So my rig is working great now. Now all I need to do is spend another hour and a half finding stuff to do here so I cam post about it in the FPGA mining project thread.... Naw, I'll spend that time optimizing it instead.
Topic: n00b question regarding submitting results (Read 393 times)
Ah, I figured it out! The python script I borrowed has multiple threads, and uses an Event to communicate between the thread the stuffs the serial port and the thread that reads from the serial port, and that was screwed up, causing all units that gor sent back after the first one to have the wrong Work associated with it.
So my rig is working great now. Now all I need to do is spend another hour and a half finding stuff to do here so I cam post about it in the FPGA mining project thread.... Naw, I'll spend that time optimizing it instead.
So my rig is working great now. Now all I need to do is spend another hour and a half finding stuff to do here so I cam post about it in the FPGA mining project thread.... Naw, I'll spend that time optimizing it instead.
Hey!
I have a question, and found this forum. I searched around a bit to find the answer, but don't have the time to do Word Association for another 3 hours to be able to post in the Mining forum, so I'm posting it here instead.
I read up about Bitcoin recently and decided to try doing some mining (because I need a break from NetHack every now and then. ). I have an intimate relationship with FPGA's (although why you folks prefer to spend time with that dirty whore Verilog is beyond me), and managed to figure out how to make a FPGA mining rig on a borrowed dev board in my spare time this week. I also found two Python scripts -- one from the Bitcoin FPGA project and another that does CPU mining -- and when combined they seem to work OK for getting the getwork info from a pool and sending it to the FPGA, and receiving the solution back from the FPGA.
My issue is that I am having issues submitting the solution back to the pool. After I send my first submission, I get RPC read errors on getting the next getwork bundles for a little while. (I need to put the time in that log so I know how long this happens for....) That submission eventually does get accepted.
But after that point, all future submissions fail with RPC send errors (and when they fail, I don't see those other errors when I get the next batch of work anymore). (I should add that my Python is not very good at all, and I am assuming I have made some mistake along the way. But why would the first submission be accepted and the others rejected?)
My questions are:
1) Stupid question: an error when sending getwork(data) means I'm doing something incorrectly, right?
2) I've saved off the data and nonce in a debug file. Can I experiment with sending the same solution back to the mining pool and will it accept it at some point, even though it is old? Do those time out at some point?
2a) If I can't do this, is there a better way to test out how I am submitting the solution?
Any help anyone can give would be greatly appreciated. I don't have enough bitcoins yet to give out meaningful tips, though.
(My goal, incidentally, is to mine enough bitcoins over the next few months to give back the borrowed board and buy a few better ones, and maybe upgrade my PC a bit. That seems entirely reasonable, I think....)
I have a question, and found this forum. I searched around a bit to find the answer, but don't have the time to do Word Association for another 3 hours to be able to post in the Mining forum, so I'm posting it here instead.
I read up about Bitcoin recently and decided to try doing some mining (because I need a break from NetHack every now and then.
). I have an intimate relationship with FPGA's (although why you folks prefer to spend time with that dirty whore Verilog is beyond me), and managed to figure out how to make a FPGA mining rig on a borrowed dev board in my spare time this week. I also found two Python scripts -- one from the Bitcoin FPGA project and another that does CPU mining -- and when combined they seem to work OK for getting the getwork info from a pool and sending it to the FPGA, and receiving the solution back from the FPGA.My issue is that I am having issues submitting the solution back to the pool. After I send my first submission, I get RPC read errors on getting the next getwork bundles for a little while. (I need to put the time in that log so I know how long this happens for....) That submission eventually does get accepted.
But after that point, all future submissions fail with RPC send errors (and when they fail, I don't see those other errors when I get the next batch of work anymore). (I should add that my Python is not very good at all, and I am assuming I have made some mistake along the way. But why would the first submission be accepted and the others rejected?)
My questions are:
1) Stupid question: an error when sending getwork(data) means I'm doing something incorrectly, right?
2) I've saved off the data and nonce in a debug file. Can I experiment with sending the same solution back to the mining pool and will it accept it at some point, even though it is old? Do those time out at some point?
2a) If I can't do this, is there a better way to test out how I am submitting the solution?
Any help anyone can give would be greatly appreciated. I don't have enough bitcoins yet to give out meaningful tips, though.
(My goal, incidentally, is to mine enough bitcoins over the next few months to give back the borrowed board and buy a few better ones, and maybe upgrade my PC a bit. That seems entirely reasonable, I think....)
Jump to: