[POOL][Scrypt][Scrypt-N][X11] Profit switching pool - wafflepool.com

[phzi]

So that would mean you were one of the miners who were taking up to 10% more of your fair share of profits from the pool, however unknowingly.

Oh, I knew what I was doing by cranking my miners the way I did - but as I said, I have discovered many blocks for WafflePool, and a rare few were orphans.

For the others who had their equipment tuned correctly for mining almost anywhere else than wafflepool, they were receiving up to 10% less of profits directly due to their efforts, depending upon the ratio of efficient to inefficient miners in the pool.  The actual percentages depend upon whether wafflepool was actually a pool of fools, and only poolwaffle knows that as he is the only one with access to the relevant statistics.

Ya... I still think that was one of the best parts about wafflepool - haha.

I too would be very curious about sfire's 20GH/s mining efficiency prior to the server changeover, which I consider to be a very welcome one.  Ironically, depending upon that statistic, it might not have been the largest miner that was keeping us from efficiently mining the lowest difficult coins, but the smaller miners instead!

Indeed.

--

I am getting around 2.5% average reject rate now after re-tuning for the new reject profile.

Looks like the new stratum server is costing me overall about 2% in decreased hashrate, and 2% additional rejects (very short sample period so far, mind you)

Guess this was the point of these changes.  Not overly pleased myself.

[Dishes]

So that would mean you were one of the miners who were taking up to 10% more of your fair share of profits from the pool, however unknowingly.

Oh, I knew what I was doing by cranking my miners the way I did - but as I said, I have discovered many blocks for WafflePool, and a rare few were orphans.

For the others who had their equipment tuned correctly for mining almost anywhere else than wafflepool, they were receiving up to 10% less of profits directly due to their efforts, depending upon the ratio of efficient to inefficient miners in the pool.  The actual percentages depend upon whether wafflepool was actually a pool of fools, and only poolwaffle knows that as he is the only one with access to the relevant statistics.

Ya... I still think that was one of the best parts about wafflepool - haha.

I too would be very curious about sfire's 20GH/s mining efficiency prior to the server changeover, which I consider to be a very welcome one.  Ironically, depending upon that statistic, it might not have been the largest miner that was keeping us from efficiently mining the lowest difficult coins, but the smaller miners instead!

Indeed.

--

I am getting around 2.5% average reject rate now after re-tuning for the new reject profile.

Looks like the new stratum server is costing me overall about 2% in decreased hashrate, and 2% additional rejects (very short sample period so far, mind you)

Guess this was the point of these changes.  Not overly pleased myself.

So you're seriously bummed out getting 2.5% rejects? Even though the server changes help your profitability overall?

[xia0mingx]

I am still getting 0% rejects and at the same hash rate. Not sure whether i need to reconfigure any settings or not, but everything looks the same to me.
WU is about 91% of hash rate
Im using Singapore server.

[dspair]

http://wafflepool.com/stats
Today's profitability is through the roof.

BTC / 1MH
0.00839364

[comeonalready]

So that would mean you were one of the miners who were taking up to 10% more of your fair share of profits from the pool, however unknowingly.

Oh, I knew what I was doing by cranking my miners the way I did - but as I said, I have discovered many blocks for WafflePool, and a rare few were orphans.

For the others who had their equipment tuned correctly for mining almost anywhere else than wafflepool, they were receiving up to 10% less of profits directly due to their efforts, depending upon the ratio of efficient to inefficient miners in the pool.  The actual percentages depend upon whether wafflepool was actually a pool of fools, and only poolwaffle knows that as he is the only one with access to the relevant statistics.

Ya... I still think that was one of the best parts about wafflepool - haha.

I too would be very curious about sfire's 20GH/s mining efficiency prior to the server changeover, which I consider to be a very welcome one.  Ironically, depending upon that statistic, it might not have been the largest miner that was keeping us from efficiently mining the lowest difficult coins, but the smaller miners instead!

Indeed.

--

I am getting around 2.5% average reject rate now after re-tuning for the new reject profile.

Looks like the new stratum server is costing me overall about 2% in decreased hashrate, and 2% additional rejects (very short sample period so far, mind you)

Guess this was the point of these changes.  Not overly pleased myself.

So your seriously bummed out getting 2.5% rejects? Even though the server changes help your profitability overall?

Dishes, you're missing the point entirely.  Phzi has admitted to having his equipment [over]tuned to submit the maximum number of shares possible, regardless of whether they were valid (not stale).  He was diluting the value of all valid shares which is the basis upon which we are all paid out, as only valid shares can be used to solve a block, thereby taking a greater portion of the total pool earnings than was actually generated by him, as he was being paid on the basis of his total shares, however worthless to the pool.  So for all those miners who had tuned their equipment to produce the greatest number of valid shares, phzi was knowingly taking profits out of your pockets.  (In fact, had someone created a modified miner to take advantage of this pool vulnerability, they could possibly have sapped many more profits away from other pool members, depending upon the server code.  And there are many smart people mining these days, so I would not be surprised if there were a good chance this was actually happening.  poolwaffle, did anyone's reject rates shoot up significantly higher than 10% immediately following the changeover?)

So this stale share handling change on the pool servers does not increase phzi's profitability at all, it increases the profitability of the more productive members of the pool.  He will see a decrease in profitability, as he can no longer leech profits from other members of the pool.

Phzi, I will not speculate upon whether your actions were ethical, as my own personal opinion of most human beings is rather low (poolwaffle excepted by my impressions of him thus far).  I think that most of us will do such things if there are no real consequences when found out.  But openly admitting it and being proud about having taken advantage of others is a rather strange twist.

Poolwaffle, as many as complaints that might come in for making this change, it was the proper thing to do so as to encourage those underperforming members of the pool to learn to tune their equipment, and properly place the fiscal burden upon them for failing to.  The old system even hurt you too, as the pool has been showing lower returns per hashpower than it could have been, and ranked below other multipools as a result.

[comeonalready]

So that would mean you were one of the miners who were taking up to 10% more of your fair share of profits from the pool, however unknowingly.

Oh, I knew what I was doing by cranking my miners the way I did - but as I said, I have discovered many blocks for WafflePool, and a rare few were orphans.

I believe you are mistaken about how you are justifying your actions by claiming that you created few orphans.  Depending upon the server code, all of the very many stale shares produced by your miners and submitted to the pool server only to increase your total share count, would never even have the opportunity to be passed on to the coin daemon, but just accounted for in the database.  Orphan blocks are created by entirely different circumstances.  Can you comment, poolwaffle?

[flukbok]

http://wafflepool.com/stats
Today's profitability is through the roof.

BTC / 1MH
0.00839364

https://i.imgur.com/wo7eMzc.png
Note : I graph directly against coins earned instead of mhash or reject rates.

Yeah it seems so, i'm seeing increased rejects, but profits seems to be going up.
I know its too early to tell but let us compare tomorrow against other pool.

Guys, i'm on 0.5% rejects on SEA server. What's yours ?

[DrunkenDonkey]

0.1% rejects overall, same config as before, that only shows it was done correctly

[DrunkenDonkey]

Sorry to disappoint but:
- I'm on EU, the first server that was updated
- Having lot of stale shares and submitting them, always accepted (my expire is set way too low, have to set it again to high)
- Producing enough - 670khz per 7950, 91% work unit efficiency

It is just good config. Yes, I can run this card and achieve the same speed on t1, I20, tc 24000, but then I get many rejects, instead I'm using t2, xi4, tc 8192 and works like charm. It is all about optimizing

[Mk2vr6]

Hey guys, after a few weeks of private alpha, followed by a few weeks of beta, I'm thrilled to announce http://wafflepool.com

Wafflepool automatically mines the most profitable alt-coin, converts the mined coins to BTC, and pays out daily.

To keep things as simple as possible, there is no registration, just point your miner to the pool with your BTC address as the username, and we do the rest!

Basic info:
Point your miner to WafflePool with your bitcoin address. Mine altcoins, get paid bitcoins
All block transaction fees are paid to miners!
Stratum for extremely low stale rate!
1% mining fee - This includes all transaction fees, and fees for converting to BTC
Automatic difficulty adjustment
No downtime when switching between profitable coins!
No registration, just point and mine!
PPLNS rewards for mining fairness

Payouts are run once daily, and automatically pay anyone with a balance of 0.01 btc.  On Sundays pay anyone with a balance of 0.001 btc.
Payouts are not at an exact time, but are typically at Noon GMT (7am EST).

Pm'd.

[comeonalready]

Sorry to disappoint but:
- I'm on EU, the first server that was updated
- Having lot of stale shares and submitting them, always accepted (my expire is set way too low, have to set it again to high)
- Producing enough - 670khz per 7950, 91% work unit efficiency

It is just good config. Yes, I can run this card and achieve the same speed on t1, I20, tc 24000, but then I get many rejects, instead I'm using t2, xi4, tc 8192 and works like charm. It is all about optimizing

Not sure why you thought I would be disappointed about anything concerning your configuration, as all I was pointing out was that your observation did not necessarily prove all that you stated it did.  What is your MH/s per KWh?

[DrunkenDonkey]

Not sure, don't have a means to measure it. Should be pretty low tho, the highest voltage running card is 1.093, the rest are spread between 1.050 and 1.070 and quite cool even in the hot weather around, don't reach 70. My point is something waffle said: cranking intensity up gets you stales, because the card has no time to submit stuff, on the other hand, having the same speed achieved with lower intensity makes the shares gets submitted very fast, thus the low rejects. Also observation between the 2 configs - one makes even the mouse lag, the other - you can freely watch movies and can forget there is even mining going on.

Edit: just checked the other 3 rigs with 280x on them - the highest reject is 0.27%, the lowest - 0.07%

[miless2111s]

Not sure, don't have a means to measure it. Should be pretty low tho, the highest voltage running card is 1.093, the rest are spread between 1.050 and 1.070 and quite cool even in the hot weather around, don't reach 70. My point is something waffle said: cranking intensity up gets you stales, because the card has no time to submit stuff, on the other hand, having the same speed achieved with lower intensity makes the shares gets submitted very fast, thus the low rejects. Also observation between the 2 configs - one makes even the mouse lag, the other - you can freely watch movies and can forget there is even mining going on.

Edit: just checked the other 3 rigs with 280x on them - the highest reject is 0.27%, the lowest - 0.07%

DM
Thanks for sharing your settings and experience - I've got a couple of questions (please forgive me if they're silly due to my newness to all this!):
1) I am using xubuntu and cgminer 3.7.2 - can I use this xi with this?
2) Given that the "power" of the graphics card is used for mining does the fact that you can play movies etc mean that you're missing out on some of the spare "power" for mining and hence there is more hash/effective work that your setup could be doing?  I'm using "power" as I don't know what to talk about in terms of watts, hash to mean "processing power!

[comeonalready]

poolwaffle, it might have been mentioned before, but is there any way we can get average rejection rates listed next to average hash rates, for both the total pool broken out by day, and individual miners broken out per worker?  Such information would help drive us all to increase our efficiency collectively.

[poolwaffle]

poolwaffle, it might have been mentioned before, but is there any way we can get average rejection rates listed next to average hash rates, for both the total pool broken out by day, and individual miners broken out per worker?  Such information would help drive us all to increase our efficiency collectively.

Yep, on my list for today.  It wasn't an issue before so we hadn't been showing it   Until I can get it added, overall pool reject rate is currently going between 2-3% overall.

0.1% rejects overall, same config as before, that only shows it was done correctly

Actually, that does not really prove anything, as the server you're on could very possibly still be running the old code.  (Though if pw says it's updated, then I would take him at his word.)  Also, it could also show that your equipment is not producing as many valid shares as it can.  Or it could show that your miner is not set to submit stale shares, and the only ones that are being rejected by the server are those it submits because it still thinks they have a valid status.

In any case, it shows that miners like phzi were taking a share of your profits by exploiting the system.

Everyone is on the new servers (about 12 hours now), minus about 3 miners (about 15MH total) that are using a custom port for people that couldn't mine on 3333 (which I honestly forgot about until now)...  I'll switch that over today.

As for phzi, certainly didn't mean to call anyone out on the miner tweaking, just noticed it was happening, and wanted to fully explain what was going to happen with the new software (and why).  I honestly expected a bit higher than 2-3% overall (expected 5-7% and thus more complaining on the forum), but 2-3% seems to be taken really well.  From his perspective, everything he did made sense.  Kept tuning the card further and further without getting any rejects (which is pretty much exactly what you're supposed to do), and everything was working fine from his side (shares accepted, blocks found, etc).  Its just a matter that due to the way the stratum endpoint was handled, it really should have been telling him (rejects) that what he was doing wasn't as efficient as he saw it as (for the pool as a whole that is).  My guess is there are a decent number of people in the same spot, he's just going to take the brunt of it because he's here a lot, and posting a lot

Over the past three days, there has been as much as a 26% rise in the value of litecoin itself, the value of which many other scrypt coins seem to track.  And as litecoin is one of the coins we have been mining, I should hope that our profits measured in BTC are increasing as well!  What you are seeing might have nothing to do with the new server, or the whale not being here, or that you recently found your lucky rabbit's foot.

(If rabbit's feet were really that lucky, then how lucky was the rabbit!)

Unfortunately, over the last 3 days there has also been close to an equal rise in difficulty   LTC profitability has increased only marginally in that time (5%?).

[comeonalready]

As for phzi, he's just going to take the brunt of it because he's here a lot, and posting a lot

And having admitted it proudly seemingly knowing exactly what he was doing!

Over the past three days, there has been as much as a 26% rise in the value of litecoin itself, the value of which many other scrypt coins seem to track.  And as litecoin is one of the coins we have been mining, I should hope that our profits measured in BTC are increasing as well!  

Unfortunately, over the last 3 days there has also been close to an equal rise in difficulty  LTC profitability has increased only marginally in that time (5%?).

An excellent point of course..  I might have omitted that part of the equation simply to make a stronger point (in the absence of someone else who knows what he is talking about) .  Or I might have simply neglected to account for the increasing difficulty.  I'll never tell (but you could look a few pages back to see if I was aware of the litecoin situation.)

Looking forward to pool rejection rate statistics!

Update, there should be another significant litecoin difficulty increase in less than one hour!

[DrunkenDonkey]

DM
Thanks for sharing your settings and experience - I've got a couple of questions (please forgive me if they're silly due to my newness to all this!):
1) I am using xubuntu and cgminer 3.7.2 - can I use this xi with this?
2) Given that the "power" of the graphics card is used for mining does the fact that you can play movies etc mean that you're missing out on some of the spare "power" for mining and hence there is more hash/effective work that your setup could be doing?  I'm using "power" as I don't know what to talk about in terms of watts, hash to mean "processing power!

1. You need sgminer for that. I'm not a linux expert, called for help a friend that installed sgminer 4.1.0 on the rigs, didn't look all that complicated, but what do I know xi is based on shaders on the card, and there are odd times that it actually has the same value as I - for example on 280x/7950 I13 is perfect as it is hitting right on the shader count.

2. Yeah that is the theory - you should be using ever bit of speed on the card to get to work. The practice, however, differs. The programming on these cards is quite complex and often what seems logical results in illogical results, not sure if it is because of the way code is written, or optimizations, you can have the card working to the booth and still don't produce as fast results as a setting that is much more relaxed and can have the card with power to spare, including allowing you to watch movies (hashrate drops some 5-8% when watching movies tho). I totally agree that the aim should be best results for given clock/temp/consumption, so what I try to optimize is to find the most relaxed settings that give me the same/better efficiency than the commonly used ones. For example 280x the commonly used one is good enough, not the case with 7950, and to a degree - 290

I heard another explanation about intensity - it is internally how much processes are running, so upping it results in taking on many tasks, but every single one is completed slower - not that much of a problem with slow coins, but on fast blocks using higher intensity you are not giving it a chance to complete and submit the job, so the general advice on shitcoin mining is to find a low intensity that does the job good enough.

[poolwaffle]

So I'm looking into why we have so much unexchanged dogecoin.  At first glance, nothing is wrong on our side... Confirmed our sends are getting to the exchange, confirmed our trader is trading them for BTC, and confirmed our scripts are processing the trades correctly.  Everything looks fine.

On second glance, our dogecoin balance (locally on our coindaemon) says 0 balance, and on dogechain, says we have a decent amount of unspent coins at our address...  Going to check to see if maybe something got confused/missed when we were upgrading the wallet...

[vals]

So I'm looking into why we have so much unexchanged dogecoin.  At first glance, nothing is wrong on our side... Confirmed our sends are getting to the exchange, confirmed our trader is trading them for BTC, and confirmed our scripts are processing the trades correctly.  Everything looks fine.

On second glance, our dogecoin balance (locally on our coindaemon) says 0 balance, and on dogechain, says we have a decent amount of unspent coins at our address...  Going to check to see if maybe something got confused/missed when we were upgrading the wallet...

They've upped the block confirmation time with the new 1.6 fork. It now takes 240 confirmations which is around 4 hours (previously it was around 30 minutes).

[poolwaffle]

So I'm looking into why we have so much unexchanged dogecoin.  At first glance, nothing is wrong on our side... Confirmed our sends are getting to the exchange, confirmed our trader is trading them for BTC, and confirmed our scripts are processing the trades correctly.  Everything looks fine.

On second glance, our dogecoin balance (locally on our coindaemon) says 0 balance, and on dogechain, says we have a decent amount of unspent coins at our address...  Going to check to see if maybe something got confused/missed when we were upgrading the wallet...

They've upped the block confirmation time with the new 1.6 fork. It now takes 240 confirmations which is around 4 hours.

Don't think thats it, I think theres a daemon problem.  Specifically I'm seeing these: https://github.com/dogecoin/dogecoin/issues/341 show up in our send logs.  I'm guessing that has something to do with it.  Going to be upgrading the wallet right now, and rescanning/indexing.  Dogecoin mining will be down for a bit here (not sure how long this takes).