[ANN][AUTO-SWITCH] Profit-switch auto-exchange pool: CleverMining.com

[pdaddy]

My last 12 hours on CleverMining. I'm am happy with he pool. Thanks Terk for the hard work!

http://i58.tinypic.com/15843lj.jpg

Btw not once did the pool disconnect in those last 12 hours.  

[1714102836]

1714102836

[1714102836]

1714102836

[1714102836]

1714102836

[Alexoc]

I think it will be more profitable if we don't mine Doge. We are to small. Please deaktivate doge.

[thedue]

I think it will be more profitable if we don't mine Doge. We are to small. Please deaktivate doge.

Just wait... it´s the right decision to mine doge or at least to test it!

Thanks Terk!! Incredible job!

[Vbs]

Started mining on this pool and my discard rate percentage is crazy.

Getworks   Accepted   Rejected   Hardware Errors   Utility   Discarded
337           39           9           0                   1.765           669


scan time = 28
expiry = 7
queue = 1

even at queue 0 it was this bad.. tried 30/30 on scan and expiry and no diff..

is this normal for the pool ?

Scan time  and Expiry shouldn't really have any effect on a Stratum pool.  But I wouldn't set my Expiry that low in any case.  Put them back to default Scan = 30, Expiry = 120 and see what happens.

You really only need to play with these setting if your solo mining or using a Getwork pool, so you would have a low scan time to detect new blocks.  But Stratum pools will tell you when there is a new block, so it's not necessary.

+1!

Try setting both to 9999 and queue to 0 (remember that the staged work, ST, is always 1+queue).

Your discarded stats are normal, as staged work is discarded on "new blocks" (real new blocks for the mined coin or coin switches, displayed as NB on cgminer's stats).

Discarded work approaches ST*Getworks. As you were using extra staged work in queue = 1 (meaning ST=2, two staged work items), you should get around 2*337 = 674 discards.

[Terk]

If you don't plan to switch away in a hour or three, then don't worry about us mining DOGE and the variance. There will be several hours worse than average but then there will be a reward like this one:



Of course we are not mining it all the time - only if it's significantly more profitable than other coins. When it's not, we are still mining other coins. This should give better profits in long-term (say 48/72-hour average).

I am aware that it doesn't look encouraging when you look at the 24-hour chart but I decided that your profits are more important than this. Most of users mining here now are not people who jump for an hour and then switch away. Most of you are mining here for days, so I decided it won't affect you in a negative way.

I am still trying to keep a balance between profitability and variance. If DOGE is only 5% more profitable than some other coin, then we will be mining this other coin. This is to ensure the flow of Bitcoins from the pool to the users won't stop. Our variance while mining DOGE is mathematically so big, that the 5% difference doesn't matter, so it's not enough profitable to switch to DOGE. With our average expected time to find a DOGE block, it is highly likely that DOGE price will change by more than this 5% (up or down) before we will find a block. This is why we switch to DOGE only if there is a bigger profitability gap between it and the second best coin. This results in percentage of our hashpower mining DOGE varying between 20% and 80% in different hours and that's fine. We mine DOGE when it's really much more profitable but we still mine other coins if they're more profitable or when the difference isn't that big.

For example, the hour that just ended had 128% LTC profitability, so we're doing our business as usual between times spent on mining DOGE. And time spent on mining DOGE isn't wasted, because we will score well enough to make up for our time when we find a block, just like when we found or first DOGE block before (on the image above).

Well, I shared more than I should about how and why the pool makes its choices, but I wanted to reassure you that our choices are good and there is a reason for them.

Side note/news: There's an IRC channel on Freenode which you can join if you want to talk about the pool with other users and/or me: #CleverMining

[Damnsammit]

Are you investigating Clevermining and leave Middlecoin? 

Absolutely not leaving Middlecoin.  I like their crappy website and awesome phone app

[roy7]

You might plug your doge mining into p2pool instead of soloing mining, if you solo mine atm. I think multipool does that on all of their coins actually (uses p2pool networks). This way they can get the lower variance p2pool provides for each coin.

[Xeon]

I just started mining on this amazing pool! Really like it!

[QNX]

I just started mining on this amazing pool! Really like it!

also i, but i have 20% rejects with 2mh 

[Terk]

You might plug your doge mining into p2pool instead of soloing mining, if you solo mine atm. I think multipool does that on all of their coins actually (uses p2pool networks). This way they can get the lower variance p2pool provides for each coin.

This might be a good idea. I know absolutely nothing about p2pool, yet alone about how to plug an existing pool into p2pool, while still getting payment into a single address owned by the pool and being able to track shares submitted by individual users at the same time. If anyone has any knowledge, I'd appreciate pointing me into the right direction. If not, I'll research it later this evening to see what's possible and how to do it.

[Terk]

I just started mining on this amazing pool! Really like it!

also i, but i have 20% rejects with 2mh 

Please read this and check if it helps you: https://bitcointalk.org/index.php?topic=448649.msg5063819#msg5063819

[roy7]

This might be a good idea. I know absolutely nothing about p2pool, yet alone about how to plug an existing pool into p2pool, while still getting payment into a single address owned by the pool and being able to track shares submitted by individual users at the same time. If anyone has any knowledge, I'd appreciate pointing me into the right direction. If not, I'll research it later this evening to see what's possible and how to do it.

I'm guessing you'd need to use it as a proxy, where all work is going through to your normal p2pool node (with payment address set to your pools address) and you were recording shares like normal on the front side of your proxy. I have no idea how multipool does it.

If it's of any help code-wise, here is a proxy setup someone wrote to act as a CPPSRB proxy between small miners and p2pool. The pool gets all the fees and then pays out miners based on their scores like any normal pool.

http://proxypool.doge.st/

[Terk]

I'm guessing you'd need to use it as a proxy, where all work is going through to your normal p2pool node (with payment address set to your pools address) and you were recording shares like normal on the front side of your proxy. I have no idea how multipool does it.

If it's of any help code-wise, here is a proxy setup someone wrote to act as a CPPSRB proxy between small miners and p2pool. The pool gets all the fees and then pays out miners based on their scores like any normal pool.

http://proxypool.doge.st/

Thanks a lot! This looks like it would do the job. I would need to check if it's stable and relatively bug-free and to change functions responsible for saving user shares data into database (because this software stores shares in a different way than my software does). This won't be difficult as I already have read related parts of its source code and identified what and where would need to be changed. Seems pretty straightforward even if I don't know Haskell language and this is actually my first time seeing any Haskell source code. But most languages are similar to all other, aren't they?

I'll wait with this until I'm sure that DOGE is here to stay. I don't mean stay as a coin but to stay near top of profitability list. DOGE rewards are halving in 1800 blocks = 30 hours and it should affect its profitability. For DOGE profitability to stay unchanged there would need one of the following to happen: 1) DOGE/BTC price remains the same but difficulty halves because half of miners went away; 2) Difficulty remains the same but price doubles; 3) Some combination of the above (e.g. difficulty drops by 50% and price rises by 50%).

I doubt any of these will happen overnight. They might happen but it will most likely take time - days or weeks - and I believe we will have enough our own hahspower at the time when DOGE profitability backs up to be able to mine them without proxying to p2pool.

In case if one of the above happened overnight, I'll start proxying our DOGE mining to p2pool using the software above. Thanks a lot for pointing me into this. I'll see what will happen when reward halves and will act accordingly then. My best guess is that DOGE will stop being that profitable for the next days/weeks, but of course I might be wrong. I don't think anyone predicted DOGE success and adoption rate so it might surprise again.

In the meantime we continue mining DOGE when it's significantly more profitable than other coins. That means using between 0% and 80% of our hashpower in any hour to mine DOGE. We'll be doing this until hitting next fat DOGE block and then I disable DOGE entirely again. Halving the reward will be coming soon and with our average time expected to mine a block it won't be worth to invest our time any more.

[chilin_dude]

If you don't plan to switch away in a hour or three, then don't worry about us mining DOGE and the variance. There will be several hours worse than average but then there will be a reward like this one:



Of course we are not mining it all the time - only if it's significantly more profitable than other coins. When it's not, we are still mining other coins. This should give better profits in long-term (say 48/72-hour average).

I am aware that it doesn't look encouraging when you look at the 24-hour chart but I decided that your profits are more important than this. Most of users mining here now are not people who jump for an hour and then switch away. Most of you are mining here for days, so I decided it won't affect you in a negative way.

I am still trying to keep a balance between profitability and variance. If DOGE is only 5% more profitable than some other coin, then we will be mining this other coin. This is to ensure the flow of Bitcoins from the pool to the users won't stop. Our variance while mining DOGE is mathematically so big, that the 5% difference doesn't matter, so it's not enough profitable to switch to DOGE. With our average expected time to find a DOGE block, it is highly likely that DOGE price will change by more than this 5% (up or down) before we will find a block. This is why we switch to DOGE only if there is a bigger profitability gap between it and the second best coin. This results in percentage of our hashpower mining DOGE varying between 20% and 80% in different hours and that's fine. We mine DOGE when it's really much more profitable but we still mine other coins if they're more profitable or when the difference isn't that big.

For example, the hour that just ended had 128% LTC profitability, so we're doing our business as usual between times spent on mining DOGE. And time spent on mining DOGE isn't wasted, because we will score well enough to make up for our time when we find a block, just like when we found or first DOGE block before (on the image above).

Well, I shared more than I should about how and why the pool makes its choices, but I wanted to reassure you that our choices are good and there is a reason for them.

Side note/news: There's an IRC channel on Freenode which you can join if you want to talk about the pool with other users and/or me: #CleverMining

^ Think that is a really good post and way to go with the pool sir re balancing profitability and variance. Bravo.

[greatwolf]

Just wondering, has the vardiff been re-enabled yet? Last I heard, that feature was turned off and set to static share diff 512 for diagnostic purposes.

[jwb7111]

Just joined the pool!  All what you are saying is making sense.  A little patience pays off!

[cokeman0]

Hi

What's going on with the btc/mh/day ? Why are we making so little ? are we mining to much doge ?

[thedue]

Hi

What's going on with the btc/mh/day ? Why are we making so little ? are we mining to much doge ?

Dude.. just read what he wrote.. based on 72h we´ll be making more profit.

[sentientmcribs]

hey guys, so i ran into a problem. i was trying to edit my pool info while drunk (the LAST time ill ever do that), and i accidentally put the user name as X instead of my username. for the past 6 hours i have been mining under that. is there anyway that balance can be transferred to my actual bitcoin account? i can verify its me by showing that my hash rate (12.8Mh/s) is the exact same for my actual user name.

[Ritual]

hey guys, so i ran into a problem. i was trying to edit my pool info while drunk (the LAST time ill ever do that), and i accidentally put the user name as X instead of my username. for the past 6 hours i have been mining under that. is there anyway that balance can be transferred to my actual bitcoin account? i can verify its me by showing that my hash rate (12.8Mh/s) is the exact same for my actual user name.

I really hope you get sorted mate, and I'm sure Terk will sort you out, but I couldn't help but laugh It's the type of thing I would do!

So to drunk-texting and drunk-facebooking, we must now add drunk-mining