Pages:
Author

Topic: [ANN] Miner Control 1.6.1 - Auto profit switching miner controller - page 9. (Read 164304 times)

sr. member
Activity: 455
Merit: 250
When its mining by "Average" how is the average calculated?  What makes this better than mining by the Net value?

Just curious how it works, otherwise been running solid for weeks on my 1070 rigs.

Jake

The main purpose for average data - to reduce amount of switches. Profitability in MC depends on two determinants: exchange rate and coins difficulty. The fluctiations of both are rather high in "net" profitability. If miner switches two offten it has no time to get a full mine speed, and its efficiency drops. Averaging makes it smother.

aaand returning to pump and dump...just my thoughts.
as far as i can see - there is not so difficult math to insert this "pumpy-dumpy" feature i guess...
WTM has current exchange rate(cer), 24h exchange rate (24h) and so on...
so...why not to add some delta ticking once in a hour or close (delta=cer/24h) and then manipulate with cweight automatically.example
if delta= 0.95...1.05 - cweghtnew=cweightold(that means tha last hour it wasn't big fluctuation in price)
if delta= 0.90...0.95 - cweightnew=cweightold-(1-delta) (means coin is falling and need to correct it's weight)
if delta= 1.05...1.1 - cweightnew=cweightold(why so?i'm not optimist and always waiting for bad news=)
if delta <0.9 or delta >1.1 - cweightnew=cweightold-(1-delta)(cause i don't want to mine pumped coin like it was with decred)
that's is not absolute numbers...just an example how i see some "defend" from pumpy-dumpy=)you're not forced to do it.but if you can implement something like this in a very-very-very beta then i will test with a great pleasure=)

It can be done in such way, but as I wrote above, not only exchange rate affects the price, but difficulty also. Normally when ex.rate rises, more miners begin to mine it and difficulty increases proportionally and vice versa. Threfore profitabillity in Whattomine and in MC drops in this case. It should be self-regulating mechanism.
yes.but decred not such a popular coin i guess...cause raising price in 140% was up diff only at 20%...right now, when pump is over new miners begin to ask about it.not a big deal...but i want that feature anyway=)so...will wait.
full member
Activity: 133
Merit: 100
When its mining by "Average" how is the average calculated?  What makes this better than mining by the Net value?

Just curious how it works, otherwise been running solid for weeks on my 1070 rigs.

Jake

The main purpose for average data - to reduce amount of switches. Profitability in MC depends on two determinants: exchange rate and coins difficulty. The fluctiations of both are rather high in "net" profitability. If miner switches two offten it has no time to get a full mine speed, and its efficiency drops. Averaging makes it smother.

aaand returning to pump and dump...just my thoughts.
as far as i can see - there is not so difficult math to insert this "pumpy-dumpy" feature i guess...
WTM has current exchange rate(cer), 24h exchange rate (24h) and so on...
so...why not to add some delta ticking once in a hour or close (delta=cer/24h) and then manipulate with cweight automatically.example
if delta= 0.95...1.05 - cweghtnew=cweightold(that means tha last hour it wasn't big fluctuation in price)
if delta= 0.90...0.95 - cweightnew=cweightold-(1-delta) (means coin is falling and need to correct it's weight)
if delta= 1.05...1.1 - cweightnew=cweightold(why so?i'm not optimist and always waiting for bad news=)
if delta <0.9 or delta >1.1 - cweightnew=cweightold-(1-delta)(cause i don't want to mine pumped coin like it was with decred)
that's is not absolute numbers...just an example how i see some "defend" from pumpy-dumpy=)you're not forced to do it.but if you can implement something like this in a very-very-very beta then i will test with a great pleasure=)

It can be done in such way, but as I wrote above, not only exchange rate affects the price, but difficulty also. Normally when ex.rate rises, more miners begin to mine it and difficulty increases proportionally and vice versa. Threfore profitabillity in Whattomine and in MC drops in this case. It should be self-regulating mechanism.
full member
Activity: 133
Merit: 100
Package updated.

-added column "PoolFee". Shows percent charged by Pool. Consists of two parts
1) pool fee itself, key "fee" in config or got from API;
2) btcfee - fee for convertion to BTC for pools like Zpool and MPH

- layout improvements;
- Updated JSON to v10, new Newtonsoft.Json.dll;
- added Pascallite for Suprnova;

Link1:
https://drive.google.com/open?id=0Bz8-ln3QUcV4U3kzVXFzMUI0VUE

Link2:
https://mega.nz/#F!rhAFCbIa!F4XoHfIYoPYgJ-V1TUyP2A
sr. member
Activity: 455
Merit: 250
aaand returning to pump and dump...just my thoughts.
as far as i can see - there is not so difficult math to insert this "pumpy-dumpy" feature i guess...
WTM has current exchange rate(cer), 24h exchange rate (24h) and so on...
so...why not to add some delta ticking once in a hour or close (delta=cer/24h) and then manipulate with cweight automatically.example
if delta= 0.95...1.05 - cweghtnew=cweightold(that means tha last hour it wasn't big fluctuation in price)
if delta= 0.90...0.95 - cweightnew=cweightold-(1-delta) (means coin is falling and need to correct it's weight)
if delta= 1.05...1.1 - cweightnew=cweightold(why so?i'm not optimist and always waiting for bad news=)
if delta <0.9 or delta >1.1 - cweightnew=cweightold-(1-delta)(cause i don't want to mine pumped coin like it was with decred)
that's is not absolute numbers...just an example how i see some "defend" from pumpy-dumpy=)you're not forced to do it.but if you can implement something like this in a very-very-very beta then i will test with a great pleasure=)
sr. member
Activity: 455
Merit: 250
When its mining by "Average" how is the average calculated?  What makes this better than mining by the Net value?

Just curious how it works, otherwise been running solid for weeks on my 1070 rigs.

Jake
MC takes it from "statwindow" value.if you set it to 60 - then MC calculates avg profitability during last hour.if 120 - 2 hours and so on...
full member
Activity: 211
Merit: 100
When its mining by "Average" how is the average calculated?  What makes this better than mining by the Net value?

Just curious how it works, otherwise been running solid for weeks on my 1070 rigs.

Jake
sr. member
Activity: 455
Merit: 250
well...in this case DCR was pumping.and now it's falling.i was mining some on coinmine but it's not enough to withdraw it to exchange.(have 0.6 dcr, min 1 dcr withdraw).so i need to mine for 1 day coin that not profitable anymore just to withdraw it.so i'm freezing about 8$ at pool now.and again - was talking about decreasing coin weight on pump only and increasing it after sometime if this raising in price is continuing...
so the main idea for me in MC -
1.to mine the most profitable coin in long term
2.to mine as much altcoins with lowest diff as it possible at this moment, withdrawing them to an exchange and waiting them grow.
BUT i don't want to mine the coin(1) with small fluctuation in diff when profitability depends on exchange and it's manipulation cause i'm not so strong in trading...
BUT it's very good to sell some "easy"(2) coins on pumps later=)
full member
Activity: 133
Merit: 100
lyolyalya, I think that your idea is over limits for reliable automatization. To address multiple issues including this one I propose one easy (for the dev)) trick. Instead of taking profitability and/or price from whatever, add new "price" tag to each coin. If filled, MC should calculate profitability internally, out of diff, reward, freq of finding, all that stuff. And so we can manually set estimated price which takes into account all uncalculable things.

If done, MC will become perfect and I will use it again))

There is such option in MC, created by previous developers.  Calling "manual" pool. You can set any price manually. Thread is here
https://bitcointalk.org/index.php?topic=769239.msg10081087;topicseen#msg10081087

As for lyolyalya's idea, I guess it contradicts the main principle of MC to earn more with highest profit. If realised, it will be applied to all the coins, not only DCR. I mean here algos with fast confirmation periods, so it will bring no big advantage. Imagine, if DCR pump continued for some days more, you would receive an extra high profit after confirmations have finished.
I put this idea on hold for the moment.
newbie
Activity: 12
Merit: 0
lyolyalya, I think that your idea is over limits for reliable automatization. To address multiple issues including this one I propose one easy (for the dev)) trick. Instead of taking profitability and/or price from whatever, add new "price" tag to each coin. If filled, MC should calculate profitability internally, out of diff, reward, freq of finding, all that stuff. And so we can manually set estimated price which takes into account all uncalculable things.

If done, MC will become perfect and I will use it again))
sr. member
Activity: 455
Merit: 250
was thinking.can you add some option to nihilate pumps on exchange rates?right now we have a pump on DCR(50%) and MC switches to it.but we know - DCR has very slow confirmations so after 2-3 days not sure that i will have same earning as i expected.example.
if an exchange rate is on pump or dump up to 25% or more in 24 hours than need to reduce cweight in a 12.5% maybe or something like that to miss that pumps and dumps...of course it has to be user defined option...
sr. member
Activity: 455
Merit: 250
can you check DCR balances on suprnova?can't see them.sib,mona,lbc - fine.
sorry.had wrong userid in conf file.
sr. member
Activity: 455
Merit: 250
well, you are right.suprnova gives me AvgSp and others rapidly...strange, why zec and zcl on MPH both broken...
sr. member
Activity: 455
Merit: 250
Can you test beta8 + ewbf 0.3.2b as experiment?
same zeros...
pool reporting my hashrate well at pool's dashboard.
when starting ewbf miner says that api - disabled.was added --api - listen on 127.0.0.1:42000 but only zeros again.
full member
Activity: 133
Merit: 100
Can you test beta8 + ewbf 0.3.2b as experiment?
sr. member
Activity: 455
Merit: 250
API not working to EWBF 0.3.3 b
https://bitcointalksearch.org/topic/m.18314713
awww...okay.not a big deal...not mining equihash as much=)just reporting=)
newbie
Activity: 19
Merit: 2
sr. member
Activity: 455
Merit: 250
aaand seems like beta 8+ewbf 0.3.3+MC=issue reading AcSpWrk and TopAvgSp and Accept and Reject.zeros in all columns for zec and zcl...
It doesn't depend on miner you are using, but depends on the pool. If you mine with MPH, they were under DDOS attack last week, so statistics is very lagging there.
no problems until updating to beta 8+ewbf 0.3.3...and all other stats are fine.
full member
Activity: 133
Merit: 100
aaand seems like beta 8+ewbf 0.3.3+MC=issue reading AcSpWrk and TopAvgSp and Accept and Reject.zeros in all columns for zec and zcl...
It doesn't depend on miner you are using, but depends on the pool. If you mine with MPH, they were under DDOS attack last week, so statistics is very lagging there.
sr. member
Activity: 455
Merit: 250
aaand seems like beta 8+ewbf 0.3.3+MC=issue reading AcSpWrk and TopAvgSp and Accept and Reject.zeros in all columns for zec and zcl...
sr. member
Activity: 455
Merit: 250
but beta 7 was normal...
Pages:
Jump to: