Yeah well that does not really make sense to me. Are they the same person ?
Why would Clipse ( who is paying 120 % ) go to Goat ( who is paying 115% ) !?
They are hopping the same pool(s) and one of them is down. |
|
|
|
Post some proof and instructions how to do this if we are to believe you. Just sayin'  You need instructions on hopping or proof of its profitability ? Just search this forum. Its not exactly new. As for proof about clipse.. I cant prove clipse is hopping, let alone which pools, but I think its proven his original (and now removed) claims were flatout lies, so the burden of proof is upon him. That said, I can prove bitclockers is being hopped by a hashrate thats higher than Clipse+Goat+GPUmax: https://bitcointalk.org/index.php?topic=61985.msg737310#msg737310Havent checked the other prop pools yet, but bitclockers is now down. If it stays down for a few days, it will be interesting to see what happens to clipse's hashredirector. Something tells me he will suddenly lose a lot of customers in South Africa and no longer have such "need" for hashrate. Time will tell. |
|
|
|
nVidia have 8 GPU limit as well.
I dont think they have. Windows, more specifically, WDDM has a 6 or 8 ?GPU limit for display devices. Fiddling with the registry can overcome that, by making some cards not be recognized as display devices (they wont function anymore if you connect a monitor). Thats also why XP doesnt have this problem, it doesnt have WDDM Anyway, I distinctly remember reading an nvidia engineer describing how to do that, and saying there was no hard driver limit to the amount of GPUs. That was in the context of CUDA, but it shouldnt be different for OpenCL. |
|
|
|
188 vs 228 I believe. The 2GB version takes an 8 pin and a 6 pin.
The price is still too high. Hopefully no one buys them and they dump them <$150.
You cant deduct power consumption from the PCie connector. 40W TDP extra at full speed and full bandwith already seems highly doubtful to me for 1GB extra (more likely, those cards are clocked higher?), but even if that is the case, lowering the memory clock to 200 Mhz will cut that powerconsumption difference by >5x and since mining uses very little memory bandwidth the actual power consumption difference is likely even far less. IOW, you are probably looking at just a few watts. |
|
|
|
Or is something wrong with my config?
No, but there is something wrong with you not reading what was discussed here ad nauseam before asking. If its asked ad nauseum, perhaps it should be put it in the first post. I read the first 4 or 5 pages of this thread before asking anything, I dont think you can expect everyone to read all 25 |
|
|
|
|
Thanks for the headsup DnT. I didnt realize BAMT now supported cgminer, or even that it was still supported. That looks really nice. Linuxcoin works fine ATM, and Im pretty familiar with ubuntu, less so with debian, but those stats look tempting enough to give BAMT a spin.
|
|
|
|
As long as everything goes according to plan, I will be bringing on about ~5Gh online to bitminter in the next few days.  I hope to be a big part of the family for a long while. Running BAMT. I have two rigs 3x5970s on ASRock 970 Extreme 4, Sempron 140 and Seasonic 1250 80+ Gold. Not worried yet, but first boot the system did not see my 2&3rd cards. I might need some help if I cannot resolve this myself. I have yet to try BAMT, but you probably just need to run aticonfig --adapter=all --initial (add sudo if its debian based without root account) For now, I need some general advice.... 1.How long to burn-in the gpus at stock? Im not convinced burning in does anything really; at most it could help settle thermal pads if its a new card, but that wouldnt take very long. A day or so? 2.How far to push OC settings after burn-in? No one can answer that for you. Whatever clocks are stable and dont push temps too high. The latter is also quite subjective. I personally dont like temps over 70C, but keeping 5970s below that may not be easy. 3.Should I underclock the CPU? edit: I misread. You asked about underclocking the CPU, not GPU. Yeah, by all means, underclock it. Below answer was for underclocking the GPU: If you have to pay for electricity, perhaps. Although its not underclocking you should be concerned about, but undervolting. Lowering voltage will significantly reduce temperatures and power consumption (the relation is quadratic), while usually only marginally lowering max safe overclocks. You will have to experiment. Underclocking at the same voltage makes little sense, unless you have a thermal problem. You will be less powerefficient at lower clocks, since power consumption of the videocards scales linear with clockspeed, and the overhead (CPU, motherboard, etc) doesnt scale at all. So it may sound strange, but given a certain voltage, higher clockspeeds will be more power efficient. Any other tips for a newbie?
Yeah. Stick around  And listen to Death and Taxes. He knows all there is to know about 5970s and I just saw he is playing with BAMT, so Im sure he will correct any mistakes I just typed . |
|
|
|
Good to know its normal. And Im sure there is a good reason for it, but arent you causing a lot of stales by having to force a restart every 10 seconds?
Yes, but those stales aren't lost. They can still be block solutions, and payouts are computed by your number of shares relative to everyone else's, so if everyone gets the same amount of stales, payouts are completely fair. However, some miners try to minimize network overhead by dropping share solutions when a long poll response comes in. This could possibly cause P2Pool to lose a true block solution (the same applies to any merged mining pool), so people should run CGminer with --submit-stales or use newer versions of CGminer that do that automatically. Poclbm also has this behaviour with no way to disable it, as of yet ... I'll put in a pull request. Okay, thanks for the explanation. I guess I have some reading to do |
|
|
|
|
Good to know its normal. And Im sure there is a good reason for it, but arent you causing a lot of stales by having to force a restart every 10 seconds?
|
|
|
|
I just pointed my non dedicated miner to p2pool to give it a try. It seems to work (hurray!), but I have at least one question: why does cgminer report so many longpoll events when mining on p2pool? Here is recent sample: 2012-02-08 21:34:41] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:34:46] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:34:52] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:34:57] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:06] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:09] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:22] Accepted 00000000.e626df82.c92c7a5e GPU 0 thread 0 pool
2012-02-08 21:35:25] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:30] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:34] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:37] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:39] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:42] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:46] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:55] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:35:58] Accepted 00000000.8e5eab72.9e42f8e9 GPU 0 thread 0 pool
2012-02-08 21:36:04] Accepted 00000000.652b3770.be6a3762 GPU 0 thread 1 pool
2012-02-08 21:36:06] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:36:07] Accepted 00000000.8072fb91.64a7cfea GPU 0 thread 0 pool
2012-02-08 21:36:12] Accepted 00000000.4dc0dc0c.77f4e052 GPU 0 thread 0 pool
2012-02-08 21:36:14] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:36:18] Accepted 00000000.9ebbb9a6.7e98eaad GPU 0 thread 1 pool
2012-02-08 21:36:26] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:36:30] LONGPOLL requested work restart, waiting on fresh work
2012-02-08 21:36:31] Accepted 00000000.9aabb60a.99a24530 GPU 0 thread 1 pool
2012-02-08 21:36:31] LONGPOLL requested work restart, waiting on fresh work On traditional pools, youd expect about 1 longpoll every ~10 minutes (or maybe 4 or so when merged mining). Why does p2pool generate so many more? Or is something wrong with my config? |
|
|
|
Pool hopping is detectable by the pool op. If a user's per-share reward value is consistently higher than average, they are gaming the payout system. In aggregate data like the total per-share reward vs round time, observation by outsiders will be masked by the total pool size and the modest counter-measures, but on a per-user basis, there will be users who earn the average expected per-share reward, and hoppers that stick out like a sore thumb.
Agreed. But since I made the post, Deepbit has linked to their own thread were is shown that hoppers are not only detected, but also labeled as such. Just no action is being taken against them - for now. I hadnt seen that when I made my analysis, I just tried proving deepbit can be and is being hopped. That much is beyond doubt now. I do agree because of deepbits size, the current impact on non-stop miners by hoppers is trivial and statistically noise, that was never my concern. |
|
|
|
Good luck getting your molex in there.
Point taken. Allow me to rephrase: I dont know anyone that stupid . At least not personally  . |
|
|
|
No, if our system detects a proxy pool (bithopper) it won't allow the purchase to start. We also do research on unknown pools to verify if they are mining bitcoin blocks. In reality most of the proxy pools can't handle the speeds we throw at them so they are blacklisted anyway.
Thats certainly a lot better than those proxy pools, and I appreciate the effort, but there is only so much you can do; there is no telling if one day some "Lucky Sr" decides to take down some alt coin or attempt a double spend attack and buys hashing power at gpumax to complement his established pool's. Of course thats true for traditional pools as well, difference being that traditional pools at least have to earn trust to gather a sizable membership over a long period of time, whereas with gpumax, they can just buy it instantly. Q: with GPUmax, do you as a miner have control over which buyers you allow to bid on your hashpower? So you can blacklist or whitelist specific pools? BTW, semi off topic, but I notice bitminter is blacklisted by you due to high stales. AFAIK, that problem was solved 1 or 2 days ago. At least Im currently seeing only 0.1% stales since the new backend went live. |
|
|
|
The users of GPUMAX know exactly where their hashing power goes
Yes (*), but he was talking about goat and clipse's proxies. anyone mining there definitely doesnt know what their hashing power is used for. I sometimes wonder if goat and clipse know themselves. Anyway, had to add a *, because, I assume (?) proxy pools could also buy hashing power at gpu max? If so, the miner has again no clue where his hashes ultimately end up. But I havent checked out gpumax yet, so I could be wrong. |
|
|
|
Would somebody who is on PPS be effected more than Proportional? Because i'm on Proportional, the avg. payout was higher within a few days time at my current Mhashs. PPS would not be affected. Of course, that has to be put in perspective, considering deepbit charges, what 10% fee for PPS? That makes hopping a total non issue by comparison, but you still get paid 10% less than on any (hop proof) zero fee pool though. |
|
|
|
no no, no one but p4man thinks deepbit is getting hopped. God you're such an idiot. "No one but me" and Deepbit themselves  Welcome to my ignore list again. I wonder why I ever let you out. |
|
|
|
I never understood the purpose to that, my payout is more or less consistent within 1-2%, very rarely (maybe once every other day) i will only get 80/160,000 shares and it'll be higher or lower than normal, but as far as i can see Pool hopping hasn't done anything to my payout if it's happening at all. (Deepbit)
If the hopping were to happen constantly, you wouldnt see it from your payout, it would always be "normal" but x% lower than 100% PPS for the current difficulty. Im not saying I think the impact from hopping is significant though, at least not compared to the fees you are paying anyway . |
|
|
|
|
Looks like a new heatsink. If they have any sense, the board is now mounted sideways, with the top and bottom fan working to together to blow air over the heatsink in the direction of the grooves. If the bottom fan is mounted to blow against the underside of the PCB, then, well, I cant imagine anyone being that stupid.
|
|
|
|
|
I just calculated the profit for poolhoppers on Bitclockers; assuming they jumped ship at 43% CDF, during my sample interval they would have achieved a payout per share thats 110% the expected payout and 121% the payout non stop miners got on that pool.
Hopping earlier, which requires more pools to hop, increased that slightly, but not by much.
Its worth noting, the more hoppers there are, the less those hoppers can benefit (if all miners are hoppers, they gain nothing); considering the enormous ratio hoppers/non hoppers on bitclockers, I suspect this is pretty much a worst case scenario. Ill try simulate deepbit later, assuming one could hop it, Im fairly certain it will be much closer to, if not over 150%.
|
|
|
|
Yeah, I highly doubted deepbit was getting heavily hopped.
Non sequitur, depending how you define "heavily".My point still stands. AFAICT, they are being hopped by something like 5-10% of the hashing power. That I cant prove an effect on the payouts of non stop miners is entirely to be expected since Id need a much bigger sample for that than just the last few days available on their site; there are many reasons you can get slightly above expected payouts (excluding their high fees of course) on such a short span, the most important one being pool luck. Other factors are stales, since I have no data for them. Since they occur more frequently at the beginning of new blocks, it could chance the result significantly. |
|
|
|
|
Show Posts
