Sun Jan 24 11:57:18 UTC 2016
status of (former) sell side gateway:
nud getliquidityinfo B | grep BETwD8nSjtj9ADSvej2na34xmsMYwPRymv -A 2
"BETwD8nSjtj9ADSvej2na34xmsMYwPRymv" : {
"buy" : 13082.98,
"sell" : 5127.7376
status of (former) buy side gateway:
nud getliquidityinfo B | grep BFGMPykfKxXZ1otrCZcsbnTwJjKHPP9dsP -A 2
"BFGMPykfKxXZ1otrCZcsbnTwJjKHPP9dsP" : {
"buy" : 20070.39,
"sell" : 10187.1611
Do you recognize how smooth liquidity provision has become?
The recent daily trading volume was still tens of thousands of USD daily, of which over 80% was on Poloniex.
The ALP was rarely zero, but some thousand USD value were still traded by the NuBots.
Now you might to glimpse how hard I had to fight to keep the peg - with almost no ALP support and a trading volume of up to $200,000 daily.
Back then we already had NuBots with market aware offsets, albeit in a manual way.
With CRFC to create T1.1 and NuBots on T1.2 and T2 the peg is safer than it is without it.
Market aware dynamic offsets might make that even more effective and at the same time cheaper.
It already works with static offset and some ALP (I haven’t touched the config of the NuBots for 2 days now).
Or do you remember the last emergency call to FLOT to deposit funds at the NuBot accounts?
Outlook
Please read and discuss this
Parts of it might be more controversial than others.
The part that deals with providing liquidity with a combination of ALP on T1.1 and NuBots on T1.2/T2 might be able to find consensus.
@masterOfDisaster. Thank you so much for your hard work over this last tumultuous period. Sorry for not being able to give the support that was needed, when it was needed.
I’m trying to catch up with everything that has happened but I can see that there is a definite need for a change in how liquidity is provided. I really do hope that the updates that are coming live up to the expectation and the need.
A lot of the topics that have been touched on by you and others are known. Trader pressure on the peg is real and is something that the Parametric order books of NuBot were specifically designed to combat (traders can still use NBT to trade, they just may have to pay a premium if they want to trade ‘Now’). By forcing the use of NuBot with the New ALP software I hope that approach to liquidity management will become more common.
You mention that the peg is safer with CRFC and NuBots than without. The good news on that is that the CRFC code is ready in ALPv2 and we are very close to a release of the NuBot code that contains the new ALP client. I’m sorry it hasn’t been ready before now as it may have lessened your load over the last few days, but it is nearly ready now. I just hope it lives up to expectations.
Bittrex buy side is low/empty.
I’m not in the position to change that
The trade volume there is 33% of total daily volume with roughly $10k.
Is there still consensus that sustaining dual side operations at all exchanges is a good idea?
Why not selling stamps anywhere, but buying them back only at specified places for guaranteed prices (maybe post offices), if you understand what I mean
And for the record: it wasn’t me (the one who traded at Bittrex)!
Nu is lucky that traders don’t fill more expensive orders. There’s almost nothing (~1 BTC) in between the current rate and $500 per BTC in the NBT/BTC pair there:
My complaints were rather directed towards NSR holders, who don’t contribute to the success of this endeavour and not towards those who already do so much.
I appreciate that you care!
I hope so, too. If they don’t we’ll analyze and refine
This is according to my current assessment of the situation based on the recent events.
CRFC can be a great line of defense on T1.1
NuBots can be a great line of defense on T1.2 and T2.
We’ve learned that NuBots alone can be sufficient. Do you perchance have the payout data for NuPool participants at Poloniex from 2016-01-15 to 2016-01-24 (total rewards per minute would be enough to derive the liquidity per minuted from it)?
I expect using a combination of CRFC ALP and NuBots to be cheaper, while being more efficient. Than one of them alone.
CRFC provides better incentives to put funds on T1.1 than the current scheme does. It should work. We’ll see.
The FC test run by @Nagalim on Bter look promising.
Besides I wouldn’t have known how valuable CRFC can be for such a situation, if I hadn’t learned it the (almost too) hard way.
Let’s learn from that together and improve the scheme together!
Maybe this is an indication that demand for NuBits at Bittrex is increasing, in which case we should keep supporting liquidity at this exchange?[quote=“masterOfDisaster, post:1256, topic:1239”]
We’ve learned that NuBots alone can be sufficient.
[/quote]
I am wondering what is the major difference between nubot and alp? Is that ALP does not put proceeds from one side to the other (besides the parametric aspect) ?
On the contrary, there are no BTC on the order book. Look at ALix or the pictures.
People dumped until the BTC were gone.
If they continue, the peg at Bittrex is lost.
Can you clarify a bit?[quote=“masterOfDisaster, post:1259, topic:1239”]
On the contrary, there are no BTC on the order book.
[/quote]
Oops. You are totally right. I have just got up from bed…
ALP is a conversation between the client with access to the private key used to place orders and a server that verifies the orders using a public key. MLP does not require the verification step and so is simpler, leaving just the client layer.
In theory, there is no reason at all to use ALP with 0% reward as opposed to using NuBot as an MLP.
As for bittrex, if anyone wants to sell me nbt for $0.999 id be happy to trade them some btc. This would of course be a great deal if nupool were running FC or CRFC. Currently, there is no motivation for a bittrex LP to trade with me to rebalance, so they won’t. There’s just no reward for balancing an under-target pool in the fixed reward model.
That comes on top of having no incentive to try earning 0.24% per day in turbulent times, when you can easily lose 10% in hours if hedgers use your funds successfully…
If you really think the peg on bittrex is lost, then you defacto must think that you can buy NBT there for less then $1. In that case, buy them there and sell them to me for a profit. If you can’t do that, then the peg isn’t broken.
And in this line of thinking, I’m actually starting to like @masterOfDisaster’s idea of only selling on exchanges like Bittrex and not buying.
But why should Nu afford to sustain a buy side liquidity provision there in this case?
Nu can save that money, the compensation for the sell side as well (like already explained: long-term facing some exchange default risk is cheaper than buying an insurance for it!) and just run a NBT entry gateway there.
And not only there, but at each exchange, except for the fully supported ones!
