Pair the proceeds from the sale with an additional 1.5% of circulating supply and provide liquidity on Balancer.
  2. If raised funds are less than $2,000,000, create a separate proposal to determine best path to obtain more liquidity (liquidity mine uni v2 LP positions, or supplement from strategic raise round 2).

2. Determination of Timing

The base assumption is that eventually, the GEAR token needs to be transferable. There are multiple factors that play into the timing of the token launch.

Benefits of an Unlocked Token

There are a few benefits of an unlocked token:

  1. It is easier for contributors and others that wish to become active to gain a financial interest in the protocol.
  2. Liquidity mining is more effective as LPers and users do not have to try to price unknown value / timeframes.
  3. Contributors have flexibility to adjust their payments each month to skew towards/away from gov tokens depending on financial needs/planning, without needing to ask for adjusted pay each month (ie, contributors can customize the % of their income that’s in GEAR vs stables)
  4. We can have token price feeds, which allows for use cases such as portfolio margin to be integrated, allows liquidations to be more efficient and less costly, and allows for GEAR integrations with other protocols.

Token Utility, Regulations, and Product Growth

Currently, when tokens unlock, holders will have 4 options: Sell, Hold, Buy more, or Provide Liquidity. Ultimately these are all forms of Buy, or Hold. Those that choose to hold the token are likely expecting a future return. The longer it takes to unlock the token, the more time Gearbox has to create this future return. However:

  1. In their short history, DeFi protocols have usually taken a few years to find product market fit and proven sustainable profitability.
  2. Even if Gearbox does find product market fit sooner, there is no guarantee that this value is immediately transferred to the token.
  3. Regulatory uncertainty creates even more of an unpredictable timeline on these items
  4. Revenue share in the growth phase of a protocol is generally not good for the long term success of the protocol.

The primary point is that there is no hard timeline on when Gearbox will start earning significant revenue, and then there is no hard timeline after that on when that value could transfer to token holders. That could lead to a scenario where two years from now GEAR is still a valueless governance token that does not benefit from the success of the protocol. At that point, if the token were to be unlocked, then 2 years would have been spent without the benefits of an unlocked token just for a big old nothing-burger of an unlock event, where holders still only have the option to sell or continue holding.

There are potential ways around this, such as direct utility of the GEAR token as cross-margin or other benefits for users. However, this still requires GEAR to find PMF (utility without revenue is akin to lipstick on a manatee, with sailors that have been at sea too long and slowly sober up. Or, to put that less ridiculously - it just creates hype / wealth transfer games that benefit those that understand early and eventually get figured out).

Vested Token Unlocks

Currently 29.2% of the supply is owned by early contributors and ex-core, with a 1 year linear vest beginning Dec 15th, 2022. Launching the token after this vest begins would potentially create bad optics for the DAO (tokens that were meant to be locked are instead effectively unlocked with every other token).

Gearbox V2 Hype

Launching into the hype of Gearbox V2 gives the DAO a better chance at having an efficient token sale and liquidity launch. There is risk of creating FOMO (and thus unhappy community member bag holders), however current market conditions will likely dampen this significantly. The token launch is also set for 5 weeks after V2 launches in order to give the crypto community time to fairly evaluate Gearbox V2.

Liquidity Mining

As mentioned above, liquidity mining is more efficient with a liquid token. There has been significant discussion around a liquidity mining program after V2 launches, so that is a major reason to launch the token sooner than later. The liquidity mining event would likely not launch with V2, to let V2 scale up a bit prior to bringing significant TVL. This means its acceptable for the token launch to also not occur with the launch of V2. Additionally, devs would probably kill us if we voted to launch liquidity mining and the token event at the same time as V2.

3. Sizing Liquidity Relative to Circulating Supply

The current state of GEAR tokens is as follows:

  • 29.2% for contributors/initial core, 1 year linear vest beginning Dec 15th 2022
  • 7% to Credit Account Mining, Testers, Discord Legends, and Retroactive Payment Round - available once transferable
  • 1.4m gear tech multisig [negligible]
  • 3.766% to Strategic Funding Round (locked for 1 year)
  • [Proposed] 0.667% Strategic Funding Round 2
  • [Discussed] 2% LM of pool suppliers over 3 months
  • [Proposed] 0.04% per month in contributor payments + tech multisig

When adding in the proposed 3% for liquidity and consolidating categories, that leads to:

  • ~11.5% “circulating” (ie, no vesting) supply
  • 30.7% vested
  • 55.3% DAO
  • 1.5% in liquidity pool

The current FDV per the vested fundraising round is $150m. We will keep this same valuation to make a funded auction a more likely outcome, but the market is free to price GEAR higher to than 150m FDV to account for the fact these tokens will be unvested. With 11.5% of supply circulating, that leads to a circulating market cap of $17m. 1.5% of supply would be $2.25m of liquidity, paired with ETH, for $4.5m total. This means 13% of circulating supply would be able to be sold into protocol-owned liquidity, in absence of other liquidity providers and buyers. This means that dumpoors will need to find willing buyers or sell GEAR back to the DAO at a cheaper cost than was likely paid for it (thus saving the DAO money on credit account creation). Separately, this also means only 1.5% of the total supply, or 13% of circulating supply, is available for purchase on the open market from the DAO.

This is also a decent liquidity base to build from for eventual listings on centralized exchanges, if desired (comparable to other current cycle DeFi protocols, good example being GMX which has ~4x the FDV and ~4x the liquidity). The liquidity would also be slightly more effective if it was utilized in a wide Uni V3 range. Lastly, funds raised from the proposed strategic funding round 2 could also be utilized for protocol owned liquidity if deemed inefficient. Inefficiency could be measured as GEAR beta (ie, volatility) relative to ETH, as too thin of liquidity creates high volatility that is undesirable for investors, holders, and contributors.

Note that token price scales with initial liquidity, ie if the initial token price is 0.045c (3x), then the initial liquidity will also be 3x ($13m).

4. Gnosis Batch Auction

Brief Overview

Docs are here: https://gnosis-auction.eth.limo/#/docs#topAnchor

A batch auction runs through a bidding system. Users deposit capital and bid a maximum GEAR price they would pay. At the end of the auction, the order book is analyzed to determine where the supply and demand curve intersect - this will be the GEAR clearing price. Bidders who specified a maximum price higher than the clearing price will receive GEAR tokens at the clearing price. There will be no whitelist.


Here are the proposed specs for a Gnosis Batch Auction from above, with context:

Parameter Value Justification
Auctioning Token 0xBa3335588D94035152 23F109EdC4eB7269a9Ab5D GEAR contract address
Bidding Token 0xC02aaA39b223FE8D0A0 e5C4F27eAD9083C756Cc2 wETH Contract Address. wETH is used as liquidity will be in wETH.
Sell Amount 150,000,000 GEAR Tokens 150 million GEAR tokens, see justification in section 4.
Minimum Buy Amount $0.015 x 150m GEAR Minimum buy amount is used to set the effective minimum sale price of $0.015c. $0.015 is set to exceed the recent strategic round to account for the gnosis auction tokens being unlocked immediately after the sale.
Minimum Funding Threshold 0 There is no minimum, as any amount of POL will be beneficial. If the raise is low, POL can be supplemented with liquidity mining.
Order Cancelation End Time Start Date/Time plus 12 hours Give a short period of time for people to re-think their bids if they FOMO in at launch.
Auction End Time Start Date/Time plus 72 hours Auction will last 3 days, which means at least one weekday will be included.
Minimal Bidding-Amount Per Order 10000 Denominated in GEAR tokens, corresponding to a minimum of $150, higher if price is higher. Minimum amount is used to offset gas costs of clearing the option ($5-10 for token transfer for each bid)
Atomic Closure Allowance No (Unchecked) unecessary for an illiquid token
Whitelisting Process No (Unchecked) There is no advantage to having multiple addresses participate, so there is no need for a whitelist.
Miner Role Gnosis Auction Contract Auction needs the miner role to be able to send GEAR tokens to buyers before transferability is enabled.
Whitelist none There will be no whitelist


Graphic is assuming 100m tokens sold at 0.0125c min price (ie, not updated for current proposal of 150m tokens at 0.015c min price).

In the example above, we are selling 100m gear tokens at a minimum buy amount of 1.25m, which corresponds to a minimum buy price of 0.0125c. The red curve is the supply curve, which is just the total supply (100m) multiplied by the price on the x axis. The blue curve is the demand curve, which represents the amount of capital willing to buy at each price (ie, 100% of committed capital is willing to buy at 0.0125c GEAR prices, because they have all bid at least that high).

The token sale price is determined by the intersection of the supply and demand curve. In the example below, this results in $5m of total capital being willing to purchase GEAR at a price at or above 0.05c, meaning they will all get GEAR at 0.05c. The remaining committed capital will get nothing.

The min buy amount determines the minimum (reserve) price. So long as demand is less than supply, tokens will be sold at the minimum price. Once there are enough orders to buy the full supply, the sale price will begin to increase. More examples can be seen on individual pages for past auctions.


While there is no price cap, anyone buying in the batch auction will know that there is a much larger quantity of circulating supply out there. If the market is somewhat rational, this should limit the FOMO. Additionally, the fact that not every order gets filled means there will still be some remaining demand, as many users may be willing to reconsider their minimum purchase price when the token hits the open market.

Users can bid on prices, meaning they can set their own price cap personally. This creates an upper limit on price for individuals. The final token price is determined by the bidding, and everyone pays the same price (meaning you will pay your bid price or less, should the token price be under your bid price). This creates an equal token price for all (similar to Degenesis) but lets the market set that price.

Additionally, the fact the system already exists saves on dev resources.


There is no cap per wallet, meaning a few whales could in theory purchase the entire 1.5% sale of circulating supply - but they would also have to outbid everyone else in order to do so (thus are less likely to have exit liquidity as they fomo’d highest)

While dev resources are saved, the system itself is more complex and therefore may result in less participation or users not understanding what they are participating in.

A Note on Degenesis

In absence of anyone willing to take a grant to built the Degenesis style token sale (and get it approved by governance), Gnosis Auction is the path of least resistance with limited tradeoffs between the two. Degenesis could still be pursued if someone steps up to take charge of building it, but for now the Gnosis Auction only requires work from marketing and multisig, whereas Degenesis would also require front and back end dev work.

5. Liquidity Home

The proposed liquidity home is Balancer, due to potential partnerships that would let us accumulate an initial stake in Aura and then vote for and farm our own pools (which would also create additional liquidity). Additionally, bribing is more efficient than standard liquidity mining and is an option on Balancer/Aura (not an option for Uniswap), if more liquidity is needed for a period of time.

The full amount of funds raised from the gnosis auction will be paired with an equivalent value of GEAR from the DAO and used to seed the liquidity pool. The DAO also may propose to use an additional $500k worth of ETH from the community raise 2 for liquidity, after the Gnosis Auction has concluded.

Note that the liquidity pair is denominated in ETH as altcoins are still very correlated to ETH/BTC, moreso than USD. Denominating in USDC tends to lead to more potential IL in both directions as well as more missed upside if ETH is to continue its lifetime longterm upward trend.

There is an option to denominate liquidity in stETH or aWETH, which would result in 75% of Balancer revenue being recycled as bribes. However, this means GEAR liquidity is dependendent on another protocol (LIDO or AAVE). As liquidity for stETH or aWETH growth on Balancer, the DAO could consider switching to either of those for a liquidity base. For now, we will keep things simple and pair with wETH.

Prior to enabling transferability, the multisig may use a nominal amount of GEAR and WETH (<0.01) to seed a new 50/50 pool on balancer as a test deposit. Should that succeed, the multisig should queue and execute two transactions:

  1. Enable GEAR token transferability
  2. Seed a NEW 50/50 GEAR/WETH pool on Balancer, at the price matching the USD price at the end of the Gnosis Auction.

Note that the old pool shouldn’t be used as someone could in theory purchase the GEAR in this pool by routing through the internal balance, thus changing the price before liquidity is seeded.

Lastly, the Gearbox DAO will apply for a BAL gauge on Balancer.


Awesome ! support Gnosis Auction and LM on V2 at the beginning

Wow this is amazing work mr Koala! Love all the points, have some questions:

a. Can the auction be done with USDC or it’s only WETH? Usually an ETH pairing is good in bear market, because then the asset can go up in price simply due to the ETH appreciation. However, in continuous bear markets it might serve the opposite effect where the asset is “tied to ETH direction”. Given that this is likely to be the biggest & only liquidity avenue for some time, we have to consider imho.

b. What happens if the desired amount is not sold? I might be not understanding something, but is there a soft minimum cap? I can only see in 1.4 that “creating a separate proposal and pairing” - not sure about that. There will be very little time to maneuver, so this better be decided in advance. I would personally (as much as my votes can matter) not use DAO funds for LPing as the treasury has a runway and reducing it doesn’t sound like a good idea? If less = let it be less? Perhaps the best idea is to indeed say "ok up to $500,000 can be used from treasury.

c. Is there a minimum price to be guaranteed? It’s 200M FDV, right or no?

d. Liquidity Home: ask who can give incentives? Is there a way we can work with Balancer (hence Aura Finance) or some other UNIv2 like protocol - to have boosted incentives to LPs without using GEAR supply at the start at least? I fully agree (due to Ribbon’s case) that at the start V3 is not a good idea, but maybe working with an established AMM that can support with some brrrrr into th GEAR pool = a good idea?

1 Like

a. Would the intent still be to convert this USDC to ETH to seed liquidity? I guess that works fine, makes the math easier to initiate the auction since our targets are in USDC. What I don’t want is USDC liquidity - too much extra IL. If others are aligned there, I’ll make that change.

b. We can put in a minimum cap, right now it’s set at 0. I initially had it at 500k, which would be $1m of liquidity. But we have this strategic round #2 we can pull from (my understanding was that these funds were most often noted to be used for liquidity if needed), as well as liquidity mining. The intent is not to pull any funds from the DAO outside of strategic round #2 (I can clarify that). So it feels like regardless of how little we raise if it is a small amount, it doesn’t exclude any of the other options (if we raise 100k , it just means we have to LM a bit less or take less from strategic). I guess we could do like 100k just to ensure it’s at least somewhat worthwhile. I agree it needs to be more clearly defined but would like to have some more discussion around it first.

c. Yes, 200m FDV aka 0.02c minimum price. This is not a direct variable in gnosis auciton. It is set by the minimum buy amount divided by the # of tokens being sold.

d. Balancer would require us to own AURA or find an aligned partner that owns vlAURA to direct rewards to us. Or, we could bribe vlAURA for liquidity - which would be more efficient for some period of time until it eventually balances out with straight LMing. Investing in AURA would mean we plan to incentivize liquidity for a LONG time, which I don’t think is our goal given we don’t plan to LM users forever. Also - just want to verify that you can do a 50/50 pool on balancer (80/20 would be bad for this use case IMO, not enough GEAR surplus).

I know Arrakis does tokenized lightly managed v3 positions (it’s basically a lightly managed wide range position like I initially suggested, but tokenized so that you can LM it & ppl will all be in the same range). Issue is they run manually thru a multisig right now - I don’t think that’s a good option unless they decentralize more and automate the process.

Of all this I think bribing vlAURA would be the best option for short term value extraction (or really, LM cost savings) but then if bribes equalize that means we either commit to Balancer as a long term home or have to at some point migrate liquidity.

1 Like

a. Why more IL in USDC? I don’t get it, maybe I am not familiar. I just think pairing against ETH puts the token in the first months at the mercy of the ETH direction. If it happens for the market to be giga rekt in October, then maybe ETH pairing is actually very smart to begin with. Otherwise, it will rekt GEAR a bit?

b+c. Good points! I don’t know the intention of part 2 anymore as it’s apparently now going into same multisig and is not a separate liquidity multisig (as initially discussed). Pulling major chunks from there seems not as good, as it hinders runway a bit? Depending on the amount raised of course. Anyway, I would suggest putting at least 1M as minimum, because below it selling makes no sense idk. It would suck big time if 1M is not reached, because then token transferability plans would be fairly… under threat. Would suggest overall: 1M as minimum, and the lowest bound at 150M FDV (not 200M FDV). Just to ensure the soft cap is met.

d. Balancer + Aura could just be better as it is extra $$ while is fairly similar to Uniswap interactions. Besides, since majority people use Uniswap - that would incentivize someone to also create a pool there. Decentralizing liquidity building then… I would stay away from bribes honestly (we have enough inflation planned as is) and just chat to Aura and Balancer if they are willing to support this intrinsically for now, or maybe even a DAO swap? At the same terms as Round 2 perhaps idk.

a. defi and crypto in general tends to be correlated to ETH (and BTC). If you pair with USDC, then if ETH moons the token will likely follow (and vice versa) = far more IL. I’ll concede that this only matters if we are LMing tho. But also based on your twitter you’re a terrible trader so if you are bearish maybe ETH pairing is the right call :wink:

EDIT: but ya elaborating on point a, it’s weird pairing with USDC because you are basically saying that long term you are shorting in ETH in a way (at least compared to ETH pairing). If ETh is wrecked so is the rest of crypto so pretty much ALL valuations and capital available is down. But if ETH moons and we are paired with USDC, then a lot of upside. So I think you need to somewhat align with the rest of the market to avoid the most punishing scenario (ETH moon, paired with USDC) and in doing so open up to aligned downside.

b/c. will let others chime in here before making a final recommendation since I’m OOTL on part 2 as well

d. Agreed it’s worth chatting - will see if I can find a contact.

a+b+c = thank you, noted, I rest my case. Understood!

d. will try to find

Why don’t we look at using the LM budget for Olympus Pro bonds to purchase POL rather than renting off mercenaries?

I think pairing with USDC initially is the right call, especially during this market. If PMF is found and its decided to migrate some liquidity to a UNIv3 position an ETH pairing could be built then.

Agree with Ivan, if we can get a balancer pool that’s ideal.

Imho, POL is a meme for non-PCV focused protocols. It’s just continuous selling.

Agree for uncapped emissions. If we have 3% of the gear supply set aside for liquidity mining, this would be better used for acquiring POL rather than paying mercenary capital. (either all of the 3% or half of the LM rewards pool). Exactly like GMX, JPEG etc did.

Depends if we are talking long term or short term liquidity. If we need long term liquidity, we sell GEAR for ETH/USDC. If we need short term liquidity, we mine.

For long term liquidity, I think olympus pro is a bad approach. You are selling the token, at a discount. People who own the token will wait for a nice discount then sell it for ETH, bond it, let it vest, then do it again. So basically ALL of the tokens end up getting dumped on the open market anyway, ON TOP of selling them at a discount. OP is not the way IMO unless you want to give a select few users the ability to constantly dump tokens to earn high %APR on holding GEAR.

Gnosis auction fills the purpose of getting POL to start with, and then we can liquidity mine (straight up or thru bribes) for short term liquidity incentivies or do a treasury swap for vlAURA (with assumption it won’t be dumped) for long term liquidity incentives.

Also pet peeve but GMX literally uses half the ETH they bond to buy GMX on the open market to support the price like ??? you sell GMX at a discount to then buy it higher? Minor in the grand scheme of a promising protocol but that one thing is pretty dumb.

Hardcore disagree. Sorry.

The fancy phrase of “3% for POL” in reality is just “sell 3% at 150M (or less or more) market price”. As such, that is the maximum $$ it can attract. Rough numbers are, this would get (again, SELL) just $4.5M. That is useless for the protocol.

Instead, 3% at that markup can be a 4% APY for 100M of capital in the protocol. And once this gives a boost, potentially do some POL at much higher level if that is ever to be achieved. Much better use of capital.

Gearbox is not a PCV stablecoin. It needs major liquidity, not incremental $4M of owned liquidity. Fancy words of “abandon mercenaries” and stuff like that doesn’t actually contribute to the protocol itself.

1 Like

RU версия будет или гугл в помощь?

Наверное, гугл достаточно понятно может тут перевести - но если @nikitakle нечего делать в воскресенье, мб он сделает перевод. Хотя по идее лучше делать перевод пропозала как основного (так как сейчас это дискуссия) чтобы ему не делать работу дважды.

1 Like

All great, moving forward is the best thing !

As a budding participant in Fund Raise 2, I wouldn’t be supportive of of a min. of $150M FDV considering the fact that Fund Raise 1 & 2 tokens are vested and these tokens will be transferable in the near term. I understand market conditions may force $GEAR below cost basis for Funds 1 & 2 in near term (nothing we can do about that), but all things being equal, knowing what we do now and at what price is being paid for $GEAR today and the terms that come along with it (vesting) there’s no logical reason to price transferable $GEAR below 0.02c.

If I may make a suggestion here; We could offer to prioritise adding a Aura.Finance and/or Balancer pool to the AllowedList in return for directing votes (and thus incentives) toward our $GEAR/ETH LP for a period of time commensurate with the effort needed to action. All of course is subject to technical and safety evaluation of pools but their Balancer Boosted Aave pool is worth considering. We’d be trading development and risk analysis time for additional liquidity whilst driving TVL (and thus fees) to both protocols.

As for what happens with Fund Raise 2 $$. This is being collected in Stablecoins and at this point (and assuming the current KYC compliant proposal passes), I don’t think there’s an immediate need to direct the funds anywhere just yet until we evaluate the outcome of this liquidity raise and auction. Let’s see how Lev. Ninja, Gnosis Auction, V2 launch and transferability goes before determining what to do with side-lined funds; I’m afraid it’s too many spinning plates if you catch my meaning?

Just my thoughts; happy to hear what others think.


Noted on all, thank you for schooling. Have no objections to anything, makes sense.

I’m just worried of the scenario (in case market is in shit but transferability MUST happen) where GA does not get enough of LP funds and then the treasury must be eaten out to facilitate the liquidity… what do then? cc @ov3rkoalafied

Btw, what does this mean? It is “total”, right?

Anyway, I would say that $3M is genuinely too much. Would personally suggest to cap it at $2M max, maybe even $1.5M. We need to stop introducing new supply, there is no point in that anymore. Perhaps (if Part 2 DAO Round goes with $2M size) $500K from that can be used for liquidity.

To be fair this was sized before the latest round. If we drop to the $0.015 min price then we will hit 2.25m, which is about your desired cap. There is actually no cap since people can bid as high as they want. The liquidity is also in line with other protocols. These tokens are being sold for a very specific purpose, whereas the other round appears to be more of “well uhh other people got to buy so we wanna buy too” with no actual purpose for the funds. The proposal doesn’t even specify a use case. I think having 4-6m of liquidity is necessary, and if we have no use case for the funds from round 2 then we can use them all for liquidity and reduce the scale of the gnosis auction. But it doesn’t make any sense to me to reduce the sale of the gnosis auction for the purpose of raising funds that now magically aren’t being used for anything. Feels pointless.

Noted! Makes sense actually then, thanks for defending the viewpoint.