I'm confused, why wouldn't you write it as OpenCL? This provides at least some level of portability.
EDIT: That being said, it's a generous offer!
Topic: [Removed Bounty] PrimeCoin CUDA miner - page 2. (Read 8690 times)
Prob everyone has already looked around and found these but just encase
http://www.macs-site.net/CudaPrimes.htm CUDA Primes
https://github.com/MNorthwind/OpenCL-Prime-Numbers-Generator OpenCL
Of course this is just some basic stuff, prob lots of stuff needs to be done, mapping to the memory, dealing with how cores work
on the work, esp since blocks are completed so fast right now.. etc..
http://www.macs-site.net/CudaPrimes.htm CUDA Primes
https://github.com/MNorthwind/OpenCL-Prime-Numbers-Generator OpenCL
Of course this is just some basic stuff, prob lots of stuff needs to be done, mapping to the memory, dealing with how cores work
on the work, esp since blocks are completed so fast right now.. etc..
I thought AMD cards were better for Bitcoin mining because they were better at integer operations, and Nvidia was mainly good at floating point. Primecoin mining doesn't involve many floating point operations as far as I know, so targeting Nvidia specifically could be a waste of time.
Or you could just go onto the research site and pull down the data for and including the first 17 million digit primes....
Then use that as a 'feeder' mechanism....
Then use that as a 'feeder' mechanism....
Not sure if this answers your 2nd question, but:
"Block hash, the value that is embedded in the child block, is derived from hashing the
header together with the proof-of-work certificate. This not only prevents the proof-of-work certificate from being tampered with, but also defeats attempt at generating a single
proof-of-work certificate usable on multiple blocks on the block chain, since the block
header hash of a descendant block then depends on the certificate itself. Note that, if an
attacker generates a different proof-of-work certificate for an existing block, the block
would then have a different block hash even though the block content remains the same
other than the certificate, and would be accepted to the block chain as a sibling block to
the existing block."
Taken from the design paper.
"Block hash, the value that is embedded in the child block, is derived from hashing the
header together with the proof-of-work certificate. This not only prevents the proof-of-work certificate from being tampered with, but also defeats attempt at generating a single
proof-of-work certificate usable on multiple blocks on the block chain, since the block
header hash of a descendant block then depends on the certificate itself. Note that, if an
attacker generates a different proof-of-work certificate for an existing block, the block
would then have a different block hash even though the block content remains the same
other than the certificate, and would be accepted to the block chain as a sibling block to
the existing block."
Taken from the design paper.
-What purpose the sieve serves
In mathematics, the sieve of Eratosthenes, one of a number of prime number sieves, is a simple, ancient algorithm for finding all prime numbers up to any given limit. It does so by iteratively marking as composite (i.e. not prime) the multiples of each prime, starting with the multiples of 2.
The multiples of a given prime are generated starting from that prime, as a sequence of numbers with the same difference, equal to that prime, between consecutive numbers. This is the sieve's key distinction from using trial division to sequentially test each candidate number for divisibility by each prime.
The sieve of Eratosthenes is one of the most efficient ways to find all of the smaller primes (below 10 million or so)
awesome bounty btw i hope someone gets it
In mathematics, the sieve of Eratosthenes, one of a number of prime number sieves, is a simple, ancient algorithm for finding all prime numbers up to any given limit. It does so by iteratively marking as composite (i.e. not prime) the multiples of each prime, starting with the multiples of 2.
The multiples of a given prime are generated starting from that prime, as a sequence of numbers with the same difference, equal to that prime, between consecutive numbers. This is the sieve's key distinction from using trial division to sequentially test each candidate number for divisibility by each prime.
The sieve of Eratosthenes is one of the most efficient ways to find all of the smaller primes (below 10 million or so)
awesome bounty btw i hope someone gets it
Yeah, during my short digging I did find the sieve of Eratosthenes, though your explanation was better than what I found. My main question with the sieve is, since we are finding very huge primes, what purpose this 'table of primes' serves the finding of large prime chains?
You want to have a big table because the more you have, the more likely one of them will work for a given block.
Very true, I had figured something like that was the reasoning. It just struck me that most in this field(mining clients) would be more familiar with Opencl than CUDA.
Why not Opencl?
Imagine if we finally had a Nvidia coin, or at least one worthwhile to mine on nvidia..... would bring tons new people into the cyrptoworld.. good for all IMO
Why not Opencl?
-What purpose the sieve serves
In mathematics, the sieve of Eratosthenes, one of a number of prime number sieves, is a simple, ancient algorithm for finding all prime numbers up to any given limit. It does so by iteratively marking as composite (i.e. not prime) the multiples of each prime, starting with the multiples of 2.
The multiples of a given prime are generated starting from that prime, as a sequence of numbers with the same difference, equal to that prime, between consecutive numbers. This is the sieve's key distinction from using trial division to sequentially test each candidate number for divisibility by each prime.
The sieve of Eratosthenes is one of the most efficient ways to find all of the smaller primes (below 10 million or so)
awesome bounty btw i hope someone gets it
In mathematics, the sieve of Eratosthenes, one of a number of prime number sieves, is a simple, ancient algorithm for finding all prime numbers up to any given limit. It does so by iteratively marking as composite (i.e. not prime) the multiples of each prime, starting with the multiples of 2.
The multiples of a given prime are generated starting from that prime, as a sequence of numbers with the same difference, equal to that prime, between consecutive numbers. This is the sieve's key distinction from using trial division to sequentially test each candidate number for divisibility by each prime.
The sieve of Eratosthenes is one of the most efficient ways to find all of the smaller primes (below 10 million or so)
awesome bounty btw i hope someone gets it
Yeah, during my short digging I did find the sieve of Eratosthenes, though your explanation was better than what I found. My main question with the sieve is, since we are finding very huge primes, what purpose this 'table of primes' serves the finding of large prime chains?
-What purpose the sieve serves
In mathematics, the sieve of Eratosthenes, one of a number of prime number sieves, is a simple, ancient algorithm for finding all prime numbers up to any given limit. It does so by iteratively marking as composite (i.e. not prime) the multiples of each prime, starting with the multiples of 2.
The multiples of a given prime are generated starting from that prime, as a sequence of numbers with the same difference, equal to that prime, between consecutive numbers. This is the sieve's key distinction from using trial division to sequentially test each candidate number for divisibility by each prime.
The sieve of Eratosthenes is one of the most efficient ways to find all of the smaller primes (below 10 million or so)
awesome bounty btw i hope someone gets it
In mathematics, the sieve of Eratosthenes, one of a number of prime number sieves, is a simple, ancient algorithm for finding all prime numbers up to any given limit. It does so by iteratively marking as composite (i.e. not prime) the multiples of each prime, starting with the multiples of 2.
The multiples of a given prime are generated starting from that prime, as a sequence of numbers with the same difference, equal to that prime, between consecutive numbers. This is the sieve's key distinction from using trial division to sequentially test each candidate number for divisibility by each prime.
The sieve of Eratosthenes is one of the most efficient ways to find all of the smaller primes (below 10 million or so)
awesome bounty btw i hope someone gets it
As a result of extensive development of OpenCL miners for Primecoin with impressive performance, this bounty is now removed. GPU mining of Primecoin is going to soon be saturated. If you were currently developing something for this bounty (unlikely as all development seems to be Opencl and it has been months) PM me.
Note: mtrlt is making a OpenCL-miner. That doesn't qualify for this bounty.
New rule: CUDA miner has to somewhat-significantly outperform mtrlt miner on equivalently-priced hardware (aka CUDA miner that gets 800 coins/day on a $400 NVidia GPU on testnet will get this bounty if mtrlt's miner will get 500 coins/day on a $400 AMD GPU).
The first person who makes a CUDA miner that outperforms a CPU of equivalent value by more than 2x ($330 NVidia GPU, like a 670, should give more than 2x the blocks of an i7-3770 on testnet over the same period of time) will get 6 Bitcoins! If anyone else wants to add to the bounty, just post a message with your 'pledge'.
Requirements: compiled windows binary, source code provided.
Optional: compiled linux binaries.
Current bounty chip-ins:
Vorksholk: 6BTC
Alternately, if someone wants to dig through the primecoin code and give me a nice concise summary of exactly how the mining process works (purpose that the sieve serves, relation of block hash to potential primes, process of getting data from primecoind, process of submitting data to primecoind) I will either write my own, or attempt to hire someone with the USD equivalent.
If you provide a great explanation of all of the following (and are the first to do so), I will give you one BTC:
-What purpose the sieve serves
-What relation the block hash has to the potential block-solving primes
-Process of obtaining all mining data from primecoind
-process of submitting prime chain to mining data
Also, at heart, I love NVidia, I'm a NVidia fanboy.
New rule: CUDA miner has to somewhat-significantly outperform mtrlt miner on equivalently-priced hardware (aka CUDA miner that gets 800 coins/day on a $400 NVidia GPU on testnet will get this bounty if mtrlt's miner will get 500 coins/day on a $400 AMD GPU).
The first person who makes a CUDA miner that outperforms a CPU of equivalent value by more than 2x ($330 NVidia GPU, like a 670, should give more than 2x the blocks of an i7-3770 on testnet over the same period of time) will get 6 Bitcoins! If anyone else wants to add to the bounty, just post a message with your 'pledge'.
Requirements: compiled windows binary, source code provided.
Optional: compiled linux binaries.
Current bounty chip-ins:
Vorksholk: 6BTC
Alternately, if someone wants to dig through the primecoin code and give me a nice concise summary of exactly how the mining process works (purpose that the sieve serves, relation of block hash to potential primes, process of getting data from primecoind, process of submitting data to primecoind) I will either write my own, or attempt to hire someone with the USD equivalent.
If you provide a great explanation of all of the following (and are the first to do so), I will give you one BTC:
-What purpose the sieve serves
-What relation the block hash has to the potential block-solving primes
-Process of obtaining all mining data from primecoind
-process of submitting prime chain to mining data
Also, at heart, I love NVidia, I'm a NVidia fanboy.
Jump to: