diff --git a/mining/arpow_miner.py b/mining/arpow_miner.py
new file mode 100644
index 0000000..8720cc9
--- /dev/null
+++ b/mining/arpow_miner.py
@@ -0,0 +1,104 @@
+from pyethereum import utils
+import random
+
+
+def sha3(x):
+    return utils.decode_int(utils.zunpad(utils.sha3(x)))
+
+
+class SeedObj():
+    def __init__(self, seed):
+        self.seed = seed
+        self.a = 3**160
+        self.c = 7**80
+        self.n = 2**256 - 4294968273 # secp256k1n, why not
+
+    def rand(self, r):
+        self.seed = (self.seed * self.a + self.c) % self.n
+        return self.seed % r
+
+
+def encode_int(x):
+    o = ''
+    for _ in range(8):
+        o = chr(x % 256) + o
+        x //= 256
+    return o
+
+
+ops = {
+    "plus": lambda x, y: (x + y) % 2**64,
+    "times": lambda x, y: (x * y) % 2**64,
+    "xor": lambda x, y: x ^ y,
+    "and": lambda x, y: x & y,
+    "or": lambda x, y: x | y,
+    "not": lambda x, y: 2**64-1-x,
+    "nxor": lambda x, y: (2**64-1-x) ^ y,
+    "rshift": lambda x, y: x >> (y % 64)
+}
+
+
+def gentape(W, H, SEED):
+    s = SeedObj(SEED)
+    tape = []
+    for i in range(H):
+        op = ops.keys()[s.rand(len(ops))]
+        r = s.rand(100)
+        if r < 20 and i > 20:
+            x1 = tape[-r]["x1"]
+        else:
+            x1 = s.rand(W)
+        x2 = s.rand(W)
+        tape.append({"op": op, "x1": x1, "x2": x2})
+    return tape
+
+
+def runtape(TAPE, SEED, params):
+    s = SeedObj(SEED)
+    # Fill up tape inputs
+    mem = [0] * params["w"]
+    for i in range(params["w"]):
+        mem[i] = s.rand(2**64)
+    # Direction variable (run forwards or backwards)
+    dir = 1
+    # Run the tape on the inputs
+    for i in range(params["h"] // 100):
+        for j in (range(100) if dir == 1 else range(99, -1, -1)):
+            t = TAPE[i * 100 + j]
+            mem[t["x1"]] = ops[t["op"]](mem[t["x1"]], mem[t["x2"]])
+        # 16% of the time, we flip the order of the next 100 ops;
+        # this adds in conditional branching
+        if 2 < mem[t["x1"]] % 37 < 9:
+            dir *= -1
+    return sha3(''.join(encode_int(x) for x in mem))
+
+
+def PoWVerify(header, nonce, params):
+    tape = gentape(params["w"], params["h"],
+                   sha3(header + encode_int(nonce // params["n"])))
+    h = runtape(tape, sha3(header + encode_int(nonce)), params)
+    print h
+    return h < 2**256 / params["diff"]
+
+
+def PoWRun(header, params):
+    # actual randomness, so that miners don't overlap
+    nonce = random.randrange(2**50) * params["n"]
+    tape = None
+    while 1:
+        print nonce
+        if nonce % params["n"] == 0:
+            tape = gentape(params["w"], params["h"],
+                           sha3(header + encode_int(nonce // params["n"])))
+        h = runtape(tape, sha3(header + encode_int(nonce)), params)
+        if h < 2**256 / params["diff"]:
+            return nonce
+        else:
+            nonce += 1
+
+params = {
+    "w": 100,
+    "h": 15000,  # generally, w*log(w) at least
+    "diff": 2**24,  # initial
+    "n": 1000
+}
diff --git a/mining/hashimoto.py b/mining/hashimoto.py
new file mode 100644
index 0000000..763e51a
--- /dev/null
+++ b/mining/hashimoto.py
@@ -0,0 +1,128 @@
+try:
+    shathree = __import__('sha3')
+except:
+    shathree = __import__('python_sha3')
+import random
+import time
+
+
+def sha3(x):
+    return decode_int(shathree.sha3_256(x).digest()) #
+
+
+def decode_int(s):
+    o = 0
+    for i in range(len(s)):
+        o = o * 256 + ord(s[i])
+    return o
+
+
+def encode_int(x):
+    o = ''
+    for _ in range(32):
+        o = chr(x % 256) + o
+        x //= 256
+    return o
+
+P = 2**256 - 4294968273
+
+
+def produce_dag(params, seed):
+    o = [sha3(seed)]
+    init = o[0]
+    picker = 1
+    for i in range(1, params["n"]):
+        x = 0
+        picker = (picker * init) % P
+        curpicker = picker
+        for j in range(params["k"]):
+            x |= o[curpicker % i]
+            curpicker >>= 10
+        o.append((x * x) % P)  # use any "hash function" here
+    return o
+
+
+def quick_calc(params, seed, pos):
+    init = sha3(seed)
+    known = {0: init}
+
+    def calc(p):
+        if p not in known:
+            picker = pow(init, p, P)
+            x = 0
+            for j in range(params["k"]):
+                x |= calc(picker % p)
+                picker >>= 10
+            known[p] = (x * x) % P
+        return known[p]
+
+    o = calc(pos)
+    print 'Calculated pos %d with %d accesses' % (pos, len(known))
+    return o
+
+
+def hashimoto(daggerset, params, header, nonce):
+    rand = sha3(header+encode_int(nonce))
+    mix = 0
+    for i in range(40):
+        shifted_A = rand >> i
+        dag = daggerset[shifted_A % params["numdags"]]
+        mix ^= dag[(shifted_A // params["numdags"]) % params["n"]]
+    return mix ^ rand
+
+
+def get_daggerset(params, block):
+    if block.number == 0:
+        return [produce_dag(params, i) for i in range(params["numdags"])]
+    elif block.number % params["epochtime"]:
+        return get_daggerset(block.parent)
+    else:
+        o = get_daggerset(block.parent)
+        o[sha3(block.parent.nonce) % params["numdags"]] = \
+            produce_dag(params, sha3(block.parent.nonce))
+        return o
+
+
+def mine(daggerset, params, header):
+    nonce = random.randrange(2**50)
+    orignonce = nonce
+    origtime = time.time()
+    while 1:
+        h = hashimoto(daggerset, params, header, nonce)
+        if h <= 2**256 / params["diff"]:
+            noncediff = nonce - orignonce
+            timediff = time.time() - origtime
+            print 'Found nonce: %d, tested %d nonces in %f seconds (%f per sec)' % \
+                (nonce, noncediff, timediff, noncediff / timediff)
+            return nonce
+        nonce += 1
+
+
+def verify(daggerset, params, header, nonce):
+    return hashimoto(daggerset, params, header, nonce) \
+        <= 2**256 / params["diff"]
+
+
+def light_hashimoto(seedset, params, header, nonce):
+    rand = sha3(header+encode_int(nonce))
+    mix = 0
+    for i in range(40):
+        shifted_A = rand >> i
+        seed = seedset[shifted_A % params["numdags"]]
+        # can further optimize with cross-round memoization
+        mix ^= quick_calc(params, seed,
+                          (shifted_A // params["numdags"]) % params["n"])
+    return mix ^ rand
+
+
+def light_verify(seedset, params, header, nonce):
+    return light_hashimoto(seedset, params, header, nonce) \
+        <= 2**256 / params["diff"]
+
+params = {
+    "numdags": 40,
+    "n": 250000,
+    "diff": 2**14,
+    "epochtime": 100,
+    "k": 3
+}
diff --git a/mining.go b/mining/mining.go
similarity index 100%
rename from mining.go
rename to mining/mining.go
diff --git a/mining.py b/mining/mining.py
similarity index 100%
rename from mining.py
rename to mining/mining.py
diff --git a/mining/python_sha3.py b/mining/python_sha3.py
new file mode 100644
index 0000000..fa72032
--- /dev/null
+++ b/mining/python_sha3.py
@@ -0,0 +1,411 @@
+#! /usr/bin/env python
+# coding: utf-8
+
+# The Keccak sponge function was designed by Guido Bertoni, Joan Daemen,
+# Michaël Peeters and Gilles Van Assche. For more information, feedback or
+# questions, please refer to their website: http://keccak.noekeon.org/
+#
+# Based on the implementation by Renaud Bauvin,
+# from http://keccak.noekeon.org/KeccakInPython-3.0.zip
+#
+# Modified by Moshe Kaplan to be hashlib-compliant
+#
+# To the extent possible under law, the implementer has waived all copyright
+# and related or neighboring rights to the source code in this file.
+# http://creativecommons.org/publicdomain/zero/1.0/
+
+
+import math
+
+
+def sha3_224(data=None):
+  return Keccak(c=448, r=1152, n=224, data=data)
+
+
+def sha3_256(data=None):
+  return Keccak(c=512, r=1088, n=256, data=data)
+
+
+def sha3_384(data=None):
+  return Keccak(c=768, r=832, n=384, data=data)
+
+
+def sha3_512(data=None):
+  return Keccak(c=1024, r=576, n=512, data=data)
+
+
+class KeccakError(Exception):
+  """Custom error Class used in the Keccak implementation"""
+
+  def __init__(self, value):
+    self.value = value
+
+  def __str__(self):
+    return repr(self.value)
+
+
+class Keccak:
+  def __init__(self, r, c, n, data=None):
+    # Initialize the constants used throughout Keccak
+    # bitrate
+    self.r = r
+    # capacity
+    self.c = c
+    # output size
+    self.n = n
+
+    self.b = r + c
+    # b = 25*w
+    self.w = self.b // 25
+     # 2**l = w
+    self.l = int(math.log(self.w, 2))
+
+    self.n_r = 12 + 2 * self.l
+
+    self.block_size = r
+    self.digest_size = n
+
+    # Initialize the state of the sponge
+    # The state is made up of 25 words, each word being w bits.
+    self.S = [[0, 0, 0, 0, 0],
+             [0, 0, 0, 0, 0],
+             [0, 0, 0, 0, 0],
+             [0, 0, 0, 0, 0],
+             [0, 0, 0, 0, 0]]
+
+    # A string of hexchars, where each char represents 4 bits.
+    self.buffered_data = ""
+
+    # Store the calculated digest.
+    # We'll only apply padding and recalculate the hash if it's modified.
+    self.last_digest = None
+
+    if data:
+      self.update(data)
+
+  # Constants
+
+  ## Round constants
+  RC = [0x0000000000000001,
+        0x0000000000008082,
+        0x800000000000808A,
+        0x8000000080008000,
+        0x000000000000808B,
+        0x0000000080000001,
+        0x8000000080008081,
+        0x8000000000008009,
+        0x000000000000008A,
+        0x0000000000000088,
+        0x0000000080008009,
+        0x000000008000000A,
+        0x000000008000808B,
+        0x800000000000008B,
+        0x8000000000008089,
+        0x8000000000008003,
+        0x8000000000008002,
+        0x8000000000000080,
+        0x000000000000800A,
+        0x800000008000000A,
+        0x8000000080008081,
+        0x8000000000008080,
+        0x0000000080000001,
+        0x8000000080008008]
+
+  ## Rotation offsets
+  r = [[0,  36,   3,  41,  18],
+       [1,  44,  10,  45,   2],
+       [62,  6,  43,  15,  61],
+       [28, 55,  25,  21,  56],
+       [27, 20,  39,   8,  14]]
+
+  @staticmethod
+  def Round(A, RCfixed, w):
+    """Perform one round of computation as defined in the Keccak-f permutation
+
+    A: current state (5x5 matrix)
+    RCfixed: value of round constant to use (integer)
+    """
+
+    #Initialization of temporary variables
+    B = [[0, 0, 0, 0, 0],
+         [0, 0, 0, 0, 0],
+         [0, 0, 0, 0, 0],
+         [0, 0, 0, 0, 0],
+         [0, 0, 0, 0, 0]]
+    C = [0, 0, 0, 0, 0]
+    D = [0, 0, 0, 0, 0]
+
+    #Theta step
+    for x in range(5):
+      C[x] = A[x][0] ^ A[x][1] ^ A[x][2] ^ A[x][3] ^ A[x][4]
+
+    for x in range(5):
+      D[x] = C[(x - 1) % 5] ^ _rot(C[(x + 1) % 5], 1, w)
+
+    for x in range(5):
+      for y in range(5):
+        A[x][y] = A[x][y] ^ D[x]
+
+    #Rho and Pi steps
+    for x in range(5):
+      for y in range(5):
+        B[y][(2 * x + 3 * y) % 5] = _rot(A[x][y], Keccak.r[x][y], w)
+
+    #Chi step
+    for x in range(5):
+      for y in range(5):
+        A[x][y] = B[x][y] ^ ((~B[(x + 1) % 5][y]) & B[(x + 2) % 5][y])
+
+    #Iota step
+    A[0][0] = A[0][0] ^ RCfixed
+
+    return A
+
+  @staticmethod
+  def KeccakF(A, n_r, w):
+    """Perform Keccak-f function on the state A
+
+    A: 5x5 matrix containing the state, where each entry is a string of hexchars that is 'w' bits long
+    n_r: number of rounds
+    w: word size
+    """
+
+    for i in xrange(n_r):
+      A = Keccak.Round(A, Keccak.RC[i] % (1 << w), w)
+
+    return A
+
+  ### Padding rule
+  # This is a disgusting piece of code. Clean it.
+  @staticmethod
+  def pad10star1(M, n):
+    """Pad M with the pad10*1 padding rule to reach a length multiple of r bits
+
+    M: message pair (length in bits, string of hex characters ('9AFC...')
+    n: length in bits (must be a multiple of 8)
+    Example: pad10star1([60, 'BA594E0FB9EBBD30'],8) returns 'BA594E0FB9EBBD93'
+    """
+
+    [my_string_length, my_string] = M
+
+    # Check the parameter n
+    if n % 8 != 0:
+      raise KeccakError.KeccakError("n must be a multiple of 8")
+
+    # Check the length of the provided string
+    if len(my_string) % 2 != 0:
+      #Pad with one '0' to reach correct length (don't know test
+      #vectors coding)
+      my_string += '0'
+    if my_string_length > (len(my_string) // 2 * 8):
+      raise KeccakError.KeccakError("the string is too short to contain the number of bits announced")
+
+    nr_bytes_filled = my_string_length // 8
+    nbr_bits_filled = my_string_length % 8
+    l = my_string_length % n
+    if ((n - 8) <= l <= (n - 2)):
+      if (nbr_bits_filled == 0):
+        my_byte = 0
+      else:
+        my_byte = int(my_string[nr_bytes_filled * 2:nr_bytes_filled * 2 + 2], 16)
+      my_byte = (my_byte >> (8 - nbr_bits_filled))
+      my_byte = my_byte + 2 ** (nbr_bits_filled) + 2 ** 7
+      my_byte = "%02X" % my_byte
+      my_string = my_string[0:nr_bytes_filled * 2] + my_byte
+    else:
+      if (nbr_bits_filled == 0):
+        my_byte = 0
+      else:
+        my_byte = int(my_string[nr_bytes_filled * 2:nr_bytes_filled * 2 + 2], 16)
+      my_byte = (my_byte >> (8 - nbr_bits_filled))
+      my_byte = my_byte + 2 ** (nbr_bits_filled)
+      my_byte = "%02X" % my_byte
+      my_string = my_string[0:nr_bytes_filled * 2] + my_byte
+      while((8 * len(my_string) // 2) % n < (n - 8)):
+        my_string = my_string + '00'
+      my_string = my_string + '80'
+
+    return my_string
+
+  def update(self, arg):
+    # Update the hash object with the string arg. Repeated calls are equivalent to a single call with the concatenation of all the arguments: m.update(a); m.update(b) is equivalent to m.update(a+b). arg is a normal bytestring.
+
+    self.last_digest = None
+    # Convert the data into a workable format, and add it to the buffer
+    self.buffered_data += arg.encode('hex')
+
+    # Absorb any blocks we can:
+    if len(self.buffered_data) * 4 >= self.r:
+      extra_bits = len(self.buffered_data) * 4 % self.r
+
+      # An exact fit!
+      if extra_bits == 0:
+        P = self.buffered_data
+        self.buffered_data = ""
+      else:
+        # Slice it up into the first r*a bits, for some constant a>=1, and the remaining total-r*a bits.
+        P = self.buffered_data[:-extra_bits // 4]
+        self.buffered_data = self.buffered_data[-extra_bits // 4:]
+
+      #Absorbing phase
+      for i in xrange((len(P) * 8 // 2) // self.r):
+        to_convert = P[i * (2 * self.r // 8):(i + 1) * (2 * self.r // 8)] + '00' * (self.c // 8)
+        P_i = _convertStrToTable(to_convert, self.w, self.b)
+
+        # First apply the XOR to the state + block
+        for y in xrange(5):
+          for x in xrange(5):
+            self.S[x][y] = self.S[x][y] ^ P_i[x][y]
+        # Then apply the block permutation, Keccak-F
+        self.S = Keccak.KeccakF(self.S, self.n_r, self.w)
+
+  def digest(self):
+    """Return the digest of the strings passed to the update() method so far.
+
+    This is a string of digest_size bytes which may contain non-ASCII
+    characters, including null bytes."""
+
+    if self.last_digest:
+      return self.last_digest
+
+    # UGLY WARNING
+    # Handle bytestring/hexstring conversions
+    M = _build_message_pair(self.buffered_data.decode('hex'))
+
+    # First finish the padding and force the final update:
+    self.buffered_data = Keccak.pad10star1(M, self.r)
+    self.update('')
+    # UGLY WARNING over
+
+    assert len(self.buffered_data) == 0, "Why is there data left in the buffer? %s with length %d" % (self.buffered_data, len(self.buffered_data) * 4)
+
+    # Squeezing time!
+    Z = ''
+    outputLength = self.n
+    while outputLength > 0:
+      string = _convertTableToStr(self.S, self.w)
+      # Read the first 'r' bits of the state
+      Z = Z + string[:self.r * 2 // 8]
+      outputLength -= self.r
+      if outputLength > 0:
+        S = KeccakF(S, verbose)
+
+    self.last_digest = Z[:2 * self.n // 8].decode('hex')
+    return self.last_digest
+
+  def hexdigest(self):
+    """Like digest() except the digest is returned as a string of hex digits
+
+    This may be used to exchange the value safely in email or other
+    non-binary environments."""
+    return self.digest().encode('hex')
+
+  def copy(self):
+    # First initialize whatever can be done normally
+    duplicate = Keccak(c=self.c, r=self.r, n=self.n)
+    # Then copy over the state.
+    for i in xrange(5):
+      for j in xrange(5):
+        duplicate.S[i][j] = self.S[i][j]
+    # and any other stored data
+    duplicate.buffered_data = self.buffered_data
+    duplicate.last_digest = self.last_digest
+    return duplicate
+
+
+## Generic utility functions
+
+def _build_message_pair(data):
+  hex_data = data.encode('hex')
+  size = len(hex_data) * 4
+  return (size, hex_data)
+
+
+def _rot(x, shift_amount, length):
+  """Rotate x shift_amount bits to the left, considering the \
+  string of bits is length bits long"""
+
+  shift_amount = shift_amount % length
+  return ((x >> (length - shift_amount)) + (x << shift_amount)) % (1 << length)
+
+### Conversion functions String <-> Table (and vice-versa)
+
+
+def _fromHexStringToLane(string):
+  """Convert a string of bytes written in hexadecimal to a lane value"""
+
+  #Check that the string has an even number of characters i.e. whole number of bytes
+  if len(string) % 2 != 0:
+    raise KeccakError.KeccakError("The provided string does not end with a full byte")
+
+  #Perform the conversion
+  temp = ''
+  nrBytes = len(string) // 2
+  for i in xrange(nrBytes):
+    offset = (nrBytes - i - 1) * 2
+    temp += string[offset:offset + 2]
+  return int(temp, 16)
+
+
+def _fromLaneToHexString(lane, w):
+  """Convert a lane value to a string of bytes written in hexadecimal"""
+
+  laneHexBE = (("%%0%dX" % (w // 4)) % lane)
+  #Perform the conversion
+  temp = ''
+  nrBytes = len(laneHexBE) // 2
+  for i in xrange(nrBytes):
+    offset = (nrBytes - i - 1) * 2
+    temp += laneHexBE[offset:offset + 2]
+  return temp.upper()
+
+
+def _convertStrToTable(string, w, b):
+  """Convert a string of hex-chars to its 5x5 matrix representation
+
+  string: string of bytes of hex-coded bytes (e.g. '9A2C...')"""
+
+  # Check that the input paramaters are expected
+  if w % 8 != 0:
+    raise KeccakError("w is not a multiple of 8")
+
+  # Each character in the string represents 4 bits.
+  # The string should have exactly 'b' bits.
+  if len(string) * 4 != b:
+    raise KeccakError.KeccakError("string can't be divided in 25 blocks of w bits\
+    i.e. string must have exactly b bits")
+
+  #Convert
+  output = [[0, 0, 0, 0, 0],
+            [0, 0, 0, 0, 0],
+            [0, 0, 0, 0, 0],
+            [0, 0, 0, 0, 0],
+            [0, 0, 0, 0, 0]]
+
+  bits_per_char = 2 * w // 8
+  for x in xrange(5):
+    for y in xrange(5):
+      # Each entry will have b/25=w bits.
+      offset = (5 * y + x) * bits_per_char
+      # Store the data into the associated word.
+      hexstring = string[offset:offset + bits_per_char]
+      output[x][y] = _fromHexStringToLane(hexstring)
+  return output
+
+
+def _convertTableToStr(table, w):
+  """Convert a 5x5 matrix representation to its string representation"""
+
+  #Check input format
+  if w % 8 != 0:
+    raise KeccakError.KeccakError("w is not a multiple of 8")
+  if (len(table) != 5) or any(len(row) != 5 for row in table):
+    raise KeccakError.KeccakError("table must be 5x5")
+
+  #Convert
+  output = [''] * 25
+  for x in xrange(5):
+    for y in xrange(5):
+      output[5 * y + x] = _fromLaneToHexString(table[x][y], w)
+  output = ''.join(output).upper()
+  return output