[CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers - page 86.

SelbyTsang

hero member

Activity: 577

Merit: 504

Quote from: -ck on May 03, 2014, 07:59:58 AM

Quote from: ManeBjorn on May 02, 2014, 09:29:20 AM

I have been researching pool centralization and manipulation, one of the things I found is that discus fish and ghash set their accepted block size very low. Discus fish obscenely low.
This allows them to plow through more of the smaller blocks faster than pools that play fair.

There's no such thing. "Smaller blocks" (presumably by that you mean ones with less transactions) take just as long on average to solve as larger ones.

The time needed to find the block should be the same, in statistical sense.
But the time needed to propagate a smaller block to the network should be slightly shorter, giving the miner a slightly advantage in orphan block race.

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Quote from: ManeBjorn on May 02, 2014, 09:29:20 AM

I have been researching pool centralization and manipulation, one of the things I found is that discus fish and ghash set their accepted block size very low. Discus fish obscenely low.
This allows them to plow through more of the smaller blocks faster than pools that play fair.

There's no such thing. "Smaller blocks" (presumably by that you mean ones with less transactions) take just as long on average to solve as larger ones.

DPoS

sr. member

Activity: 462

Merit: 250

Quote from: organofcorti on May 02, 2014, 04:38:04 AM

eleuthria, people like this guy clearly don't understand probability, and given that the community can't stop people like that mining, these sorts of posts might end if you add a confidence interval to the luck chart for context indicating how likely an average "luck" is.

I've worked with systems my whole life and many times had to point out issues in other departments when I detected a change in behavior. And almost always I get blasted for 'not knowing what I am talking about' so I have think skin on this

Or you can say I really do understand probability since I predicted a long downtrend... we know that isn't true so more likely I sensed a change in the pulse and left the table

ManeBjorn

legendary

Activity: 1288

Merit: 1004

I have been researching pool centralization and manipulation, one of the things I found is that discus fish and ghash set their accepted block size very low. Discus fish obscenely low.
This allows them to plow through more of the smaller blocks faster than pools that play fair.
I have more but I have to finish verification first before I say anything more.

Quote from: DPoS on May 01, 2014, 10:57:34 PM

Quote from: BrandonMcPherson on May 01, 2014, 09:46:23 PM

DPoS, I appreciate that you are genuinely concerned about this, but variance happens.

If you have any evidence beyond the de facto luck of the pool, I'd personally be delighted to see it, but claiming to be the only guy who can see the conspiracy while the math nerds politely nod and smile...

I say this without malice, but it reminds me of the scores of (to a person, bad) poker players I've met over the years who talk about how dealers screw them, how winners are just lucky, how they usually win when they sit in their lucky seat, how they never win with pocket aces, and a gillion other things that all come down to the reality that variance happens.

They don't understand it, and so the game must be rigged.

nope - I've mined with BTCGuild 2nd half of last year and right into this situation. I totally know the types you are talking about.. I am talking about a step down in the overall dynamic.

I said a few times I don't run pools and can't be an expert. But I can notice when a game balance has changed from a mile away so I put that out there right when it started to happen.. It continues and will continue until taken seriously

I'll put it another way.. do you think those folks that raped wall street with High Frequency Trading haven't cast their eyes on mining pools to look for any way shape or form to get and edge?

Call me chicken little and I am not attacking BTCGuild, it is just this is the place I mined for so long so this where my gut threw the red flag that something was amiss..

perhaps Ghash or Discus Fish or KNC figured something out

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote from: Sir Alan on May 02, 2014, 06:27:18 AM

Quote from: organofcorti on May 02, 2014, 04:38:04 AM

people like this guy clearly don't understand probability

but they do love innuendo.

It'll be in your endo in a minute! *heh heh*

Sir Alan

full member

Activity: 221

Merit: 100

Quote from: organofcorti on May 02, 2014, 04:38:04 AM

people like this guy clearly don't understand probability

but they do love innuendo.

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote from: DPoS on May 02, 2014, 01:28:14 AM

No problem - I'm not here to fight or anything just mentioned it 2 months ago when I notice the change and here we are. See ya in another two months

regards

eleuthria, people like this guy clearly don't understand probability, and given that the community can't stop people like that mining, these sorts of posts might end if you add a confidence interval to the luck chart for context indicating how likely an average "luck" is.

DPoS

sr. member

Activity: 462

Merit: 250

No problem - I'm not here to fight or anything just mentioned it 2 months ago when I notice the change and here we are. See ya in another two months

regards

eleuthria

legendary

Activity: 1750

Merit: 1007

You've already admitted you're not an expert. That is also obvious by what you've said. I have kept what you've posted in this thread even though it is making assumptions that are completely false.

Ghash, Discus Fish, and KNC do not post their luck figures. They also don't publish any historical data to calculate it externally. For all we know their luck is as bad or worse. Even if it isn't bad though, that doesn't mean anything because variance doesn't only swing one way. Just like how BTC Guild had over 6 consecutive months of positive luck. But nobody ever bitches when things are working better than expected, but they sure do forget it quickly.

However, they did not "figure something out" and the fact that you even mention that shows you have no clue what you're talking about. If they "figured something out" it would mean SHA256 is fundamentally broken and Bitcoin is insecure. Mining is a RANDOM process. That is a *FACT* unless SHA256 is broken. There is no way to "more efficiently distribute work". Finding a difficulty 1 share is a 1 in ~4.2 billion chance per hash. Finding a share at current network difficulty is a 1 in ~33,600,000,000,000,000,000 chance per hash.

DPoS

sr. member

Activity: 462

Merit: 250

Quote from: BrandonMcPherson on May 01, 2014, 09:46:23 PM

DPoS, I appreciate that you are genuinely concerned about this, but variance happens.

If you have any evidence beyond the de facto luck of the pool, I'd personally be delighted to see it, but claiming to be the only guy who can see the conspiracy while the math nerds politely nod and smile...

I say this without malice, but it reminds me of the scores of (to a person, bad) poker players I've met over the years who talk about how dealers screw them, how winners are just lucky, how they usually win when they sit in their lucky seat, how they never win with pocket aces, and a gillion other things that all come down to the reality that variance happens.

They don't understand it, and so the game must be rigged.

nope - I've mined with BTCGuild 2nd half of last year and right into this situation. I totally know the types you are talking about.. I am talking about a step down in the overall dynamic.

I said a few times I don't run pools and can't be an expert. But I can notice when a game balance has changed from a mile away so I put that out there right when it started to happen.. It continues and will continue until taken seriously

I'll put it another way.. do you think those folks that raped wall street with High Frequency Trading haven't cast their eyes on mining pools to look for any way shape or form to get and edge?

Call me chicken little and I am not attacking BTCGuild, it is just this is the place I mined for so long so this where my gut threw the red flag that something was amiss..

perhaps Ghash or Discus Fish or KNC figured something out

MoreBloodWine

legendary

Activity: 1064

Merit: 1001

Quote from: DPoS on May 01, 2014, 09:04:44 PM

Quote from: eleuthria on May 01, 2014, 08:56:44 PM

I really just need to remove that chart. Eligius' thread doesn't get derailed this frequently over this crap and their 90 day luck is almost identical to ours. (Guild: 92.3%, Eligius 92.5% for the last 2000 blocks [calculated yesterday], which is ~88 days, p2pool 90.8%).

I'd say Eliguis is suffering from the same BS - you seem to want to run away from the stuff I mentioned

I'll go away and be back in two months since you want to call it 'crap'

Don't take it the wrong way, it's just eleuthria fields luck stuff more than a goalie at hockey game.

Ok, that analogy didn't word the way I hoped it would but the points still made. He gets it a lot.

BrandonMcPherson

full member

Activity: 205

Merit: 100

DPoS, I appreciate that you are genuinely concerned about this, but variance happens.

If you have any evidence beyond the de facto luck of the pool, I'd personally be delighted to see it, but claiming to be the only guy who can see the conspiracy while the math nerds politely nod and smile...

I say this without malice, but it reminds me of the scores of (to a person, bad) poker players I've met over the years who talk about how dealers screw them, how winners are just lucky, how they usually win when they sit in their lucky seat, how they never win with pocket aces, and a gillion other things that all come down to the reality that variance happens.

They don't understand it, and so the game must be rigged.

DPoS

sr. member

Activity: 462

Merit: 250

Quote from: eleuthria on May 01, 2014, 08:56:44 PM

I really just need to remove that chart. Eligius' thread doesn't get derailed this frequently over this crap and their 90 day luck is almost identical to ours. (Guild: 92.3%, Eligius 92.5% for the last 2000 blocks [calculated yesterday], which is ~88 days, p2pool 90.8%).

I'd say Eliguis is suffering from the same BS - you seem to want to run away from the stuff I mentioned

I'll go away and be back in two months since you want to call it 'crap'

demonmaestro

hero member

Activity: 574

Merit: 500

Mining for the hell of it.

well its not really the chart but 3 blocks.... Undecided

eleuthria

legendary

Activity: 1750

Merit: 1007

I really just need to remove that chart. Eligius' thread doesn't get derailed this frequently over this crap and their 90 day luck is almost identical to ours. (Guild: 92.3%, Eligius 92.5% for the last 2000 blocks [calculated yesterday], which is ~88 days, p2pool 90.8%).

guytechie

hero member

Activity: 677

Merit: 500

Quote from: DPoS on May 01, 2014, 12:02:06 PM

Quote from: navigator on May 01, 2014, 08:45:55 AM

In the past btcguild seems like it was always slightly above 100% luck. It's nothing unusual for it to have a bad luck streak like this.

Say you played blackjack for 30 years and then the dealer took out the Kings.. you would notice a change in the dynamic quickly before even realizing the kings were out. That was the same type of feeling I had when this downtrend started and why I made a fuss and then left.

Perhaps ghash is doing that 'don't report found blocks right away' ploy. Or something else like hashing on BTCGuild and disrupting solved blocks somehow.. or maybe they just figured out a better way to distribute work in their pool.

I don't know but I do know that if you keep playing blackjack without kings, eventually you'll just forget how it was when there were kings in blackjack and then make a new sense of what's normal

also.. if that blackjack dealer wanted to go for the long con, they would put kings back in the deck once every 4 shuffles or just take out two kings and two queens so things can still appear normal a little bit more...

just saying

Also a weird coincidence luck seem to have gone down around the time PPS was discontinued. Don't know how to say that without being taken the wrong way...

DPoS

sr. member

Activity: 462

Merit: 250

Quote from: navigator on May 01, 2014, 08:45:55 AM

In the past btcguild seems like it was always slightly above 100% luck. It's nothing unusual for it to have a bad luck streak like this.

Say you played blackjack for 30 years and then the dealer took out the Kings.. you would notice a change in the dynamic quickly before even realizing the kings were out. That was the same type of feeling I had when this downtrend started and why I made a fuss and then left.

Perhaps ghash is doing that 'don't report found blocks right away' ploy. Or something else like hashing on BTCGuild and disrupting solved blocks somehow.. or maybe they just figured out a better way to distribute work in their pool.

I don't know but I do know that if you keep playing blackjack without kings, eventually you'll just forget how it was when there were kings in blackjack and then make a new sense of what's normal

also.. if that blackjack dealer wanted to go for the long con, they would put kings back in the deck once every 4 shuffles or just take out two kings and two queens so things can still appear normal a little bit more...

just saying

navigator

sr. member

Activity: 362

Merit: 250

In the past btcguild seems like it was always slightly above 100% luck. It's nothing unusual for it to have a bad luck streak like this.
I also feel like the pool with the most hash will be the one that's slightly more lucky than it should be, but I don't see how that can be true.
Ghash is already at 18PH today and will likely be over 19PH tomorrow or the next day. It's growing at an astounding rate.
I think that since sometimes Ghash will find a block in a few seconds, that those blocks are what are 'taking away' the luck from the other pools.
But in reality it's just lady luck and variance.

Has anyone calculated Ghash's luck?

Edit: I forgot about http://organofcorti.blogspot.com/ lots of useful pool info there.

DPoS

sr. member

Activity: 462

Merit: 250

Quote from: eleuthria on April 29, 2014, 05:24:14 PM

Quote from: DPoS on April 29, 2014, 04:30:09 PM

Quote from: eleuthria on April 28, 2014, 09:35:12 PM

Sometimes I feel like Bitcoin's hashing/difficulty algorithm is actually a cruel joke. 256 minute block -> 325 minute block -> 9 minute -> 14 minute -> 13 minute -> 3 minute.

2 blocks took 9 hours followed by the pool getting 5 of the last 6 on the entire network.

maybe the cruel joke is *whatever* is affecting your pool and that good stretch was a hiccup/reset/reconfig of *whatever* is affecting your pool

maybe I am wearing a tinfoil hat but maybe you have blinders on..I don't know since I never ran a pool but it is obvious *something* has pulled a few cards out of your deck

Oh, it's "obvious" something is happening? The post was tongue in cheek. EVERY POOL has stretches and then bursts like I posted above, and it's annoying for all of them.

If you want an idea of how far variance can swing: BTC Guild found 11 blocks in a row across the network a few months ago. That was with the pool only have ~25% of the entire network hash rate.

EDIT: The blocks: 275644 - 275654. 11 blocks by BTC Guild in 1h 10m with 0 blocks from other pools in that time.

As I said I don't run a pool but as an outsider just observing patterns it does seem like the game is being rigged.. perhaps those great streaks you still get are the only way you can fight through *whatever* is going on... and many of the shorter streaks are getting hijacked or whatever

I haven't analyzed your graphs completely but it seems that you never have good stretches anymore where you only dip to 100% and keep spiking. Your pool used to do that frequently before and the last time was first week in March

That's close to two months ago. The recent good stretch still dipped many times under 100% a week ago. Just seems like lower lows and lower highs.

I used to mainly mine PPS on your pool but always kept an eye on the performance of the PPLNS since I would try to take a few runs in there from time to time.... and I can say that a few months ago *something* changed

Sadly there are only a few big pools so variance is still going to be used to explain it away... I'll check back in a few months but if nothing is discovered I expect the pool to stay around 80% luck

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: MoreBloodWine on May 01, 2014, 03:09:54 AM

Quote from: eleuthria on May 01, 2014, 03:07:13 AM

Quote from: MoreBloodWine on May 01, 2014, 03:02:06 AM

Quote from: eleuthria on May 01, 2014, 02:57:26 AM

Quote from: MoreBloodWine on May 01, 2014, 02:56:28 AM

Waiting for a confirmation 2014-05-01 03:54 AM

Where exactly does this "confirmation" come from ?

When another block on the network confirms it. That only shows up when BTC Guild just found a block and no new block on the network has built off it (IE: We don't know if it's going to be orphaned or not yet).

Ahh, so when the next block is found / built off of it its a yay or nay to boot it off the island.

Yep. The pool no longer pays for orphans, but it doesn't require a 100-120 confirmation wait. The likelihood of a block getting orphaned after it is confirmed is extremely low, to the point it just doesn't seem worth delaying the application of rewards to wait for even more confirmations.

I was actually sad to see this go.

In the past, oh lets say 6 months, how many orphans has BTCGuild seen ?

It's been less than 1% since the paid orphans were removed. We haven't had one in the last 300 blocks now. The historical orphan rate (not counting the 0.8 hardfork event) has been about as close to 1% you can get given the sample size in comparison to the expected frequency.

Topic: [CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers - page 86. (Read 903163 times)