BTW, if anyone else wants to implement ROCKMINER support, here are my notes from reverse engineering the protocol: Device->Host:
32-bit nonce
4-bit status
4-bit command
0 = nonce found
1 = Task complete
2 = Get task
2-bit product id
0 = R-Box (4 chips, freq 200-270-290)
6-bit chip id
8-bit task id
8-bit temperature
Host->Device:
256-bit midstate
128-bit unused/zero
8-bit 0xaa
8-bit (MHz / 10) - 1
8-bit chip id
8-bit 0x55
96-bit datatail
|
|
|
|
Thanks for the fast replies.
I bought my R-Box from Canary along with the power supply, so I assume it is adequate.
IF the cause is an unstable pool, I am wondering why that problem doesn't happen to my Antminer U2's? I should have been more specific, that there are 3 devices (R-Box and 2 U2's) but the only one that dies is the R-box.
It was a driver-specific bug when it is starved for work (which happens when the pools are unstable). |
|
|
|
My R-Box is "dying" after running a day or two. This has happened at least 3 times the first week I've owned it. I'm running with 2 Antminers U2's as well with the following command line bfgminer.exe -S rockminer:all --set rockminer:clock=270 -S antminer:all --set-device antminer:clock=x0A81 -o <pool URL> -u <username> -p <password> And it doesn't seem what frequency I run at, this is about all I can get. I've tried freq=270, 280, and 290. Running at lower frequency doesn't seem to help the stability. Am I doing anything wrong or is this typical? RKM 0: 42.0C | 32.54/32.48/32.12Gh/s | A:13 R:0+0(none) HW:4/none I think this can happen when pools aren't completely stable. I've put some improvements in git to try to handle it better, and hopefully recover no matter what stops it. |
|
|
|
Are you located in China? Is the selfish miner located in China? He claims to be named LiYi, and located in GuangZhou, China. 2 - Individual miners may not have standing to sue the selfish miner. In a civil case (involving money/damages) you must prove that damages be caused, but also that he damages were against you. There is clearly a relationship between the miners and the pool (the miners provide work for the pool and in exchange for each unit of work the pool provides a maximum amount of payment, if payment is less then the maximum then when the pool can afford to pay more then the maximum the units that got paid less get paid more). The relationship between miners at the pool are not as clear. I am not an attorney, but I think a likely ruling would be if a miner tried to sue another miner at the same pool, the judge would say that their "beef" is with the pool operator, not the selfish miner. On the other hand if the pool operator were to sue a miner the damages are more clear, as the miner did not provide the work, the miner said they provided the work, and the pool operator paid for the work that was not done. There is clearly a fraud here. Pools don't pay miners for work, merely coordinate cooperation between miners who pay each other. This is especially clear-cut on Eligius, where most of the funds never pass through the pool operator's hands. That may be how it is on a logistics standpoint, but is that how it is in the eyes of the law? If you were BTC Guild or ghash I would say defiantly no, as both of those pools have block rewards (and tx fees) paid to the "pool" wallet, and the BTC is then eventually transferred to miners' wallets via automatic payouts. Eligius is very different in that it pays the block rewards (and tx fees) directly to miners via a TX in the found block. Someone could argue what you are saying but they could also argue that since the pool determines who gets paid how much via the payout cue (this being embedded into the header of work provided by the pool - I think this is how it works) that the pool does really control the found blocks. Even a attorney could likely not answer this question with certainty, as I don't think this kind of dispute has been litigated before. The only person who can answer would be the judge that hears the case (and any appellate panel of judges that hear any appeals). I know for tax purposes, other pools are using this same interpretation. In theory he paid good money for this equipment. Supposedly he made it all himself. That means his primary cost is electricity (actual chips and PCBs do not cost very much to produce). As far as I can tell he has done this to multiple pools. Do you think it would be possible to modify mining software so that only the stratum shares are sent back to the pool with the correct header, but the other shares could use a different header (one that pay out to another address)? Do you have a way to determine when he withheld a block from the pool? If so can you compare that to other blocks found around that time, is there any consistency as to who found the blocks? I know that it has previously been determined that you cannot modify block headers to make a found block payout to your own address as the hash would be invalid, but someone who has the resources to have millions of dollars worth of mining equipment might have the resources to make this happen. It's not possible. I don't understand this part of your post entirely. |
|
|
|
You previously said that eligius lost ~400 BTC, based on $571 for 1 BTC that comes out to ~$228,400 that was stolen. Your largest mining address lost ~$14,850 from the selfish miner. If you look at your number 5 mining address they only lost ~$3,300 from the selfish miner. The point is that the amounts of individual miners are relatively small and probably would not be worth hiring an attorney over, also attorneys would probably want to be paid by the hour for a case with that much is dispute. If you were to hire an attorney to bring a case trying to recover the entire $228,400 (400 BTC) then there would be a better chance that an attorney would work on a contingent basis (agree to only get paid if they win and the payment would be taken out of the settlement/judgment). Probably right. Which would mean it'd have to be a class-action case (if there even is such a thing in China). As far as I know, nobody ever wins in class-action lawsuits...  2 - Individual miners may not have standing to sue the selfish miner. In a civil case (involving money/damages) you must prove that damages be caused, but also that he damages were against you. There is clearly a relationship between the miners and the pool (the miners provide work for the pool and in exchange for each unit of work the pool provides a maximum amount of payment, if payment is less then the maximum then when the pool can afford to pay more then the maximum the units that got paid less get paid more). The relationship between miners at the pool are not as clear. I am not an attorney, but I think a likely ruling would be if a miner tried to sue another miner at the same pool, the judge would say that their "beef" is with the pool operator, not the selfish miner. On the other hand if the pool operator were to sue a miner the damages are more clear, as the miner did not provide the work, the miner said they provided the work, and the pool operator paid for the work that was not done. There is clearly a fraud here. Pools don't pay miners for work, merely coordinate cooperation between miners who pay each other. This is especially clear-cut on Eligius, where most of the funds never pass through the pool operator's hands. |
|
|
|
In regards to the proof holding up in a criminal case it is important to understand how complex Bitcoin is. You need to be smart to understand even much of the basics as to how Bitcoin works. I would be surprised if you could explain to a jury (made of up "average" people, most of which likely would not have any technical background) how a pool works or how miners work in enough detail that would allow you to explain the evidence. I don't think courts usually require a full explanation of the technical details, just expert witness testimony that such and such is fact. If you were to testify, the defense attorney would ask what you think he did. Your response would be something along the lines of he withheld blocks that he founds while mining on our pool (you would explain what mining, pools are and what with holding blocks mean). The next thing he would ask is "how do you know" you would respond by saying something along the lines of "I looked at our pool records and saw x y and z" The defense attorney would ask to see the records and for you to explain what they mean. Having the records in a presettable format may be the difference between guilty and not guilty or 400 BTC or 0 BTC Perhaps. But should criminal charges be filed, the prosecutors probably have a budget for discovery, which I presume would include doing this kind of organisation of data. In regards to should he be paid if he is withholding blocks, if it appears that he withheld three blocks (for example) then 76 BTC (I would be aggressive with TX fees) should be withheld from his payment, at the very least. This is regardless if he was doing this intentionally or not and is especially true for such a large mining farm. Unfortunately, even after withholding the ~200 BTC, he still owes us like ~400 BTC. If you could find out his identity with relative certainty you could pursue civil charges against him. Assuming he was not mining via tor finding his identity shouldn't be more difficult then filing a lawsuit against the alias, then sending a subpoena to the ISP, data center until you can connect the dots to his identity. Yes, but arguably it should be some high-loss miner who would file these charges. I think wizkid057 is prepared to provide IP addresses to assist in any such lawsuit. |
|
|
|
whew...I'm stuffed on popcorn...this is getting good  Yea this is ridiculous. Not only that but noone has any real proof of any of this. There is sufficient proof to easily hold up in a criminal court case, IMO. I respect eligius a lot and would say that the reputation of eligius would be sufficient to believe you if you say that there is proof. However I would be very interested to see proof. That is something up to wizkid057 to try to organise and present. If it were me, I'd consider the effort to try to put it in a form understandable by anyone to be more trouble than it's worth. In regards to the proof holding up in a criminal case it is important to understand how complex Bitcoin is. You need to be smart to understand even much of the basics as to how Bitcoin works. I would be surprised if you could explain to a jury (made of up "average" people, most of which likely would not have any technical background) how a pool works or how miners work in enough detail that would allow you to explain the evidence. I don't think courts usually require a full explanation of the technical details, just expert witness testimony that such and such is fact. In regards to should he be paid if he is withholding blocks, if it appears that he withheld three blocks (for example) then 76 BTC (I would be aggressive with TX fees) should be withheld from his payment, at the very least. This is regardless if he was doing this intentionally or not and is especially true for such a large mining farm. Unfortunately, even after withholding the ~200 BTC, he still owes us like ~400 BTC. If it is apparent that he was doing this on purpose (if this person is who he says he is then he was doing it on purpose) then there is no reason to provide payment at all IMO. Intentionally withholding blocks from pools will degrade confidence in pools, which would lead to a decreased number of people mining in the first place (they would only solo mine and only if they could do it when they have enough hahspower that luck will not be a big issue), which would lead to centralization of mining. The more he talks, the more I get convinced it was intentional. But we may never know for sure. |
|
|
|
I already did it and now....i wonder if can still do something.
I never received my first pay, can anyone help me in this situation? The address did, so you will have to deal with whoever controls it. |
|
|
|
Question: This may sound super stupid, but I have been checking the docs and I don't see it.
Can BFGMiner or CGMiner issue explicit commands to the miner while running? I'd like to do a ZCX on a BFL Jalapeno while it is running (trying to lower the 1.0 volt voltage while hashing to see what effect it has) and I can't figure it out. Or do I have to use screen in some way I'm not (I can run Windows or PiUnix)
Thanks
C
There's a hidden RPC option to issue arbitrary commands, but I don't think it can handle multiple line responses, so not sure how it would affect things. |
|
|
|
Large hashing power will always be very carefully on selecting pools.
The eligius pool is now a "clean" one with a "potential" tiny black corner, and others 3 big is totally clean like the sky until now.
If you own 1000+T, which one will you choose?
What does Bruce mean by that? He's trolling. Basically saying when someone with 1Ph/s is looking for a pool, they will skip Eligius because of his slander. Too bad for him, Eligius is a no-fee pool, and the pool operator couldn't care less if people go elsewhere... (or at least that was my attitude when people tried crap like that) |
|
|
|
whew...I'm stuffed on popcorn...this is getting good Yea this is ridiculous. Not only that but noone has any real proof of any of this. There is sufficient proof to easily hold up in a criminal court case, IMO. |
|
|
|
[Do the people in your mining group know that you were involved in this withholding attack?
This kind of statement is typically pejorative - and its not been just you. Who knows for sure if this was an attack (as in the sense of it being willful)? Block withholding due to a bug in mining code that has now been fixed is one thing, an intentional block withholding attack is another. Brucexie does not appear to deny the former - the extant narrative/claims suggests that there was a cgminer bug in their kit which got fixed - the time line on this is still a little unclear and it would be helpful if this was clarified. So far as I can tell there is no evidence (that any one has seen presented so far on this forum) to support that this was willful. Surely one needs this evidence in order to assess it; otherwise one is guessing. Not really. Whether it was intentional or not is irrelevant. They didn't contribute to the pool, that is a fact. Therefore, they should not benefit from the pool. |
|
|
|
cex has zero incentive to give up any hashrate considering it's their main profit now (fees from trades within their system) Why do you think they have a 0% pool in the first place?
The hashrate they employ for the Gh/s traded on CEX is not the problem. Many miners unrelated to their services choose their pool to mine and that's contributes causes the majority of their pools hashrate. It's because people choose to mine there that it has all this hashpower, not because they are using half of the asics in this planet. You're wrong. Both are serious problems. Having half the ASICs on the planet is the bigger problem, since miners using the pool can switch to another one. Pointing their ASICs at other pools does not help the problem at all, only hides it. |
|
|
|
I can probably create one for Raspi SD card - have never done this before and not quite sure where to upload and share this rather large image. Let me see what I can do.
Should Hashra be listed in README as a GUI/OS based on BFGMiner? Let me know where to put it  |
|
|
|
CEX.IO has been working on possible ways of decentralising Bitcoin mining since February. We will present our solutions to the Bitcoin community in the nearest time and ask you for understanding our intentions in the right way. The Bitcoin community has been working on this since 2012. Why are you working in secret instead of collaborating with the ongoing efforts? In fact, your pool doesn't even have support for the current progress in decentralisation (GBT) today... |
|
|
|
Hi Luke
Vladimir say that is better connect miso-mosi-sck of atmel328 with misoout mosiout sckout of bitfury chip. Then write a driver like nf1_init (sorry but in vb.net because it is the language that I know better) and you should translate in C.
About the algorithm that manage the energetic behaviour of device, it should be written in external file "PowerManagementScript" so anyone could change it.
What do you think?
Anyone can change C code for now. Maybe RPC can be used to enable scripting it... |
|
|
|
Just curious, how can you be confident this was a hack? I mean what is the probability that they've just had bad luck and never got a winning answer but have genuinely been playing fairly? I've seen the evidence and am convinced (multiple different ways, even) that the miner in question was in fact block withholding. Whether it was intentional or accidental, I consider irrelevant. wizkid057 plans to release more information at some point publicly, I'm not sure when. For your " demanding 1164%-APY interest", Let me paste my last post here:
...
2. Pay under a installment plan: pay 10% per week, or pay 5% a week, with a extra 5% interest.
I calculate 5%/week as 1264% APY... 1.05 52 = 12.642808 Also, any interest beyond inflation is evil. If I really wants to make eligius "bad", The right way is to use 10 btc to hire a buck of hacker in china to DDOS that website, instead of risking my 200+ BTC on pretending to be a normal miner. Eligius is pretty much under constant DDoS these days, so you'd just be wasting your 10 BTC. And he would had had to have planned this in advance -- it would be trivial to find blocks that paid out to Eligius miners that are not in that list. So either Eligius is attacking you, or you are attacking Eligius. (And if WK thought that withholding blocks was a good strategy, why would he focus on one user?)
Edit: Dammit. I suppose EvilEligius could just assign blocks to another user -- On reflection, there is nothing really to tie the block the finder.
Actually, this is one of the items planned to be improved for GBT2... But even now, it's possible to store enough share data to prove that the block was found by the same person finding <bunch of other shares> - though I doubt wizkid057 stores that much data Oh, you give me an idea.
To hire a PR professional to dress up myself as a selfless man, after being trusted, then start to steel 200+ btc from miner.
200+ btc is about 120K USD, which is enough to hire such part-time expert in mid euro, india, or china.
So it is really profitable.
If you are educated to think independently, my word is for your consider. If wizkid057 wanted to do this, why wouldn't he take more BTC? 200 BTC isn't that much to be worth the risks of theft... |
|
|
|
Sorry for being not too clued up on the software that runs all the hashing to do with Bitcoin, but
why is there no time limit for work to be returned ? There is, 1-2 minutes. Miners cannot lie positively (claiming there is a share at X when there isn't) because the pool double-checks every share, but they CAN lie negatively (claiming there is no share at X where there is) which is what block withholding does. |
|
|
|
they threatened putting a 200 BTC bounty on hacking Eligius. LOL, if someone could hack Eligius to get 200 BTC, why would they be waiting for a bounty? |
|
|
|
I should connect miso mosi sck in miso out mosi out sck out and a free gpio as chip select?
I'm not a hardware person, sorry. I would think you'd want MISO/MOSI/SCK to connect to the mining chips when GPIO-N is low, and connected to the microcontroller when GPIO-N is high. But maybe it makes it easier or more sense to just chain the microcontroller on the end of the chips? |
|
|
|
|
Show Posts
