- cgminer uses its own version of the HID driver (Zadig). So in order to use cgminer you have to trick Windows to replace the driver for a particular device with Zadig's. I hate to be pedantic but this is not quite true. The device talks USB while HID is a driver designed for user interface devices like keyboards and mice, so when using software that uses this driver, it goes device -> usb -> hid driver -> software. Zadig is not actually a driver; it's a utility to tell windows to remove the driver and use the microsoft direct usb communication protocol (effectively a Microsoft driver called WinUSB). So while WinUSB appears to be a "driver" it's actually telling windows to just give the raw usb data to the software. So software that uses WinUSB (like cgminer which uses it for every single USB device) is talking directly to the usb device, bypassing the need for any driver, so it is going device -> usb -> cgminer. Having said that, we don't have support for 2+ chip nanofury based devices, but VS3 is kindly sending me some devices to add support to cgminer, thanks! |
|
|
|
There was a conflict of interest as far as I was concerned with there being an apparent "unofficial amt users thread" which should be posted in and run by the users but was actually a moderated thread by AMT, and there were already 2 or 3 other AMT user threads that served that purpose. Hence I deleted it.
|
|
|
|
So wait, everybody is talking about the unofficial AMT users thread.
Is it still up and I have been locked out or has it bee removed?
"The topic or board you are looking for appears to be either missing or off limits to you."
This topic has been moved to Trashcan.
by ckolivas There was a conflict of interest as far as I was concerned with there being an apparent "unofficial amt users thread" which should be posted in and run by the users but was actually a moderated thread by AMT, and there were already 2 or 3 other AMT user threads that served that purpose. Hence I deleted it. |
|
|
|
That would then discriminate against any other cscape based bitfury mining hardware that comes out with >6 chips and alas they all have no unique identifier apart from: iManufacturer 1 c-scape
iProduct 2 bi?fury
Which the bi*fury, hex*fury and these OSMs have as well. Now if you wanted to get the designer to change the iProduct in your devices it would be mighty helpful for both of us. I agree, we had thought about it but valued compatibility with existing cgminer program higher than the additional time required to create our own driver in cgminer. Now that you are supporting our OSMs, it's a different story. (It was our choice to not get you on board earlier). What do you think of naming the product OneStringMiner15, or OSM15 to have a unique iProduct and an indication of the number of chips on it ? The existing customers can chose to update their firmware once there is an advantage in doing so. In the mean time their OneStringMiner just keeps on running with the default Bitfury driver and settings. OSM or even a longer name without the number is enough since the communication protocol already reports back the number of chips, otherwise I'd need to manually add any unique combination that comes out every time you design a new board  |
|
|
|
It's 7 days. I could care less who he calls a scammer but the off topic nonsense and trolling is excessive, has been for a while.
Only seven days? The way he's complaining I thought it was a permaban. Given the amount of time he spends on these forums, 7 days is a lifetime... |
|
|
|
|
Just don't mine there and discourage anyone from mining there and they'll never have enough hashrate to successfully undo anything.
|
|
|
|
I don't know how the hell he comes up with most of what he says, it's gibberish. But I don't think he should be banned. We don't even ban scammers like bfl.
Having the highest post count on the forum full of mostly trash, sounds like a good reason to ban him permanently IMO. That 20k post count doesn't count deleted posts and probably half his posts have been deleted so you're seeing the best of the bunch  |
|
|
|
|
Most questions from newcomers to the forum are repeat questions and whenever people are asked why they didn't search first they insist they did and did not find what they were looking for...
Insert discussion about possible improvements to search facility here:
|
|
|
|
The changes in 4.2.3 were generic before I received one. It has since been fixed in git but there is no release with the new changes yet. I do not do nightly builds at this stage, but I'll likely release a new version tomorrow. It being reported as a BXF will NOT adversely affect its performance by the way; it's only cosmetic and the extra information in the API if you are reading it. |
|
|
|
What action did you guys take when one of the 4 cores stopped working? As though its not already bad with 20% less hashrate now one of the cores has stopped working reducing the hash rate to 1.2 Th/s.
Just restart cgminer from the config tab. |
|
|
|
|
It's 6.2 billion * 2^32 hashes on average to find a block.
|
|
|
|
|
Been asked a thousand times: No; you are mining to a particular address and can't change it afterwards.
|
|
|
|
I have a quick question.
I have a clean Win system all updated and no issues running my OneString Miner setup and cgminer every 8 hrs or so crashes. I have ran it alone and on another system and it just says it stops working. It will take multiple tries to get it restarted.
4.2.3 did not fix it either. I have no problems with any other miners or software I have tried on it to test if it was the system.
I've not seen this on linux, so try a debug build with the instructions here: http://ck.kolivas.org/apps/cgminer/debug/ |
|
|
|
Honestly, I'd much rather help decentralization.
Easy then, throw everything at p2pool. |
|
|
|
Who would know the difficulty would be 1000x harder now than last spring. Even though the btc price increased 6x.
Most informed miners should have known the difficulty would be 1000x more than last spring; this is why so many of us tried to warn newer miners - they just didn't listen and assumed we just wanted the mining to ourselves. The price no one can predict though. |
|
|
|
I've been watching the pool hashrate for several days now. On 04/09, it was at 201 TH/s, and 4/10 at 2 am CST, it was 214 TH/s. Right now it's at 126 TH/s. I think people are jumping ship, but won't that just create a huge variance payout for the ones left behind? What can we do to make p2pool more efficient? Are there any ideas? I don't understand how it works enough to say. I'm not giving up, I'm staying on p2pool indefinitely. I'm thinking something else is wrong, some kind of configuration problem maybe? The expected numbers on the block list are so high after a found block the day before. I've tweaked latency issues on my own computer, and I get about 92% consistently to maybe 97% efficiency. And I run nothing else but p2pool and BTC wallet.
The simple fact is that most miners are a combination of lazy, fickle, uninformed and just all round suspicious and seem to be unable to understand payout schemes, variance and luck no matter how much it's explained to them, and their care factor for things like security of the network, distributing the hashrate etc. etc. just isn't high enough to offset the fact that the primary thing they care about is profits. Then they do pseudo experiments ignoring all the things I said above and show a better payout at some random other pool for that day and they're lost for good. |
|
|
|
|
I think bearing in mind how busy a thread is as well. Some of those mining hardware threads can have a hundred posts in a day and they're impossible to navigate if they're also interspersed with extra posts of just LOL, +1, -1, awesome, etc...
|
|
|
|
|
CPU mining bitcoin stopped being profitable 2 and a half years ago. Today you'd spend $1000 in electricity to earn $1 worth of bitcoin, so no...
|
|
|
|
I have received 3 of these bare boards courtesy of BenTuras to aid software development. They are working fine on the current cgminer, though I have added some device specific tweaks which will be in the next release of cgminer due out soon. An example from each board follows: On screen: (BXF was based on Bi*fury and M specifies multiple, hence they're called MXF in cgminer) 5: MXF 3 : 48.2C | 30.74G / 30.68Gh/s WU: 430.0/m
And API stats output: ... Thanks! And I think the abbreviation should be OSM Looking forward to the next version of cgminer. That would then discriminate against any other cscape based bitfury mining hardware that comes out with >6 chips and alas they all have no unique identifier apart from: iManufacturer 1 c-scape
iProduct 2 bi?fury
Which the bi*fury, hex*fury and these OSMs have as well. Now if you wanted to get the designer to change the iProduct in your devices it would be mighty helpful for both of us. |
|
|
|
Thanks for your investigating. I build all the official binaries with mxe, so it works either way, but I have shuffled things slightly based on that info and removed the windows.h include entirely. I have no idea what effect that will have on other build environments I see, now i need to ask google about mxe  or you maybe can add some guide to windows-build.txt  That file is horribly outdated from the GPU days and since I started cross compiling with mxe there is no way I'm going to try and build it directly on windows or any other way any more, sorry. In fact that windows build file wasn't even made by me if you read it, but a very dedicated user... |
|
|
|
|
Show Posts
