BitShares Forum

Main => General Discussion => Topic started by: bitcrab on April 11, 2019, 06:10:22 pm

Title: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 11, 2019, 06:10:22 pm
I open this thread and poll to provide a place for community to express their opinion on bitCNY price feeding.

good price feeding is so important, without it we cannot make bitCNY pegged well enough, and cannot convince users that BTS is a dependent financial system.

I hope that anyone vote here can at the same time provide proof and explanation on your selection.

 
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: binggo on April 12, 2019, 12:16:18 am
(https://user-images.githubusercontent.com/34892308/56003792-d4022300-5cfa-11e9-908c-b465a6d5f95a.png)(https://user-images.githubusercontent.com/34892308/56003794-d6fd1380-5cfa-11e9-9904-7fa2fa88c2e2.png)

0.4292 cny can buy 1 bts?
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: lin5464 on April 12, 2019, 12:32:45 am
(https://raw.githubusercontent.com/lin5464/git-img/master/bts/20190412082222.png)
Elmato is still feeding 1.1 MSSR so long after bsip59 took effect.
This kind of witnesses who ignore the community consensus are still supported by the proxy, why?
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: ljk424 on April 12, 2019, 01:46:17 am
当前bitusd兑bitcny实际汇率已经高达7.2:1,但是很多见证人仍然粗暴的使用6.8的汇率,将bitusd喂价转换为bitcny喂价发布,无视真实的cny交易价格,对bitcny交易对的用户造成了极大的不公和伤害。
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 12, 2019, 02:59:12 am
update my selection:  now I vote magicwallet.witness, verbaltech2, sahkan-bitshares and elmato.

as shown below, at the moment DEX price is 0.4157bitCNY, bitCNY has about 1.8% premium,  a good feed price should be about 0.4157*1.018 = 0.423.

If an witness feed price which is 4%+ lower than such a price, I don't think he did a good job.

and elmato seems pay little focus on BTS now, he even feed MSSR=1.1 now, ignoring community consensus.

(http://i1.fuimg.com/523014/d68dbc94212d2c80.png)

(http://i1.fuimg.com/523014/a5a1dd158c890235.png)

(http://i1.fuimg.com/523014/1add5afb72db1112.png)

(http://i1.fuimg.com/523014/84f737a809d95dc6.png)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: sahkan on April 12, 2019, 08:10:09 am
I think you meant to say that you "unvoted" us. Seems to me that  everyone is within the specified 4%. Also I am one of the few (if not the only one) that uses paid price feeds so my feeds report changes a bit faster than the rest of the network.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: clockwork on April 12, 2019, 08:55:40 am
I think you meant to say that you "unvoted" us. Seems to me that  everyone is within the specified 4%. Also I am one of the few (if not the only one) that uses paid price feeds so my feeds report changes a bit faster than the rest of the network.

mine are paid too fwiw
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 13, 2019, 08:35:37 am
I think you meant to say that you "unvoted" us. Seems to me that  everyone is within the specified 4%. Also I am one of the few (if not the only one) that uses paid price feeds so my feeds report changes a bit faster than the rest of the network.

mine are paid too fwiw

paid price feeds does not mean good price feeds.

it's not difficult to know what is the exact BTS price at a moment without rapid price change.

when I do below screenshoting, DEX price is of 0.4266, bitCNY has a premium of about 1.8%, BTS price should be about 0.434CNY.
at huobi BTS has a price of 0.0629 USDT and in OTC market 1USDT = 6.92CNY, so BTS price should be 0.435CNY.
but both paid price feeds gave price 1%+ lower than the market price.


(http://i1.fuimg.com/523014/fa02b93dbfb497d7.png)

(http://i1.fuimg.com/523014/b8cee9b8f6dc7910.png)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: clockwork on April 13, 2019, 09:10:06 am
I think you meant to say that you "unvoted" us. Seems to me that  everyone is within the specified 4%. Also I am one of the few (if not the only one) that uses paid price feeds so my feeds report changes a bit faster than the rest of the network.

mine are paid too fwiw

paid price feeds does not mean good price feeds.

it's not difficult to know what is the exact BTS price at a moment without rapid price change.

when I do below screenshoting, DEX price is of 0.4266, bitCNY has a premium of about 1.8%, BTS price should be about 0.434CNY.
at huobi BTS has a price of 0.0629 USDT and in OTC market 1USDT = 6.92CNY, so BTS price should be 0.435CNY.
but both paid price feeds gave price 1%+ lower than the market price.


(http://i1.fuimg.com/523014/fa02b93dbfb497d7.png)

(http://i1.fuimg.com/523014/b8cee9b8f6dc7910.png)

you do realise the prices we publish  are not random right?

I could cherry pick a combination of exchange routes from USD -> BTS or CNY -> BTS that result in even lower price or even higher than the ones you mention.

Thats why we take all of them into account.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: clockwork on April 13, 2019, 09:19:12 am
Here...take your pick....

(https://i.imgur.com/ojdK4sA.png)


I'm getting tired of this.

We can't simply publish high prices because it suits some people.

same goes for low prices.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 13, 2019, 09:21:11 am
you do realise the prices we publish  are not random right?

I could cherry pick a combination of exchange routes from USD -> BTS or CNY -> BTS that result in even lower price or even higher than the ones you mention.

Thats why we take all of them into account.

my point is that we need to pay more weight on DEX price.
when market is stable and no unusual issues happen in DEX, price feed should not be lower than Pdex.
here, Pdex = DEX price in bitCNY * bitCNY premium/discount factor
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 13, 2019, 09:24:58 am
Here...take your pick....

(https://i.imgur.com/ojdK4sA.png)


I'm getting tired of this.

We can't simply publish high prices because it suits some people.

same goes for low prices.

that's why I suggest to feed price based on DEX price and bitCNY premium/discount.

who know where there is an CEX has a price of 0.399? if there is, just arbitrage between it and DEX.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: clockwork on April 13, 2019, 09:28:02 am
Here...take your pick....

(https://i.imgur.com/ojdK4sA.png)


I'm getting tired of this.

We can't simply publish high prices because it suits some people.

same goes for low prices.

that's why I suggest to feed price based on DEX price and bitCNY premium/discount.

who know where there is an CEX has a price of 0.399? if there is, just arbitrage between it and DEX.

i believe that's coinegg IIRC
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 13, 2019, 09:42:49 am
Here...take your pick....

(https://i.imgur.com/ojdK4sA.png)


I'm getting tired of this.

We can't simply publish high prices because it suits some people.

same goes for low prices.

that's why I suggest to feed price based on DEX price and bitCNY premium/discount.

who know where there is an CEX has a price of 0.399? if there is, just arbitrage between it and DEX.

and I greatly doubt that your calculation does not considered the premium of USDT, recently, 1USDT=6.92CNY but the official USD/CNY rate is about 6.71.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: binggo on April 13, 2019, 09:49:59 am
0.399 ::)?

 USD -> BTS? where can buy bts with USD?

 BTS ->BTC -> USD ->CNY? is this some of witness feed price?!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

 if can't make a true fiat price of BTS, please don't make the bitasset,

Now the feed price just become the fake price, i can't use the feed price to buy 1 bts from the CEX.

(https://user-images.githubusercontent.com/34892308/56077851-aa88ea80-5e13-11e9-9536-a66d8a38ebea.png)
(https://user-images.githubusercontent.com/34892308/56077854-bffe1480-5e13-11e9-8491-9036b90819a8.png)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 13, 2019, 03:13:59 pm
unvoted clockwork,  blckchnd,  delegate-1.lafona.

cannot bear this, bitCNY has a 1.8% premium at this moment, but the DEX price is always higher than the feed price!

(http://i2.tiimg.com/523014/2980fa660f0960db.png)

Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 14, 2019, 08:47:16 am
The reason for the premium is that (once again) shorters are deliberately keeping their positions in margin call territory, which means all orders below MSSR get eaten up by margin calls. Using the DEX price as the feed in this situation would be completely wrong, because the margin calls *define* the DEX price in terms of feed price and MSSR. There is your feedback loop.

Blaming this on the witnesses is short-sighted and unfair IMO.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 14, 2019, 09:10:28 am
The reason for the premium is that (once again) shorters are deliberately keeping their positions in margin call territory, which means all orders below MSSR get eaten up by margin calls. Using the DEX price as the feed in this situation would be completely wrong, because the margin calls *define* the DEX price in terms of feed price and MSSR. There is your feedback loop.

Blaming this on the witnesses is short-sighted and unfair IMO.

I don't care what cause the premium, the fact is the premium is there.

feed price + MSSR does not define the DEX price, we have seen a lot that there is no determination relationship between these. especially for bitCNY, which has a good depth that now margin call orders will be eaten instantly while they appear.

feed price is one key part of the whole rules, if feed price does not reflect the real market price, the whole rule sets will not work well.

feed price is serving DEX trading, DEX trading has enough big real volume, it's not acceptable that the feed price conflict with DEX price.

CEX price always involve many uncertain factors, naked short selling, faked volume, USDT premium...we need to be very careful while calculate feed price from CEX price.


Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: xeroc on April 14, 2019, 02:14:52 pm
Blaming this on the witnesses is short-sighted and unfair IMO.
I agree with @pc
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: binggo on April 14, 2019, 02:30:31 pm
Blaming this on the witnesses is short-sighted and unfair IMO.
I agree with @pc

If they can't find the “fair” feed price, they should be blamed as they accepted the work.

The feed price is the foundation of Market Pegged Assets, a “fair” feed price is important.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 14, 2019, 03:41:03 pm
I don't care what cause the premium, the fact is the premium is there.

You can't repair it if you don't find out why it's broken.

And IMO it is quite obvious why it's broken: https://wallet.bitshares.org/#/asset/CNY lists over 200 short positions with a CR of less than 1.8. Whenever BTS goes down a tiny bit in relation to CNY, all of these get margin called, which eats all orders up to MSSR from the market.

Also, almost all of them have set a target CR of 1.75, so it's clear that these short positions are *deliberately* kept at such a low collateralization.

Fiddling with the feed price is not going to change any of this. Effectively you are trying to re-apply BSIP-42, with the same reasoning, the same justification and probably also the same outcome.

If they can't find the “fair” feed price, they should be blamed as they accepted the work.

Like I said, in the presence of skewed external markets there is no such thing as *the correct price*. What you call fair may be unfair on others.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 14, 2019, 05:44:17 pm
I don't care what cause the premium, the fact is the premium is there.

You can't repair it if you don't find out why it's broken.

And IMO it is quite obvious why it's broken: https://wallet.bitshares.org/#/asset/CNY lists over 200 short positions with a CR of less than 1.8. Whenever BTS goes down a tiny bit in relation to CNY, all of these get margin called, which eats all orders up to MSSR from the market.

Also, almost all of them have set a target CR of 1.75, so it's clear that these short positions are *deliberately* kept at such a low collateralization.

Fiddling with the feed price is not going to change any of this. Effectively you are trying to re-apply BSIP-42, with the same reasoning, the same justification and probably also the same outcome.

If they can't find the “fair” feed price, they should be blamed as they accepted the work.

Like I said, in the presence of skewed external markets there is no such thing as *the correct price*. What you call fair may be unfair on others.

I feel you misunderstood what I mean, I do not want to "repair" the premium here, I just want the feed price reflect the right price.

it's quite easy to know the "right" price, at least when I am sure that bitCNY has premium and your feed price is far lower than DEX price definitely you are wrong.

at least we need to ensure that the feed price do not conflict with DEX price.

orders are filled every minutes, some margin call orders does not change the fact.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 14, 2019, 07:15:19 pm
If I were a trader and saw that margin calls happen regularly, I would place my sell orders 2% above the price at which the least collateralized short position is margin called, for an instant 2% profit. If I wanted to buy CNY on the DEX I would place my order 2% above that price, because otherwise I have no chance of getting it filled.

Thus, margin calls and feed price and MSSR all interact with each other. The DEX price is driven by margin calls, which are in turn driven by the feed price and MSSR.

If you reduce the MSSR to 1% the premium will also go down to 1%. Does changing MSSR change the "fair" price of bitCNY? I think not, but according to your logic the fair price would then be 1% higher than it is now, not 2%.

On an external exchange you don't have that effect, because the external exchange is not influenced by margin calls in the way that the DEX is. The result is that the external trade price is much more natural than the DEX price, and so can more reasonably be called a "fair" price.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: binggo on April 14, 2019, 11:41:39 pm
The result is that the external trade price is much more natural than the DEX price, and so can more reasonably be called a "fair" price.

So if the witness can get the "fair" price from the CEX? half of the witness give the lower feed price than the real price, the witnees think they give the "fair" price, can they prove it?

The fact is not, so what we are taking about? now the feed price lower more 2% than the "fair" price, is it fair for the margin call? make the lower feed price is very dangerous for the Market Pegged Assets, harming the balance.

now the "fair" price is 0.4380, what are they doing? just give a BTC price?!

(https://user-images.githubusercontent.com/34892308/56100963-8501fb00-5f51-11e9-83ad-98fb3ef18ccd.png)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 15, 2019, 01:23:21 am
If I were a trader and saw that margin calls happen regularly, I would place my sell orders 2% above the price at which the least collateralized short position is margin called, for an instant 2% profit. If I wanted to buy CNY on the DEX I would place my order 2% above that price, because otherwise I have no chance of getting it filled.

Thus, margin calls and feed price and MSSR all interact with each other. The DEX price is driven by margin calls, which are in turn driven by the feed price and MSSR.

If you reduce the MSSR to 1% the premium will also go down to 1%. Does changing MSSR change the "fair" price of bitCNY? I think not, but according to your logic the fair price would then be 1% higher than it is now, not 2%.

On an external exchange you don't have that effect, because the external exchange is not influenced by margin calls in the way that the DEX is. The result is that the external trade price is much more natural than the DEX price, and so can more reasonably be called a "fair" price.

premium is not a fixed value, 1.8% is the value in these 2 days, 4 days before, the premium is about 0.5%. you can only get this value in the market.

MSSR has impact on premium but it does not determine the premium.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: lin5464 on April 15, 2019, 02:14:36 am
Every time people in communities outside China talk about the premium of bitcny, they seem to have no idea about the premium at all. I think this should be because they have never tried to buy BTS with direct RMB. They simply cannot understand what the Chinese community has been complaining about.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 15, 2019, 02:43:42 am
Every time people in communities outside China talk about the premium of bitcny, they seem to have no idea about the premium at all. I think this should be because they have never tried to buy BTS with direct RMB. They simply cannot understand what the Chinese community has been complaining about.

it's easy to understand if  they want to.

you can get the premium simply if you install magicwallet in your mobile.

(http://i2.tiimg.com/523014/1e570c9d93917d8a.png)

(http://i2.tiimg.com/523014/77d40cb6e96d2f51.png)

you can also contact magicwallet team to request an API to get the data with program - yes, you need to do that if you are a witness.
 
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 15, 2019, 06:27:52 am
we are always telling the traders like this:

"hi guys, look at this, now you need to pay 0.441*1.015(current bitCNY premium) = 0.4476 CNY to buy 1BTS, but if you margin call price is above 0.4388, your debt position will be margin called, so wonderful, right?"


(http://i1.fuimg.com/523014/981f3f3aa2417782.png)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 15, 2019, 07:36:14 pm
So if the witness can get the "fair" price from the CEX? half of the witness give the lower feed price than the real price, the witnees think they give the "fair" price, can they prove it?

You cannot prove something to be "fair" because "fair" is subjective. Witnesses can explain how they calculate their feeds. Some will find the result fair, some won't.

premium is not a fixed value, 1.8% is the value in these 2 days, 4 days before, the premium is about 0.5%. you can only get this value in the market.

MSSR has impact on premium but it does not determine the premium.

The effect I described applies only when BTS price is going down relative to CNY. If the BTS price fluctuates, the premium will fluctuate as well and can vanish completely if BTS is in an uptrend.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: sahkan on April 15, 2019, 10:46:41 pm
First off, I would like to say that it is important to me as a witness to do as much as I can to help Bitshares and businesses based on Bitshares to succeed. You and other companies put a lot of work into this and by no means am I ever trying to intentionally derail those efforts. You can rest assured that I care about all my feeds and I want them to be as accurate as possible (thus the paid services), and I do that across all price feeds on the DEX. In the past few months we have done a lot of changes on the Bitshares to accommodate BSIP42, MSSR changes etc. Sometimes I don't agree with those changes but as the witness I follow the majority (or whatever proposal gets voted in). As an example I wrote a warning about BSIP42: https://bitsharestalk.org/index.php?topic=20792.msg322796#msg322796 (https://bitsharestalk.org/index.php?topic=20792.msg322796#msg322796) and a Black Swan on bitUSD and unfortunately we did have the Black Swan because BSIP42 was voted in.
I started to monitor my CNY feed more closely and will work with it to make sure that the feeds reflect true pricing from the markets. But let us remember that few months ago price feeds were within 4% range from the median and now most of the feeds are within 1% with spikes to 2%. The DEX with a settlement of 24hrs was not designed for REAL TIME PRICING, most witnesses feed every 20 -40 minutes.

2nd off this is just a personal comment, I think all witnesses have an obligation to provide a reliable infrastructure and price feeds for ALL committee owned smart coins. Is it too difficult for some to provide more than 2 price feeds?

and 3rd off, just wanted to say that your approach to constantly vote and unvote witnesses to pressure them to see things your way is not really well received. I think we are all open to conversations to make things better and improve on the things that we do, because as I said; we all want Bitshares to succeed and bitCNY is big part of that.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 21, 2019, 01:17:33 pm
I am considering to push one principle for bitCNY price feed -  price feed should not conflict with DEX price.

exactly speaking, price feed should not conflict with DEX price modified with bitCNY premium/discount.

as shown below, at the time bitCNY has a premium of about 1.7%, so the price feed should be about 0.4069*1.017=0.4138

(http://i1.fuimg.com/523014/b9a932a1c85c2372.jpg)

but most of the witnesses feed a price which is even under 0.4069 at the time, this is killing the bitCNY ecosystem!

(http://i1.fuimg.com/523014/5139f182c2937aaf.jpg)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: sahkan on April 21, 2019, 10:27:53 pm
Even though currently the DEX price more or less reflex the market price; based on buys and sells it can spike significantly in relation to global markets. And of course we can't just feed the DEX price by itself. I have been working on my tuning my prices a bit. I  have also noticed that GDEX updates their price feeds every 2 minutes. Is that were we are trying to go with this? Provide frequent up to a minute feeds?
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: ljk424 on April 22, 2019, 03:25:04 am
witnesses are the god of BTS.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 22, 2019, 03:48:21 am
Even though currently the DEX price more or less reflex the market price; based on buys and sells it can spike significantly in relation to global markets. And of course we can't just feed the DEX price by itself. I have been working on my tuning my prices a bit. I  have also noticed that GDEX updates their price feeds every 2 minutes. Is that were we are trying to go with this? Provide frequent up to a minute feeds?

I don't think each witness should feed every 2 minutes, however it is not acceptable that the market price have changed a lot but the feed price is not updated for long time.

feed price should not conflict with DEX price modified with bitCNY premium/discount, otherwise feed price will be the tool for shorting.

Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 22, 2019, 06:28:32 am
Is that were we are trying to go with this? Provide frequent up to a minute feeds?

*Publishing* a feed every 2 minutes is normally not required.

*Checking* the feed often is a good idea though. Technically you need to publish only if
a) your feed is about to expire, or
b) if the new feed price would change your position relative to the median (i. e. old price was below median and new price is above, or vice versa), or
c) your feed price *is* the median

You may want to publish more often than that to avoid being accused of neglecting your duties. :-/
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: alt on April 22, 2019, 11:52:19 am
you are definitely wrong.
every witness should input the feed price without any dependence from other witness or median price.
this is what the meaning of "witness".

here is a possible result when follow your suggest:
suppose there are 3 witness, a gives price 0.059USD, b gives 0.06 USD, c gives 0.061 USD
when the price rise, b change to 0.061, c change to 0.062, a get a new price at 0.06, so he don't need to update feed price.
when b and c goes to 0.062, 0.063, a still keep unchange.
...
when b and c goes to 0.1, 0.101, a still keep unchange.
now b and c got a chance to control median price between 0.059 to 0.1.

for 21 witness, when there are 11 witness always get a little bit higher price, while other 10 always get a little bit lower, this could happen too.

Is that were we are trying to go with this? Provide frequent up to a minute feeds?

*Publishing* a feed every 2 minutes is normally not required.

*Checking* the feed often is a good idea though. Technically you need to publish only if
a) your feed is about to expire, or
b) if the new feed price would change your position relative to the median (i. e. old price was below median and new price is above, or vice versa), or
c) your feed price *is* the median

You may want to publish more often than that to avoid being accused of neglecting your duties. :-/
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: sschiessl on April 22, 2019, 04:21:49 pm
 I have two questions:

If I offboarding 1 bitCNY to my bank account, how much CNY will I get?

If I onboard now 1 CNY, how much bitCNY do I get?

This must include all fees in the process.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: zhouxiaobao on April 22, 2019, 04:26:54 pm
      The feed price should be  “how much RMB does it cost to buy 1 BTS?”
      But ,all along, I have been unable to use RMB to buy BTS  at this feed price .It is too low.
     (DEX price *bitCNY premium )is the actual price I paid.
     It is totally unfair that the feed price is lower than the price I actually paid.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on April 22, 2019, 04:31:05 pm
I have two questions:

If I offboarding 1 bitCNY to my bank account, how much CNY will I get?

If I onboard now 1 CNY, how much bitCNY do I get?

This must include all fees in the process.

the deposit/withdraw fee in magicwallet can answer these questions.
you can mail to pluswave@magicwallet.io to get API for that info.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: zhouxiaobao on April 22, 2019, 04:34:39 pm
Is the price of the DEX fake? The daily turnover is tens of millions!
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: Thul3 on April 22, 2019, 04:59:05 pm
Quote
Is the price of the DEX fake? The daily turnover is tens of millions!

Increasing the feedprice would take with a MSSR of 1.02 the last rest of possible arbitrage from margin calls.

I monitored feedprice ,cex price (from mainly real volume excluding the big fake ones) and will post my thoughts after holiday.

However i have to disagree with bitcrab.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: sahkan on April 22, 2019, 09:07:04 pm
There are many crypto exchanges around the world. They trade many coins and fiat currency, and they charge different fees. For example coinbase charges about .5% to trade and there are withdrawal fees. We can't feed prices based on all the fees that traders pay to acquire a particular coin, it would be impossible to get an accurate feed that way. So this is not how the feed price works or should work:

      The feed price should be  “how much RMB does it cost to buy 1 BTS?”
      But ,all along, I have been unable to use RMB to buy BTS  at this feed price .It is too low.
     (DEX price *bitCNY premium )is the actual price I paid.
     It is totally unfair that the feed price is lower than the price I actually paid.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: zhouxiaobao on April 23, 2019, 02:42:02 am


you do realise the prices we publish  are not random right?

I could cherry pick a combination of exchange routes from USD -> BTS or CNY -> BTS that result in even lower price or even higher than the ones you mention.

Thats why we take all of them into account.

       Don't you think there is no handling charge for such many price changes?All major CEX give you free exchange?
       This is why the feed price thus calculated is always surprisingly low.
        The feed price should not be lower than the actual price paid by investors to buy a BTS in the DEX  market.
        In fact, the feed price should take the highest price of several major CEX and DEX, because this is the actual amount of RMB paid by investors.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: zhouxiaobao on April 23, 2019, 03:08:24 am
There are many crypto exchanges around the world. They trade many coins and fiat currency, and they charge different fees. For example coinbase charges about .5% to trade and there are withdrawal fees. We can't feed prices based on all the fees that traders pay to acquire a particular coin, it would be impossible to get an accurate feed that way. So this is not how the feed price works or should work:

      The feed price should be  “how much RMB does it cost to buy 1 BTS?”
      But ,all along, I have been unable to use RMB to buy BTS  at this feed price .It is too low.
     (DEX price *bitCNY premium )is the actual price I paid.
     It is totally unfair that the feed price is lower than the price I actually paid.

       This is ridiculous!
       For example,There are two exchanges, one handling fee is 10%, BTS price is 0.9; The other one handling fee is 1%, BTS's price is 1. How to determine the feeding price?0.95?
       
       It is true that there are too many transactions that cannot be counted accurately. should we take the higher one more reasonably?
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 23, 2019, 05:54:04 am
you are definitely wrong.
every witness should input the feed price without any dependence from other witness or median price.
this is what the meaning of "witness".

here is a possible result when follow your suggest:
suppose there are 3 witness, a gives price 0.059USD, b gives 0.06 USD, c gives 0.061 USD
when the price rise, b change to 0.061, c change to 0.062, a get a new price at 0.06, so he don't need to update feed price.
when b and c goes to 0.062, 0.063, a still keep unchange.
...
when b and c goes to 0.1, 0.101, a still keep unchange.
now b and c got a chance to control median price between 0.059 to 0.1.

for 21 witness, when there are 11 witness always get a little bit higher price, while other 10 always get a little bit lower, this could happen too.

Is that were we are trying to go with this? Provide frequent up to a minute feeds?

*Publishing* a feed every 2 minutes is normally not required.

*Checking* the feed often is a good idea though. Technically you need to publish only if
a) your feed is about to expire, or
b) if the new feed price would change your position relative to the median (i. e. old price was below median and new price is above, or vice versa), or
c) your feed price *is* the median

You may want to publish more often than that to avoid being accused of neglecting your duties. :-/

In your example a majority of witnesses collude to manipulate the feed price. That always works, independent from what the minority does.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: alt on April 23, 2019, 07:58:06 am
In your example a majority of witnesses collude to manipulate the feed price. That always works, independent from what the minority does.
nobody manipulate the feed price, they just been supposed to follow your suggestion, didn't update price when it can't change the median price.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 23, 2019, 01:29:03 pm
Then I do not understand where you see a problem.

If b feeds 0.1 and c feeds 0.101 then it doesn't matter if a feeds 0.06 or 0.09.

It is only important that a *checks* frequently what the price is. Then if b suddenly feeds 0.059, a can update quickly.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: alt on April 23, 2019, 01:45:28 pm
Then I do not understand where you see a problem.

If b feeds 0.1 and c feeds 0.101 then it doesn't matter if a feeds 0.06 or 0.09.

It is only important that a *checks* frequently what the price is. Then if b suddenly feeds 0.059, a can update quickly.
a feeds 0.05, b feeds 0.1, c feeds 0.101, then either b or c got a chance to manipulate the median price from 0.101 to 0.05.
if there are 21 witness, 10 feeds 0.05, 11 feeds 0.10, either of the 11 have a chance to maniuplate the median price to 0.05.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: pc on April 24, 2019, 06:24:29 am
If the median is manipulated to 0.05 then a will update his feed when he checks next. That's why I said you should *check* frequently.

I admit that in your example the feed would be very much off for a short period of time, which could have catastrophic effects. It is an extreme example though, and I think that in practice such a situation would not occur.

Still, it makes sense to add another rule:

d) if your calculated feed price is off from your published feed price by more than 1-2%, you *should* publish the new price.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: alt on April 24, 2019, 06:53:20 am
correct, witness should compare with his old feed price.
there will be a big problem when we got a big wrong feed price even only 1 sec.

If the median is manipulated to 0.05 then a will update his feed when he checks next. That's why I said you should *check* frequently.

I admit that in your example the feed would be very much off for a short period of time, which could have catastrophic effects. It is an extreme example though, and I think that in practice such a situation would not occur.

Still, it makes sense to add another rule:

d) if your calculated feed price is off from your published feed price by more than 1-2%, you *should* publish the new price.
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: binggo on May 09, 2019, 12:13:07 am
(https://user-images.githubusercontent.com/34892308/57416204-c96f7680-7231-11e9-9c20-85f87db88170.png)

Who is the bad guy?
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: binggo on May 09, 2019, 12:48:14 am
If you can't provide a quality feed price and  timely view and maintenance it, and didn't understand how important to the smarcoin,  please withdraw the witness position.

and this also talk to some committee and big Proxy,don't play like a zombie.

The witness should have a threshold,which one he want to become a witness should lock 5M bts in the system, i didn't want someone didn't have 1 bts to become the witness and sunk blood from the system.

If the fuction of the witness is producing the block only,give Snoopy a computer it will also can do this.

How about let Snoopy become the witness?

(https://user-images.githubusercontent.com/34892308/57416931-eb1e2d00-7234-11e9-9fb5-f6861ca631b4.png)
(https://user-images.githubusercontent.com/34892308/57421215-6daaea80-723d-11e9-9877-094f745f0985.png)
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: finn-bts on May 09, 2019, 01:01:10 am
If you can't provide a quality feed price and  timely view and maintenance it, and didn't understand how important to the smarcoin,  please withdraw the witness position.

and this also talk to some committee and big Proxy,don't play like a zombie.

The witness should have a threshold,which one he want to become a witness should lock 5M bts in the system, i didn't want someone didn't have 1 bts to become the witness and sunk blood from the system.

If the fuction of the witness is producing the block only,give Snoopy a computer it wil also can do this.

(https://user-images.githubusercontent.com/34892308/57416931-eb1e2d00-7234-11e9-9fb5-f6861ca631b4.png)
I agree with you
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: dirnet on May 09, 2019, 03:06:04 am
If you can't provide a quality feed price and  timely view and maintenance it, and didn't understand how important to the smarcoin,  please withdraw the witness position.

and this also talk to some committee and big Proxy,don't play like a zombie.

The witness should have a threshold,which one he want to become a witness should lock 5M bts in the system, i didn't want someone didn't have 1 bts to become the witness and sunk blood from the system.

If the fuction of the witness is producing the block only,give Snoopy a computer it wil also can do this.

How about let Snoopy become the witness?

(https://user-images.githubusercontent.com/34892308/57416931-eb1e2d00-7234-11e9-9fb5-f6861ca631b4.png)
(https://user-images.githubusercontent.com/34892308/57421215-6daaea80-723d-11e9-9877-094f745f0985.png)
I agree
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: finn-bts on May 09, 2019, 04:26:46 am
If you can't provide a quality feed price and  timely view and maintenance it, and didn't understand how important to the smarcoin,  please withdraw the witness position.

and this also talk to some committee and big Proxy,don't play like a zombie.

The witness should have a threshold,which one he want to become a witness should lock 5M bts in the system, i didn't want someone didn't have 1 bts to become the witness and sunk blood from the system.

If the fuction of the witness is producing the block only,give Snoopy a computer it wil also can do this.

How about let Snoopy become the witness?

(https://user-images.githubusercontent.com/34892308/57416931-eb1e2d00-7234-11e9-9fb5-f6861ca631b4.png)
(https://user-images.githubusercontent.com/34892308/57421215-6daaea80-723d-11e9-9877-094f745f0985.png)
I agree
pretty good idea
Title: Re: Which witnesses do not doing well in CNY price feeding?
Post by: bitcrab on May 09, 2019, 05:28:16 am
开了个新帖:https://bitsharestalk.org/index.php?topic=28383.0
 (https://bitsharestalk.org/index.php?topic=28383.0)
以后喂价问题去那里提供证据,投诉喂价不合理的见证人