[Passed] NuPool 9

[Nagalim]

The cost is what matters, the target is essentially meaningless for this operation.

@JordanLee i hope you reference the current spread motion in the new one and don’t just steamroll the current legislation.

We are currently paying 35 nbt/day for nupool on poloniex. We pay 150 nbt/day for nulagoon. We pay 8 nbt/day for nupond. What is our budget? How much bigger is nupool than nupond? I’d suggest we can spend between 200 and 300 nbt/day on liquidity. In that case, we can probably increase nupool cost on polo to something like 50 nbt/day.

[Cybnate]

prefer to see the target (and cost) doubled to 30,000

I don’t think that is wise. We should keep the target more or less the same imo.
The benefits of increasing targets are too limited.

while the tolerance is reduced to 0.005

Let’s start with that and see whether the trading will be increasing as assumed.
More trading makes money for the LPs, so the focus should be here in the first place.

[Nagalim]

Since Bittrex is often not at target, lowering target while keeping the ‘cost’ the same will have the effect of increasing provision by withholding a smaller portion of the funds than with a higher target.

[cryptog]

So is the target of 15k on Polo enough? And why?
On which criteria do we need to rely on to make a decision?

[huafei]

guys

check stauts of your nubot

"poloniex": {"btc": {"ask": {"rank_1": 2.85499024, "rank_2": 0.0, "total": 2.85499024}, "bid": {"rank_1": 10743.01812202, "rank_2": 0.0, "total": 10743.01812202}, "total": 10745.873112259998}

[woolly_sammoth]

Morning everyone, a busy weekend meant no chance for some computer time to update the proposal. Given the last few comments though that’s perhaps no bad thing.

I would agree with Jordan in this thread. LArge amounts of liquidity are the headline for NuBits. It may be that depleted walls are regularly topped up so that liquidity is kept to an average level in the long term but it is only those who are familiar with the inner workings of Nu that will understand this. The general trader on an exchange will only see transient walls.
@Nagalim has a point about the target in a CRFC system being fairly arbitrary, there is still an incentive on liquidity providers to provide as much as they can to get a larger percentage of the reward. However I think it does serve a useful purpose. It broadcasts a figure that we would expect to see on the exchange at most times. I’m still not totally sure about what to do with targets. traditionally the pool fee for NuPool has been 1% of the target so I’m not convinced that I have a completely impartial view.

@Nagalim is also correct in his assessment of the Tiers. In the last operation, NuPool operated with just 2 tiers. Tier1 was everything within the 1.05% tolerance. Tier 2, everything outside of that. The design of the ALPv2 software allows for any number of tiers, each with their own tolerance and allowing for a specific reward for each. The figures associated with the tiers can be updated dynamically during the operation instead of being set in stone during the proposal phase. This could allow for a tightening or relaxing of the peg as required, we as shareholder just need to decide which triggers cause which actions. There is some work needed on the feedback mechanism in NuBot so that it can react to changes in targets and rewards and place funds in a way that corresponds with the LPs appetite for risk vs reward.

In light of the discussions around liquidity I am altering the OP.

• The target will be less of a factor in deciding the reward and fee. The rewards will reflect the support targeted to the different exchnages rather than being a function of the target.
• The targets will be reduced to make it easier for the full reward to be shared between participants.
• The tolerance will be reduced to 1%.
• The operational period will be reduced to 30 days as I expect there to be more discussion around these issues over the coming weeks and being able to respond to the outcomes in 30 days makes more sense than waiting 60.

The effect of the changes mentioned will be to reduce the cost to Nu of this operation. They also make more sense in a CRFC system. The lowering of the targets should make it easier for LPs to make the full reward available.

I will leave the Daft tag on the proposal so that further conversation can take place around these changes.

[Nagalim]

reduce the cost to Nu

But none of those 4 points mentioned cost…

[woolly_sammoth]

Not directly no, but through decoupling the reward, fee and target I have reduced the costs incurred.
They have been rationalised to make the maths easier.

[Nagalim]

So for everyone’s understanding:
Targets in CRFC are a minimum, as opposed to FR where they are maximums. This is a fundamental difference. We should expect to pay higher premiums to establish a minimum than when we were establishing maximums.

[huafei]

@woolly_sammoth

not add the reward ?

09:43:00.852 [ALPTask] ERROR - Method=ALP createRequest failed too many times and timed out. attempts = 8 [c.n.n.t.ALPTas

[woolly_sammoth]

Do you have some more information about the error? which exchange is this for? it shouldn;t have anything to do with the reward.

[huafei]

09:48:00.437 [ALPTask] ERROR - Method=ALP createRequest failed too many times and timed out. attempts = 8 [c.n.n.t.ALPTask:94]
09:49:00.654 [ALPTask] ERROR - Method=ALP createRequest failed too many times and timed out. attempts = 8 [c.n.n.t.ALPTask:94]
09:49:31.165 [priceTriggerTask] ERROR - IOException: java.net.ConnectException: Connection timed out [c.n.n.t.w.PoloniexWrapper:964]
09:49:31.166 [priceTriggerTask] WARN  - Cannot save orderID of BUY wall, problem placing it : ApiError [4 : Null Return] [c.n.n.t.TradeUtils:240

at polo

[woolly_sammoth]

Well, I’m not seeing any errors in the ALP logs, There are others placing orders at polo successfully too. Could ytou restart your NuBot please?

[huafei]

always restart my nubot , but get the error few minute late

{“provided”: 0.0, “reward”: 0.0, “round_time”: 1463410429}

do not know the error from POLO or ALP server ,
only my bid order show at the status page

[Nagalim]

So this new contract would give 28.8 nbt/day for poloniex, which is a 17% reduction of poloniex support.

[woolly_sammoth]

@huafei. It seems there are two issues. NuBot attempts to make connection with the NuPool server and fails after 8 attempts. It also then seems to have an issue connecting to the Poloniex API. Is your internet functioning correctly?

@Nagalim That;s correct the reward has come down to save cost. In return the target has been lowered so that the full reward ids accessed much sooner for providers.

[huafei]

Is your internet functioning correctly?

my internet is good , only see this issue for today, will see what happen after couple hours

[zoro]

Tolerance: 0.01 means offset 0.01 in each wall thus spread 2%?

[huafei]

Tolerance=(offset+fee)*2

[Nagalim]

There are concepts of deviation. With a 0.1% tolerance, you’re likely to only get something like 0.6% offset, which is like 1.2% spread. It depends on the inner workings of the ALP and where the fee comes in. I’m not familiar enough with it to know how it handles deviation.