diff --git a/algo_dev.txt b/algo_dev.txt
index 5f3fc9d..cf084ea 100644
--- a/algo_dev.txt
+++ b/algo_dev.txt
@@ -1,3 +1,9 @@
+In order to implement a new algo in a pool you need three things
+   1) the C code to do the actual hashing - which I convert into a native node addon
+   2) the max difficulty (diff1) of the coin used to check if shares/blocks are valid
+   3) the share multiplier (2^16 for example) to determine the hashrate of a miner/entire pool
+
+
 Most coins have diff1 in format like (~uint256(0) >> 20)
 I believe you can determine the hashrate multiplier by subtracting the 20 used in bitshift from the number 32
 so it would be 2^12 for this one. or for (~uint256(0) >> 16) it would be 2^(32 - 16)
@@ -20,6 +26,12 @@ I believe these are the true diff1 values that should be used on pools for each
  bcrypt:        0x00f8ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
 
 
+So remember how that list to add a new algo had 3 items, well the maxcoin devs very unforutentely did something
+that added a fourth step:
+ Step 4) Determine which parts of how work-generation / share-processing / creating-blocks work were arbitrarily
+ changed and implement them....
+
+
 Maxcoin devs, it seems, did not simply implement keccek in the way that litecoin devs implemented scrypt.
 Or how the darkcoin devs implements x11, or how the quarkcoin devs implemented quark. The maxcoin devs
 decided to change several parts of how the the currency works - so that instead of simply changing the block