Tested some of my hashing power here for a while since Terk said the reject rates had improved...  at max, CleverMining reported slightly under my actual hashrate, and rejects averaged over 20% still (and I am very close to the server...)!

Looking less profitable then wafflepool, and the BTC/day/MH rate published appears to be an utter joke since the site is definitely under reporting my hashrate (tested continuously for about half a day), and then I am losing an extra 20% to rejects.  Looks like nearly a 35% inflation of BTC/day/MH compared to wafflepool which has been a consistent a slight under estimate.

All my hashing power is back on wafflepool for now.  Will try again if reject rates actually improve.

Comes on guys... try looking thru his post history before asking questions that are already answered.

Seriously, where did you learn logic?  That makes little to no sense... If you're running a mining botnet, that hashing ultimately has to all be directed somewhere.  Some server is handling it, it is no less reasonable that a botnet operator would run a server to organize their mining power and fetch work from pools then just directing your drones directly at a pool - there is a single point handling everything either way.

You seem to be talking about consumer hardware, which makes little sense.  A key point poolwaffle made was regarding economies of scale - it would certainly not be 5mil USD, probably nowhere close.  Imagine a factory that produces Sapphire cards deciding to direct their high-binned chips to a special production line that makes equipment for mining for themselves.  Based on how much R9 290 chips were going for before this crazy mining craze (which means they would still be available at those prices in bulk if you have the right contacts), a farm like this could easily be built for under $2 million - and that's presuming it isn't an ODM just re-purposing their production line or something.

Furthermore, creating an effective multipool with excellent switching is damn complicated... if you're in the hardware industry, that doesn't mean you have great connections to the software industry.  And that money spent on a software developer would probably be better spent on firmware development to optimize your card BIOS for mining.  Why not offload the pool management and coin switching work to someone like poolwaffle for a measly 1% fee?  That seems perfectly reasonable to me.

Ultimately, I think this is none of our business.  We aren't asking you anyone else for proof that they are not running a botnet, why would that be different for someone with a few magnitudes more hashing power?


Ah there is sfire.  Good stuff.  Hope that resolves this conversation once and for-all.  I had a feeling it was him - refer to his post history for the hints he dropped about their operation in the MC thread.

Working fine here... No issues, solid acceptance rates from all my rigs, and good profit rate so far today.


No...  they aren't anywhere near tape-out yet.  It could be private ASIC development, or gridseed or something tho.

--

PoolWaffle has addressed the botnet question, so please please stop cluttering the thread about this.  It's just super annoying and serves no useful purpose.

Poolwaffle:

I generally agree with your thoughts regarding the efforts required to mine smaller coins.  But I think it is significant to consider that wafflepool can contribute at least part of its success because we didn't avoid high-diff coins like Doge (smart logic).  I also think multi-pools are generally damaging to alt-coins, and don't have a problem with that because I think there are simply too many of them popping up all the time.  But, if we never mine low-diff coins, I think we will be impacting the profitability of scrypt mining in general, much more rapidly.  

Maybe it would be worth redirecting a small portion of hashrate to low-diff coins that are relatively stable.  I am envisioning a system that would put a small subset of our miners on low-diff coins somewhat persistently.  Take cosmocoin - if you directed a small amount of the pool's hashrate there with the intent to only play a small roll in the network consistently (think, much smaller then 50% of the network hashrate), and try to target coins that we can just leave a chunk of hashrate on for the majority of the time. The quick math is my head seems to suggest this would increase the pool's profits overall by a few percentage.

Does that make any sense to you?  Maybe I will try to run some numbers to validate (or disprove) my thoughts.


You have to consider the potential behind scrypt mining in general, and how ASICs will impact scrypt alt-coins and big networks like LTC or DOGE.  At the moment, it seems unlikely that ASICs will have any real impact on global scrypt hashrates until Q3 at the earliest, so that gives us another 4-6 months minimum of GPU dominance.

But, will alt-coins remain this profitable to mine?  Ask yourself if you think there will be another DOGE, or if DOGE will remain profitable given another block halving.  Do we even need all these alt-coins, what is there real purpose?  Everyone will probably have their own speculative answers, and you need to decide what is most likely in your mind.

If you're thinking about investing in GPUs, also remember that miners use them as "Generic" Processing Units... so even if ASICs make GPU mining scrypt totally unprofitable, that doesn't mean that scrypt-jane, sha3, or various other algorithms won't be profitable to mine with GPUs, as ASICs will have no impact on those algorithms.  That is, if you think alt-coins will actually stick around and be profitable...

All this said, this is probably a much better topic for a separate thread, if you want to continue the discussion.

This also doesn't prove it's not a botnet of course (you can relay all your drone mining traffic thru a single server), but really people just need to just stop caring about these big mining accounts.  Nothing you can do about it really...  And it doesn't hurt us any.

Moreover, let's please keep the middlecoin-style unfounded speculation about botnets, or ASIC farms, or etc out of this thread.  It would be nice to think this thread can stay more mature then we've seen over at the MC bitch fest thread.

There are absolutely GPU farms this big in China or Korea... and there are also ASIC manufactures like gridseed that could be testing chips.  Stop beating a dead horse and please just leave the botnet vs whatever topic alone.  Who cares?

There are also people with 100s of GPUs in their garage... and this kind of hashing power would be only an order of magnitude larger.  Also consider that GPUs like R9 290s can potentially reach 1MH/s+ with effective tuning and BIOS modification.  No reason an ODM couldn't be building super performing MPU (Mining Processing Units) for themselves out of top binned Hawaii chips.  Not everyone is limited to consumer grade crud.

Lol... expecting coinwarz or coinchoose to actually give you a realistic profit rate is just silly. If it was that easy, these multi-pools wouldn't exist.

Anyway... variance... nobody will be interested in what you are saying unless you have at least a few days mining behind you for analysis.  Pools have good days and bad days. Luck and variance are huge in mining.  I have earned over .02BTC/MH some days recently,  and less then. 008BTC/MH on other days.  Overall, wafflepool has averaged around .012BTC/MH for me, which I consider fairly excellent.

This happened to eu the other day, and the shares were lost.  Oh well, knock it up to bad luck and be glad you switched to useast when you did.

Lol... mining is mining.  I take it you haven't been doing this very long?  If you want a reliable wage, then work a job.  Dogecoin halving cut profitability of multi-coin pools a bit, but wafflepool is still maintaining almost the same profit rates - impressive I would say.  And yes, a consistent .01BTC/MH/day is great.

If you think you can do better, then find a pool that will pay you better... good luck.  And please do post about your results.

Are you guys fast switching for variable block reward coins yet?  Ala, only mine when the next block is worth a certain amount.  Or are you still calculating profitability based on average coin reward?

Switch to another server.  Are you using uswest?  It seems to be borked at the moment.

No...  poolwaffle has addressed this in a recent post.  The reason we aren't mining small coins is because poolwaffle is changing the coin balancing system to more easily mine multiple coins.  Profits shouldn't be hurt much, if at all, because we aren't mining them, since the majority of our profits come from coins we are still mining.

I don't get what you are talking about... so far we are on target for just a bit under .01BTC/MH today.  If your shares are being reported on the pool, then you're getting paid at a pretty good rate.

5hr/24hr =~ .2 ...

Expected payout per day at .01BTC/MH with 2MH = .02BTC ...

0.00225 / .2 = .01125BTC / day ...

Perfectly reasonable given variance...  Stop trying to judge a pool over a 6hr period, that's just ignorant.

uswest isn't working for me right now.  I had to move my rigs to useast otherwise their shares aren't getting counted.

It is called variance...  Chill out and judge your stats on a weekly basis.  Even daily payouts won't be consistent.  If it shows your hashrate on the pool, then everything is working fine. Poolwaffle hasn't done anything because of the hashrate increase that would negatively impact profits in a big way.

The Fat Miner: I doubt my current issue is the same as yours.  I could restart my rigs and would have 100% shares counted again, but restarting these rigs is a royal pain in the ass because they are at a remote location and sometimes they don't restart properly.  So, I'd rather lose hashing power for a bit and have poolwaffle figure out why this is happening.

Looks like it might be affecting our big miner as well?  Their hashrate just fell down to about half...

Edit: switched my rigs to useast after losing 40 minutes of shares...  hashrate climbing on the site again.

Test for longer then 5 hours before complaining?  Yeesh...

That said, something is weird PoolWaffle - half of my shares aren't being counted right now, started just after 18:01 server time.  Shares are still be submitted by all my rigs, but half aren't being counted by the site stats (it appears 2 rigs in particular probably aren't getting their shares counted).  I have seen this before once a few days ago, and had to restart the machines.  Everything is pointed to uswest, but my rigs are divided between two locations - the second location seem to be what isn't being counted right now.  I will leave them without rebooting for a bit, so you can maybe take a look.

Wafflepool hash: 1625dbc1
worker names I suspect are not having their shares counted: phzi1, phzi2

Work restarts and new blocks have _nothing_ to do with your worker diff.  That's just because we are mining a fast coin, or switching between coins quickly.  Periodic rejects are normal (it happens when 2 blocks are found within seconds of each other, or a new block is found and then we switch coins right away).  Under 1% reject rate is great.

That is a fine reject rate.  You could probably tune your rig a bit for a slightly lower rejects (set expiry to say 5, and queue to 0, and maybe lower xintensity slightly or use 2 threads if possible), but for say a 1% gain in accepted shares, it's not likely to be worth your time.

Try wafflepool instead of just going back to middlecoin...  Then watch your rejects drop to almost nothing, and profits increase along side.

Thanks, I somehow missed that comment.

Poolwaffle: can we get block finder stats again? It is showing 0 with no history for a while now. That transparency seems very important in my mind, as it allows for miner stats to be compared to the pool stats to ensure all found blocks are being distributed to the pool.

Everyone: I don't know why you are worried about the pool growing. It is just decreasing our variance, and even with the doge block halving our profits are excellent right now.  Limiting membership or something similar is just silly - miners will just go to another profit switching pool and the small alt coins will still get mined either way. I don't know why anyone thinks the several GH address is a problem - servers seem stable snd responsive, and the user is probably mining at max diff.

Poolwaffle: can we get block finder stats again? It is showing 0 with no history for a while now. That transparency seems very important in my mind, as it allows for miner stats to be compared to the pool stats to ensure all found blocks are being distributed to the pool.

Everyone: I don't know why you are worried about the pool growing. It is just decreasing our variance, and even with the doge block halving our profits are excellent right now.  Limiting membership or something similar is just silly - miners will just go to another profit switching pool and the small alt coins will still get mined either way. I don't know why anyone thinks the several GH address is a problem - servers seem stable snd responsive, and the user is probably mining at max diff.