Topic: [500 GH/s]HHTT -Selected Diff/Stratum/PPLNS/Paid Stales/High Availability/Tor - page 25. (Read 56478 times)
October 11, 2012, 06:52:24 PM
Your posting style reminds me of the OP of the Mt.Red-Thread
October 11, 2012, 05:50:08 PM
Rather than doing anything useful I've implemented dynamically stupid colors.
The code is horrible but it actually does a fairly good job.
The code is horrible but it actually does a fairly good job.
October 11, 2012, 09:44:23 AM
Hey FD - how about a bonus
i have found 5 blocks for the pool (4 in the last 5 days) and have been paid 89 BTC
I guess this means I just have to stay on the pool for a long time as this "luck" will revert and I will probably not find a block for 3 weeks but keep earning
EDIT - seems we found another block today - now almost 62 BTC positive!
i have found 5 blocks for the pool (4 in the last 5 days) and have been paid 89 BTC
I guess this means I just have to stay on the pool for a long time as this "luck" will revert and I will probably not find a block for 3 weeks but keep earning
EDIT - seems we found another block today - now almost 62 BTC positive!
October 11, 2012, 07:46:55 AM
Interesting. I haven't really used Chebyshev's inequality before.
What I did was simple though - I'm just used Poisson quantiles for various lambda at p < 0.025 and p < 0.975 (lower tail) to generate a 95% confidence interval. The rest is just making it into a format that readers can follow.
Now, on to Chebyshev ....
What I did was simple though - I'm just used Poisson quantiles for various lambda at p < 0.025 and p < 0.975 (lower tail) to generate a 95% confidence interval. The rest is just making it into a format that readers can follow.
Now, on to Chebyshev ....
October 11, 2012, 07:39:27 AM
Organofcorti work is awesome, so are his tables.
But, if nothing more than a raw indication is needed, one could use Chebyshev alone: http://en.wikipedia.org/wiki/Chebyshev%27s_inequality
With as little as about 1.35GH/s mining power, I mine with a share-difficulty of 169.
So this is the math.
In one day I find about 1.35GH/s * 24h * 60m * 60s = 116,640,000,000,000 hashes.
In one day, in average, only 116,640,000,000,000 / ( 169 * 2^32 ) = 161 will be accepted. About one share every 9 minutes.
So my Poisson distribution (for a day) will have a mean λ = 161 and also a variance σ² = 161 (of course, in Poisson distribution mean and variance are equal: http://en.wikipedia.org/wiki/Poisson_distribution).
Chebyshev says that:
Pr{ |X-λ| < kσ } >= 1 - 1/k²
Let's choose k=2, so that we can have 75%:
Pr{ |X-161| < 25 } >= 75%
that is:
Pr{ 136 < X < 186 } >= 75%
that is:
Pr{ 0.85λ < X < 1.15λ } >= 75%
So there's and extremely good chance (at least 75%) that my daily earnings will be not less than 85% of my average
- and not more than 115% of my average.
For me it's enough: I didn't know about organofcorti tables, I didn't want to go back to my old statistics books, and so I chose an easy way to know that yes, I can afford 169 difficulty.
But, if nothing more than a raw indication is needed, one could use Chebyshev alone: http://en.wikipedia.org/wiki/Chebyshev%27s_inequality
With as little as about 1.35GH/s mining power, I mine with a share-difficulty of 169.
So this is the math.
In one day I find about 1.35GH/s * 24h * 60m * 60s = 116,640,000,000,000 hashes.
In one day, in average, only 116,640,000,000,000 / ( 169 * 2^32 ) = 161 will be accepted. About one share every 9 minutes.
So my Poisson distribution (for a day) will have a mean λ = 161 and also a variance σ² = 161 (of course, in Poisson distribution mean and variance are equal: http://en.wikipedia.org/wiki/Poisson_distribution).
Chebyshev says that:
Pr{ |X-λ| < kσ } >= 1 - 1/k²
Let's choose k=2, so that we can have 75%:
Pr{ |X-161| < 25 } >= 75%
that is:
Pr{ 136 < X < 186 } >= 75%
that is:
Pr{ 0.85λ < X < 1.15λ } >= 75%
So there's and extremely good chance (at least 75%) that my daily earnings will be not less than 85% of my average
- and not more than 115% of my average. For me it's enough: I didn't know about organofcorti tables, I didn't want to go back to my old statistics books, and so I chose an easy way to know that yes, I can afford 169 difficulty.
October 10, 2012, 10:27:58 AM
Looks like I need to put my big brain on before reading that post above...
Or just follow the worked example in part three of the blog post. You don't have to follow the theory to use the tables and work out how much variation in average per minute/hourly/daily hashrate you'll experience.
October 10, 2012, 10:24:40 AM
Looks like I need to put my big brain on before reading that post above...
October 10, 2012, 08:52:17 AM
Anyone not sure about which difficulty to select can find out how to choose a difficulty that will give you an acceptable variation in average pool hashrate here:
http://organofcorti.blogspot.com.au/2012/10/71-variable-pool-difficulty.html
http://organofcorti.blogspot.com.au/2012/10/71-variable-pool-difficulty.html
October 10, 2012, 07:29:43 AM
OK - I am really hating myself now
3 blocks found in 3 days (4 total) and I have been paid out less than 80 BTC
Glad the pool is in better financial shape but wishing I was going solo right about now
3 blocks found in 3 days (4 total) and I have been paid out less than 80 BTC
Glad the pool is in better financial shape but wishing I was going solo right about now
October 08, 2012, 10:01:19 PM
Yep, have been hitting the open file limit. My guess is that either someone is messing with me or a big new user is trying the pool with their several hundred connections.
I've increased max conns and added some tracking to see who is doing this. Hopefully it is legit (or at least a well operated bot net).
I've increased max conns and added some tracking to see who is doing this. Hopefully it is legit (or at least a well operated bot net).
October 08, 2012, 09:47:45 PM
looks like something is messed again. was just checking to see if i should swap back, shows 10 ghash over the last 5m
October 08, 2012, 01:00:30 PM
Yeah, someone was going awry. If it happens again I'll start getting drastic on it.
October 08, 2012, 07:39:23 AM
Back it seems...
And then as soon as i typed this it went down again. Seems back now (13:22 UTC).
And then as soon as i typed this it went down again. Seems back now (13:22 UTC).
October 08, 2012, 07:36:49 AM
Pool offline now.
October 08, 2012, 07:10:57 AM
And I found another block
Now I am feeling unlucky - 150 BTC found for the pool and I have less than 65 BTC to show for it
Now I am feeling unlucky - 150 BTC found for the pool and I have less than 65 BTC to show for it
October 07, 2012, 03:43:26 PM
We just got a block and I threw in the rest of my personal savings so it should be good for a while.
I'll release the code in a little bit. It isn't anything magical.
I'll release the code in a little bit. It isn't anything magical.
October 07, 2012, 01:55:06 PM
Stopping payouts is a double-edged sword but I understand your reasoning...
My concern, is users will abandon the pool and then the time to find a block will increase substantially with only 2 singles mining. At current diff that is just shy of 90 days (assuming 1.6 Gh/s) and we know diff will increase at a crazy pace once ASIC arrives = never finding a block?!?
My concern, is users will abandon the pool and then the time to find a block will increase substantially with only 2 singles mining. At current diff that is just shy of 90 days (assuming 1.6 Gh/s) and we know diff will increase at a crazy pace once ASIC arrives = never finding a block?!?
It would make more sense to allow others to invest in the pool. Create 1000 bonds with a starting face value of 1BTC, valid for one month.
If the pool loses money, those bonds lose value, if the pool makes money, those bonds gain value. With 2% PPS the bonds are probably worth the risk for some investors, and bondholders would have incentive to recommend your pool to others.
Or sell me the code and I'll set it up
October 07, 2012, 08:34:06 AM
Stopping payouts is a double-edged sword but I understand your reasoning...
My concern, is users will abandon the pool and then the time to find a block will increase substantially with only 2 singles mining. At current diff that is just shy of 90 days (assuming 1.6 Gh/s) and we know diff will increase at a crazy pace once ASIC arrives = never finding a block?!?
So I continue to hope luck changes as the pool has been rock solid - so congrats on your efforts!
EDIT - I just found a block
My concern, is users will abandon the pool and then the time to find a block will increase substantially with only 2 singles mining. At current diff that is just shy of 90 days (assuming 1.6 Gh/s) and we know diff will increase at a crazy pace once ASIC arrives = never finding a block?!?
So I continue to hope luck changes as the pool has been rock solid - so congrats on your efforts!
EDIT - I just found a block
October 07, 2012, 07:21:15 AM
Are you really 60 BTC in the hole? This has me worried. How can you keep going with those kind of losses?
Code:
Income Paid Net
400.00000000 460.36536787 -60.36536787
I am absolutely that far in the hole.
I have some sort of organic poptart that tastes like sadness. I nibble it as I ponder what I have done to get myself where I am. I have regrets but not that many.
Can I make a suggestion?
How about varying the pool fee according to the net financial position?
I made a little pool sim in matlab, for a bit of visual representation of just how savage operator variance can be on a 2% PPS pool, here's 100 runs over 1000 blocks:
As you can see, the risk to you is fairly substantial - however I was planning on pointing ASICs at your pool when they release, and I do think you could make a great success of this thing if you had some investors on board to give you a bit of a war-chest.
October 07, 2012, 12:49:27 AM
@FD,
Sorry to see the pool stats as they are - cross my fingers that a good luck streak is hit.
Is there a donation address for the pool?
Sorry to see the pool stats as they are - cross my fingers that a good luck streak is hit.
Is there a donation address for the pool?
I certainly wouldn't seek donations, I am intending this as a for-profit operation. However, I wouldn't turn down donations. If you are so inclined, you can send too the main pool pot: 1HHTTPaz86gcxqntnjZMHbaoF4EvyYZCka
If the luck doesn't improve, here is what will happen:
Whenever there are funds, the pool will send what it can proportionally to what is owed. This will keep on until everyone is paid, including whoever (reasonably) elects to stop mining on the pool. Regardless of how long that takes, at the very least my two singles will be on the pool and my two SC singles when they come so some blocks will come in and payments will go out.
I certainly understand anyone who decides to switch pools if that happens. I know most people are not looking for variance. Really I started this pool to raise awareness of lowering mining overhead by increasing difficulty and that goal is accomplished.
Jump to: