65536 與 32

一個是2^16一個是2^5，就這樣，但是在一個很簡單的SRAM程式碼(實作非常非常簡單的記憶體)與MAX PLUS II卻有不同的結果

LIBRARY ieee;
USE ieee.std_logic_1164.all;
use ieee.std_logic_unsigned.all;
use ieee.std_logic_arith.all;

ENTITY SRAM IS
generic(S_size: integer:= 65536;
        S_address: integer:= 16;
        S_length: integer:= 16);
PORT(enable: in std_logic;
     M_rw: in std_logic;
     address: in std_logic_vector(S_address-1 downto 0);
     in_data: in std_logic_vector(S_length-1 downto 0);
     out_data: out std_logic_vector(S_length-1 downto 0));
END SRAM;

ARCHITECTURE beheavor OF SRAM IS
    type memory_array is array(0 to S_size-1) of std_logic_vector(S_length-1 downto 0);
    signal sram_unit: memory_array;
BEGIN
    process(enable)
    begin
        if(enable = '1') then
            if(M_rw = '1') then
                out_data <= sram_unit(conv_integer(address));
            else
                sram_unit(conv_integer(address)) <= in_data;
            end if;
        end if;
    end process;
END beheavor;

其中

generic(S_size: integer:= 65536;
        S_address: integer:= 16;
        S_length: integer:= 16);

這個在MAX PLUS II我放了將近兩個小時complier還是沒跑完，後來將S_size改成32，S_address改成5，才能夠有點lag的complie過。

---
實驗講義寫的2^16 * 16bit 是怎麼回事XD