Project

General

Profile

LIBRARY ieee;
USE ieee.std_logic_1164.all;
--USE ieee.std_logic_arith.all;
--USE ieee.std_logic_unsigned.all;
USE ieee.numeric_std.all;
use work.pixels.all;

--TODO: scanlines!!
--out_colour(7 downto 4) <= out_colour_raw(7 downto 4);
--out_colour(3 downto 0) <= out_colour_raw(3 downto 0) when (not(out_scanlines) or vcnt(0))='1' else '0'&out_colour_raw(3 downto 1);

ENTITY polyphasicscale IS
PORT(
clock : IN STD_LOGIC; --system clock
reset_n : IN STD_LOGIC; --asynchronous reset

-- input pixels
pixels : in t_Pixel4x4;

-- next output line
next_y_in : in std_logic;
next_frame_in : in std_logic;
field2 : in std_logic;
-- sync/blank from crtc -> need to be delayed in line with pipeline
hsync_in : in std_logic;
vsync_in : in std_logic;
blank_in : in std_logic;
-- need to provide control signals:
next_x : OUT STD_LOGIC;
next_x_size : out std_logic_vector(1 downto 0);
next_y : OUT STD_LOGIC;

-- need to output
-- current pixel
r : out std_logic_vector(7 downto 0);
g : out std_logic_vector(7 downto 0);
b : out std_logic_vector(7 downto 0);
hsync : out std_logic;
vsync : out std_logic;
blank : out std_logic;
-- to set up params
sda : inout std_logic;
scl : inout std_logic
);
END polyphasicscale;

-- customscalex:568 customscaley:256 xdelta:640 ydelta:102 TA:65536
ARCHITECTURE vhdl OF polyphasicscale IS
signal fixed_xthreshold_reg : unsigned(10 downto 0);
signal fixed_ythreshold_reg : unsigned(10 downto 0);
signal param_xdelta_reg : unsigned(10 downto 0);
signal param_ydeltaeach_reg : unsigned(10 downto 0);
signal param_xaddrskip_reg : unsigned(1 downto 0);-- We only support upscaling, so for some cases need to skip 2 pixels
signal params : std_logic_vector((11*3)-1 downto 0);
-- stage 1:5 Delay syncs in line with pipeline
signal hsync_next : std_logic_vector(6 downto 1);
signal hsync_reg : std_logic_vector(6 downto 1);
signal vsync_next : std_logic_vector(6 downto 1);
signal vsync_reg : std_logic_vector(6 downto 1);
signal blank_next : std_logic_vector(6 downto 1);
signal blank_reg : std_logic_vector(6 downto 1);
-- stage 1 : Accumulators/ data request
signal xacc_next : unsigned(11 downto 0);
signal xacc_reg : unsigned(11 downto 0);
signal yacc_next : unsigned(11 downto 0);
signal yacc_reg : unsigned(11 downto 0);
signal next_x_delay1_next : std_logic;
signal next_x_delay1_reg : std_logic;
-- stage 2 : Coefs calc
type t_Coef is record
c1 : signed(8 downto 0);
c2 : signed(8 downto 0);
c3 : signed(8 downto 0);
c4 : signed(8 downto 0);
end record t_Coef;
signal coefx_next: t_Coef;
signal coefx_reg : t_Coef;
signal coefx3_next: t_Coef;
signal coefx3_reg : t_Coef;
signal coefx4_next: t_Coef;
signal coefx4_reg : t_Coef;
signal coefy_next: t_Coef;
signal coefy_reg : t_Coef;
signal next_x_delay2_next : std_logic;
signal next_x_delay2_reg : std_logic;

-- stage 3 : Y mult
signal ry1_next : signed(17 downto 0);
signal ry1_reg : signed(17 downto 0);
signal ry2_next : signed(17 downto 0);
signal ry2_reg : signed(17 downto 0);
signal ry3_next : signed(17 downto 0);
signal ry3_reg : signed(17 downto 0);
signal ry4_next : signed(17 downto 0);
signal ry4_reg : signed(17 downto 0);

signal gy1_next : signed(17 downto 0);
signal gy1_reg : signed(17 downto 0);
signal gy2_next : signed(17 downto 0);
signal gy2_reg : signed(17 downto 0);
signal gy3_next : signed(17 downto 0);
signal gy3_reg : signed(17 downto 0);
signal gy4_next : signed(17 downto 0);
signal gy4_reg : signed(17 downto 0);

signal by1_next : signed(17 downto 0);
signal by1_reg : signed(17 downto 0);
signal by2_next : signed(17 downto 0);
signal by2_reg : signed(17 downto 0);
signal by3_next : signed(17 downto 0);
signal by3_reg : signed(17 downto 0);
signal by4_next : signed(17 downto 0);
signal by4_reg : signed(17 downto 0);
signal next_x_delay3_next : std_logic;
signal next_x_delay3_reg : std_logic;

-- stage 4: X pixels
signal pixel_rx1_next : signed(8 downto 0);
signal pixel_rx1_reg : signed(8 downto 0);
signal pixel_rx2_next : signed(8 downto 0);
signal pixel_rx2_reg : signed(8 downto 0);
signal pixel_rx3_next : signed(8 downto 0);
signal pixel_rx3_reg : signed(8 downto 0);
signal pixel_rx4_next : signed(8 downto 0);
signal pixel_rx4_reg : signed(8 downto 0);

signal pixel_gx1_next : signed(8 downto 0);
signal pixel_gx1_reg : signed(8 downto 0);
signal pixel_gx2_next : signed(8 downto 0);
signal pixel_gx2_reg : signed(8 downto 0);
signal pixel_gx3_next : signed(8 downto 0);
signal pixel_gx3_reg : signed(8 downto 0);
signal pixel_gx4_next : signed(8 downto 0);
signal pixel_gx4_reg : signed(8 downto 0);

signal pixel_bx1_next : signed(8 downto 0);
signal pixel_bx1_reg : signed(8 downto 0);
signal pixel_bx2_next : signed(8 downto 0);
signal pixel_bx2_reg : signed(8 downto 0);
signal pixel_bx3_next : signed(8 downto 0);
signal pixel_bx3_reg : signed(8 downto 0);
signal pixel_bx4_next : signed(8 downto 0);
signal pixel_bx4_reg : signed(8 downto 0);

-- stage 5: X mult
signal rx1_next : signed(17 downto 0);
signal rx1_reg : signed(17 downto 0);
signal rx2_next : signed(17 downto 0);
signal rx2_reg : signed(17 downto 0);
signal rx3_next : signed(17 downto 0);
signal rx3_reg : signed(17 downto 0);
signal rx4_next : signed(17 downto 0);
signal rx4_reg : signed(17 downto 0);

signal gx1_next : signed(17 downto 0);
signal gx1_reg : signed(17 downto 0);
signal gx2_next : signed(17 downto 0);
signal gx2_reg : signed(17 downto 0);
signal gx3_next : signed(17 downto 0);
signal gx3_reg : signed(17 downto 0);
signal gx4_next : signed(17 downto 0);
signal gx4_reg : signed(17 downto 0);

signal bx1_next : signed(17 downto 0);
signal bx1_reg : signed(17 downto 0);
signal bx2_next : signed(17 downto 0);
signal bx2_reg : signed(17 downto 0);
signal bx3_next : signed(17 downto 0);
signal bx3_reg : signed(17 downto 0);
signal bx4_next : signed(17 downto 0);
signal bx4_reg : signed(17 downto 0);

-- stage 6: Pixel output (sum)
signal r_reg : std_logic_vector(7 downto 0);
signal r_next : std_logic_vector(7 downto 0);
signal g_reg : std_logic_vector(7 downto 0);
signal g_next : std_logic_vector(7 downto 0);
signal b_reg : std_logic_vector(7 downto 0);
signal b_next : std_logic_vector(7 downto 0);

BEGIN
process(clock,reset_n)
begin
if (reset_n='0') then
-- multi stage 1-5
hsync_reg <= (others=>'0');
vsync_reg <= (others=>'0');
blank_reg <= (others=>'0');

-- stage 1
xacc_reg <= (others=>'0');
yacc_reg <= (others=>'0');
next_x_delay1_reg <= '0';

-- stage 2
coefx_reg.c1 <= (others=>'0');
coefx_reg.c2 <= (others=>'0');
coefx_reg.c3 <= (others=>'0');
coefx_reg.c4 <= (others=>'0');
coefy_reg.c1 <= (others=>'0');
coefy_reg.c2 <= (others=>'0');
coefy_reg.c3 <= (others=>'0');
coefy_reg.c4 <= (others=>'0');
next_x_delay2_reg <= '0';

-- stage 3
ry1_reg <= (others=>'0');
ry2_reg <= (others=>'0');
ry3_reg <= (others=>'0');
ry4_reg <= (others=>'0');

gy1_reg <= (others=>'0');
gy2_reg <= (others=>'0');
gy3_reg <= (others=>'0');
gy4_reg <= (others=>'0');

by1_reg <= (others=>'0');
by2_reg <= (others=>'0');
by3_reg <= (others=>'0');
by4_reg <= (others=>'0');
coefx3_reg.c1 <= (others=>'0');
coefx3_reg.c2 <= (others=>'0');
coefx3_reg.c3 <= (others=>'0');
coefx3_reg.c4 <= (others=>'0');
next_x_delay3_reg <= '0';

-- stage 4
pixel_rx1_reg <= (others=>'0');
pixel_gx1_reg <= (others=>'0');
pixel_bx1_reg <= (others=>'0');

pixel_rx2_reg <= (others=>'0');
pixel_gx2_reg <= (others=>'0');
pixel_bx2_reg <= (others=>'0');

pixel_rx3_reg <= (others=>'0');
pixel_gx3_reg <= (others=>'0');
pixel_bx3_reg <= (others=>'0');

pixel_rx4_reg <= (others=>'0');
pixel_gx4_reg <= (others=>'0');
pixel_bx4_reg <= (others=>'0');
coefx4_reg.c1 <= (others=>'0');
coefx4_reg.c2 <= (others=>'0');
coefx4_reg.c3 <= (others=>'0');
coefx4_reg.c4 <= (others=>'0');

-- stage 5
rx1_reg <= (others=>'0');
rx2_reg <= (others=>'0');
rx3_reg <= (others=>'0');
rx4_reg <= (others=>'0');

gx1_reg <= (others=>'0');
gx2_reg <= (others=>'0');
gx3_reg <= (others=>'0');
gx4_reg <= (others=>'0');

bx1_reg <= (others=>'0');
bx2_reg <= (others=>'0');
bx3_reg <= (others=>'0');
bx4_reg <= (others=>'0');

-- stage 6
r_reg <= (others=>'0');
g_reg <= (others=>'0');
b_reg <= (others=>'0');
elsif (clock'event and clock='1') then
-- multi stage 1-5
hsync_reg <= hsync_next;
vsync_reg <= vsync_next;
blank_reg <= blank_next;
-- stage 1
xacc_reg <= xacc_next;
yacc_reg <= yacc_next;
next_x_delay1_reg <= next_x_delay1_next;

-- stage 2
coefx_reg <= coefx_next;
coefy_reg <= coefy_next;
next_x_delay2_reg <= next_x_delay2_next;

-- stage 3
ry1_reg <= ry1_next;
ry2_reg <= ry2_next;
ry3_reg <= ry3_next;
ry4_reg <= ry4_next;

gy1_reg <= gy1_next;
gy2_reg <= gy2_next;
gy3_reg <= gy3_next;
gy4_reg <= gy4_next;

by1_reg <= by1_next;
by2_reg <= by2_next;
by3_reg <= by3_next;
by4_reg <= by4_next;
coefx3_reg <= coefx3_next;
next_x_delay3_reg <= next_x_delay3_next;

-- stage 4
pixel_rx1_reg <= pixel_rx1_next;
pixel_gx1_reg <= pixel_gx1_next;
pixel_bx1_reg <= pixel_bx1_next;

pixel_rx2_reg <= pixel_rx2_next;
pixel_gx2_reg <= pixel_gx2_next;
pixel_bx2_reg <= pixel_bx2_next;

pixel_rx3_reg <= pixel_rx3_next;
pixel_gx3_reg <= pixel_gx3_next;
pixel_bx3_reg <= pixel_bx3_next;

pixel_rx4_reg <= pixel_rx4_next;
pixel_gx4_reg <= pixel_gx4_next;
pixel_bx4_reg <= pixel_bx4_next;
coefx4_reg <= coefx4_next;

-- stage 5
rx1_reg <= rx1_next;
rx2_reg <= rx2_next;
rx3_reg <= rx3_next;
rx4_reg <= rx4_next;

gx1_reg <= gx1_next;
gx2_reg <= gx2_next;
gx3_reg <= gx3_next;
gx4_reg <= gx4_next;

bx1_reg <= bx1_next;
bx2_reg <= bx2_next;
bx3_reg <= bx3_next;
bx4_reg <= bx4_next;

-- stage 6
r_reg <= r_next;
g_reg <= g_next;
b_reg <= b_next;
end if;
end process;
-- temporary params, until i2c implemented
-- 720p
-- param_xthreshold_next <= to_unsigned(699,10); -- these need to be inputs
-- param_xdelta_next <= to_unsigned(393,10);
-- param_ythreshold_next <= to_unsigned(400,9);
-- param_ydelta_next <= to_unsigned(166,9);
-- param_ydeltaeach_next <= to_unsigned(166,9);
-- param_xaddrskip_next <= "10";

-- 1080i
-- param_xthreshold_next <= to_unsigned(400,11); -- these need to be inputs
-- param_xdelta_next <= to_unsigned(300,11);
-- param_ythreshold_next <= to_unsigned(786,11);
-- param_ydelta_next <= to_unsigned(218,11);
-- param_ydeltaeach_next <= to_unsigned(437,11); --interlace, need to skip a line
-- param_xaddrskip_next <= "01";


-- param_xthreshold_next <= to_unsigned(1024,11); -- these need to be inputs
-- param_xdelta_next <= to_unsigned(768,11);
-- --param_xdelta_next <= to_unsigned(64,11);
-- param_ythreshold_next <= to_unsigned(1024,11);
-- --param_ydelta_next <= to_unsigned(284,11);
-- param_ydeltaeach_next <= to_unsigned(569,11); --interlace, need to skip a line
-- param_xaddrskip_next <= "01";
-- params set from i2c
i2cregs : entity work.I2C_regs
generic map (
SLAVE_ADDR => "0000011", --will we activate this and areascale at once? i.e. do I give it another address?
regs => 3,
bits => 11
)
port map (
scl => scl,
sda => sda,
clk => clock,
rst => not(reset_n),
reg => params
);
fixed_xthreshold_reg <= to_unsigned(1024,11);
fixed_ythreshold_reg <= to_unsigned(1024,11);
param_xdelta_reg <= unsigned(params(11*1-1 downto 11*0));
param_ydeltaeach_reg <= unsigned(params(11*2-1 downto 11*1));
param_xaddrskip_reg <= unsigned(params(11*2+2-1 downto 11*2));
-- TODO: how about filter params too?
-- Delay sync in line with pipeline: multi stage 1-5
process(hsync_reg,vsync_reg,blank_reg,hsync_in,vsync_in,blank_in)
begin
hsync_next <= hsync_reg(5 downto 1)&hsync_in;
vsync_next <= vsync_reg(5 downto 1)&vsync_in;
blank_next <= blank_reg(5 downto 1)&blank_in;
end process;

-- Compute accumulator - pipeline stage 1
process(xacc_reg,param_xdelta_reg,fixed_xthreshold_reg,next_y_in)
begin
next_x <= '0';
next_x_delay1_next <= '0';
xacc_next <= xacc_reg+param_xdelta_reg;
if ((xacc_reg+param_xdelta_reg)>=fixed_xthreshold_reg) then
next_x <= '1'; -- when I ask for data, its on p1 in 2 cycles, i.e. pipeline stage 3
next_x_delay1_next <= '1';
xacc_next <= xacc_reg+param_xdelta_reg-fixed_xthreshold_reg;
end if;

if (next_y_in ='1') then
xacc_next <= (others=>'0');
end if;
end process;

process(yacc_reg,param_ydeltaeach_reg,fixed_ythreshold_reg,next_y_in,next_frame_in,field2)
begin
yacc_next <= yacc_reg;
next_y <= '0';

if (next_y_in='1') then
yacc_next <= yacc_reg+param_ydeltaeach_reg;
if ((yacc_reg+param_ydeltaeach_reg)>=fixed_ythreshold_reg) then
next_y <= '1';
yacc_next <= yacc_reg+param_ydeltaeach_reg-fixed_ythreshold_reg;
end if;
end if;

if (next_frame_in='1') then
yacc_next <= (others=>'0');
if (field2='1') then --slightly lower on interlace field 2
yacc_next(9 downto 0) <= fixed_ythreshold_reg(10 downto 1);
end if;
end if;
end process;

-- Coef computation - pipeline stage 2
process(
xacc_reg,
yacc_reg)
variable xphase : std_logic_vector(3 downto 0);
variable yphase : std_logic_vector(3 downto 0);
begin
coefx_next.c1 <= (others=>'0');
coefx_next.c2 <= (others=>'0');
coefx_next.c3 <= (others=>'0');
coefx_next.c4 <= (others=>'0');
coefy_next.c1 <= (others=>'0');
coefy_next.c2 <= (others=>'0');
coefy_next.c3 <= (others=>'0');
coefy_next.c4 <= (others=>'0');

xphase := std_logic_vector(xacc_reg(9 downto 6));
yphase := std_logic_vector(yacc_reg(9 downto 6));

-- case xphase is
-- when "0000" =>
-- coefx_next <= x"7f000000";
-- when "0001" =>
-- coefx_next <= x"7f000000";
-- when "0010" =>
-- coefx_next <= x"7f000000";
-- when "0011" =>
-- coefx_next <= x"7f000000";
-- when "0100" =>
-- coefx_next <= x"7f000000";
-- when "0101" =>
-- coefx_next <= x"7f000000";
-- when "0110" =>
-- coefx_next <= x"7f000000";
-- when "0111" =>
-- coefx_next <= x"7f000000";
-- when "1000" =>
-- coefx_next <= x"7f000000";
-- when "1001" =>
-- coefx_next <= x"7f000000";
-- when "1010" =>
-- coefx_next <= x"7f000000";
-- when "1011" =>
-- coefx_next <= x"7f000000";
-- when "1100" =>
-- coefx_next <= x"7f000000";
-- when "1101" =>
-- coefx_next <= x"7f000000";
-- when "1110" =>
-- coefx_next <= x"7f000000";
-- when "1111" =>
-- coefx_next <= x"7f000000";
-- when others=>
-- coefx_next <= (others=>'0');
-- end case;
--
-- case yphase is
-- when "0000" =>
-- coefy_next <= x"7f000000";
-- when "0001" =>
-- coefy_next <= x"7f000000";
-- when "0010" =>
-- coefy_next <= x"7f000000";
-- when "0011" =>
-- coefy_next <= x"7f000000";
-- when "0100" =>
-- coefy_next <= x"7f000000";
-- when "0101" =>
-- coefy_next <= x"7f000000";
-- when "0110" =>
-- coefy_next <= x"7f000000";
-- when "0111" =>
-- coefy_next <= x"7f000000";
-- when "1000" =>
-- coefy_next <= x"7f000000";
-- when "1001" =>
-- coefy_next <= x"7f000000";
-- when "1010" =>
-- coefy_next <= x"7f000000";
-- when "1011" =>
-- coefy_next <= x"7f000000";
-- when "1100" =>
-- coefy_next <= x"7f000000";
-- when "1101" =>
-- coefy_next <= x"7f000000";
-- when "1110" =>
-- coefy_next <= x"7f000000";
-- when "1111" =>
-- coefy_next <= x"7f000000";
-- when others=>
-- coefy_next <= (others=>'0');
-- end case;
case xphase is
when "0000" =>
coefx_next.c1 <= to_signed(-8,9);
coefx_next.c2 <= to_signed(147,9);
coefx_next.c3 <= to_signed(-8,9);
coefx_next.c4 <= to_signed(1,9);
when "0001" =>
coefx_next.c1 <= to_signed(-6,9);
coefx_next.c2 <= to_signed(148,9);
coefx_next.c3 <= to_signed(-11,9);
coefx_next.c4 <= to_signed(1,9);
when "0010" =>
coefx_next.c1 <= to_signed(-3,9);
coefx_next.c2 <= to_signed(147,9);
coefx_next.c3 <= to_signed(-13,9);
coefx_next.c4 <= to_signed(1,9);
when "0011" =>
coefx_next.c1 <= to_signed(-1,9);
coefx_next.c2 <= to_signed(145,9);
coefx_next.c3 <= to_signed(-13,9);
coefx_next.c4 <= to_signed(1,9);
when "0100" =>
coefx_next.c1 <= to_signed(0,9);
coefx_next.c2 <= to_signed(141,9);
coefx_next.c3 <= to_signed(-10,9);
coefx_next.c4 <= to_signed(1,9);
when "0101" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(133,9);
coefx_next.c3 <= to_signed(-3,9);
coefx_next.c4 <= to_signed(1,9);
when "0110" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(118,9);
coefx_next.c3 <= to_signed(12,9);
coefx_next.c4 <= to_signed(1,9);
when "0111" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(95,9);
coefx_next.c3 <= to_signed(35,9);
coefx_next.c4 <= to_signed(1,9);
when "1000" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(65,9);
coefx_next.c3 <= to_signed(65,9);
coefx_next.c4 <= to_signed(1,9);
when "1001" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(35,9);
coefx_next.c3 <= to_signed(95,9);
coefx_next.c4 <= to_signed(1,9);
when "1010" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(12,9);
coefx_next.c3 <= to_signed(118,9);
coefx_next.c4 <= to_signed(1,9);
when "1011" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(-3,9);
coefx_next.c3 <= to_signed(133,9);
coefx_next.c4 <= to_signed(1,9);
when "1100" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(-10,9);
coefx_next.c3 <= to_signed(141,9);
coefx_next.c4 <= to_signed(0,9);
when "1101" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(-13,9);
coefx_next.c3 <= to_signed(145,9);
coefx_next.c4 <= to_signed(-1,9);
when "1110" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(-13,9);
coefx_next.c3 <= to_signed(147,9);
coefx_next.c4 <= to_signed(-3,9);
when "1111" =>
coefx_next.c1 <= to_signed(1,9);
coefx_next.c2 <= to_signed(-11,9);
coefx_next.c3 <= to_signed(148,9);
coefx_next.c4 <= to_signed(-6,9);
when others =>
coefx_next.c1 <= (others=>'0');
coefx_next.c2 <= (others=>'0');
coefx_next.c3 <= (others=>'0');
coefx_next.c4 <= (others=>'0');
end case;
case yphase is
when "0000" =>
coefy_next.c1 <= to_signed(-8,9);
coefy_next.c2 <= to_signed(147,9);
coefy_next.c3 <= to_signed(-8,9);
coefy_next.c4 <= to_signed(1,9);
when "0001" =>
coefy_next.c1 <= to_signed(-6,9);
coefy_next.c2 <= to_signed(148,9);
coefy_next.c3 <= to_signed(-11,9);
coefy_next.c4 <= to_signed(1,9);
when "0010" =>
coefy_next.c1 <= to_signed(-3,9);
coefy_next.c2 <= to_signed(147,9);
coefy_next.c3 <= to_signed(-13,9);
coefy_next.c4 <= to_signed(1,9);
when "0011" =>
coefy_next.c1 <= to_signed(-1,9);
coefy_next.c2 <= to_signed(145,9);
coefy_next.c3 <= to_signed(-13,9);
coefy_next.c4 <= to_signed(1,9);
when "0100" =>
coefy_next.c1 <= to_signed(0,9);
coefy_next.c2 <= to_signed(141,9);
coefy_next.c3 <= to_signed(-10,9);
coefy_next.c4 <= to_signed(1,9);
when "0101" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(133,9);
coefy_next.c3 <= to_signed(-3,9);
coefy_next.c4 <= to_signed(1,9);
when "0110" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(118,9);
coefy_next.c3 <= to_signed(12,9);
coefy_next.c4 <= to_signed(1,9);
when "0111" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(95,9);
coefy_next.c3 <= to_signed(35,9);
coefy_next.c4 <= to_signed(1,9);
when "1000" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(65,9);
coefy_next.c3 <= to_signed(65,9);
coefy_next.c4 <= to_signed(1,9);
when "1001" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(35,9);
coefy_next.c3 <= to_signed(95,9);
coefy_next.c4 <= to_signed(1,9);
when "1010" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(12,9);
coefy_next.c3 <= to_signed(118,9);
coefy_next.c4 <= to_signed(1,9);
when "1011" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(-3,9);
coefy_next.c3 <= to_signed(133,9);
coefy_next.c4 <= to_signed(1,9);
when "1100" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(-10,9);
coefy_next.c3 <= to_signed(141,9);
coefy_next.c4 <= to_signed(0,9);
when "1101" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(-13,9);
coefy_next.c3 <= to_signed(145,9);
coefy_next.c4 <= to_signed(-1,9);
when "1110" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(-13,9);
coefy_next.c3 <= to_signed(147,9);
coefy_next.c4 <= to_signed(-3,9);
when "1111" =>
coefy_next.c1 <= to_signed(1,9);
coefy_next.c2 <= to_signed(-11,9);
coefy_next.c3 <= to_signed(148,9);
coefy_next.c4 <= to_signed(-6,9);
when others =>
coefy_next.c1 <= (others=>'0');
coefy_next.c2 <= (others=>'0');
coefy_next.c3 <= (others=>'0');
coefy_next.c4 <= (others=>'0');
end case;


end process;

next_x_delay2_next <= next_x_delay1_reg;
-- YMult - pipeline stage 3
process(pixels, coefy_reg)
variable p1 : t_pixel;
variable p2 : t_pixel;
variable p3 : t_pixel;
variable p4 : t_pixel;
begin
p1 := pixels(pixel4x4_idx(0,0));
p2 := pixels(pixel4x4_idx(0,1));
p3 := pixels(pixel4x4_idx(0,2));
p4 := pixels(pixel4x4_idx(0,3));

-- dsp blocks 9x9 multipliers
-- 3 per block -> 4 blocks
ry1_next <= signed('0'&p1.red)*coefy_reg.c1;
ry2_next <= signed('0'&p2.red)*coefy_reg.c2;
ry3_next <= signed('0'&p3.red)*coefy_reg.c3;
ry4_next <= signed('0'&p4.red)*coefy_reg.c4;

gy1_next <= signed('0'&p1.green)*coefy_reg.c1;
gy2_next <= signed('0'&p2.green)*coefy_reg.c2;
gy3_next <= signed('0'&p3.green)*coefy_reg.c3;
gy4_next <= signed('0'&p4.green)*coefy_reg.c4;

by1_next <= signed('0'&p1.blue)*coefy_reg.c1;
by2_next <= signed('0'&p2.blue)*coefy_reg.c2;
by3_next <= signed('0'&p3.blue)*coefy_reg.c3;
by4_next <= signed('0'&p4.blue)*coefy_reg.c4;
end process;
next_x_delay3_next <= next_x_delay2_reg;
coefx3_next <= coefx_reg;

-- X pixels - pipeline stage 4
process(
pixel_rx1_reg,pixel_rx2_reg,pixel_rx3_reg,pixel_rx4_reg,
pixel_gx1_reg,pixel_gx2_reg,pixel_gx3_reg,pixel_gx4_reg,
pixel_bx1_reg,pixel_bx2_reg,pixel_bx3_reg,pixel_bx4_reg,
ry1_reg,ry2_reg,ry3_reg,ry4_reg,
gy1_reg,gy2_reg,gy3_reg,gy4_reg,
by1_reg,by2_reg,by3_reg,by4_reg,
next_x_delay3_reg
)
variable ry_sum : signed (17 downto 0);
variable gy_sum : signed (17 downto 0);
variable by_sum : signed (17 downto 0);
begin
pixel_rx1_next <= pixel_rx1_reg;
pixel_rx2_next <= pixel_rx2_reg;
pixel_rx3_next <= pixel_rx3_reg;
pixel_rx4_next <= pixel_rx4_reg;

pixel_gx1_next <= pixel_gx1_reg;
pixel_gx2_next <= pixel_gx2_reg;
pixel_gx3_next <= pixel_gx3_reg;
pixel_gx4_next <= pixel_gx4_reg;

pixel_bx1_next <= pixel_bx1_reg;
pixel_bx2_next <= pixel_bx2_reg;
pixel_bx3_next <= pixel_bx3_reg;
pixel_bx4_next <= pixel_bx4_reg;

if (next_x_delay3_reg='1') then
-- ry_sum := ry1_reg+ry2_reg+ry3_reg+ry4_reg;
-- pixel_rx1_next <= ry_sum(17)&ry_sum(15 downto 8);
-- pixel_rx2_next <= pixel_rx1_reg;
-- pixel_rx3_next <= pixel_rx2_reg;
-- pixel_rx4_next <= pixel_rx3_reg;
--
-- gy_sum := gy1_reg+gy2_reg+gy3_reg+gy4_reg;
-- pixel_gx1_next <= gy_sum(17)&gy_sum(15 downto 8);
-- pixel_gx2_next <= pixel_gx1_reg;
-- pixel_gx3_next <= pixel_gx2_reg;
-- pixel_gx4_next <= pixel_gx3_reg;
--
-- by_sum := by1_reg+by2_reg+by3_reg+by4_reg;
-- pixel_bx1_next <= by_sum(17)&by_sum(15 downto 8);
-- pixel_bx2_next <= pixel_bx1_reg;
-- pixel_bx3_next <= pixel_bx2_reg;
-- pixel_bx4_next <= pixel_bx3_reg;
ry_sum := ry1_reg+ry2_reg+ry3_reg+ry4_reg;
pixel_rx4_next <= ry_sum(17)&ry_sum(15 downto 8);
pixel_rx3_next <= pixel_rx4_reg;
pixel_rx2_next <= pixel_rx3_reg;
pixel_rx1_next <= pixel_rx2_reg;

gy_sum := gy1_reg+gy2_reg+gy3_reg+gy4_reg;
pixel_gx4_next <= gy_sum(17)&gy_sum(15 downto 8);
pixel_gx3_next <= pixel_gx4_reg;
pixel_gx2_next <= pixel_gx3_reg;
pixel_gx1_next <= pixel_gx2_reg;

by_sum := by1_reg+by2_reg+by3_reg+by4_reg;
pixel_bx4_next <= by_sum(17)&by_sum(15 downto 8);
pixel_bx3_next <= pixel_bx4_reg;
pixel_bx2_next <= pixel_bx3_reg;
pixel_bx1_next <= pixel_bx2_reg;
end if;
end process;
coefx4_next <= coefx3_reg;

-- X mult - pipeline stage 5
process(
pixel_rx1_reg,pixel_rx2_reg,pixel_rx3_reg,pixel_rx4_reg,
pixel_gx1_reg,pixel_gx2_reg,pixel_gx3_reg,pixel_gx4_reg,
pixel_bx1_reg,pixel_bx2_reg,pixel_bx3_reg,pixel_bx4_reg,
coefx4_reg)
begin
-- dsp blocks 9x9 multipliers
-- 3 per block -> 4 blocks
rx1_next <= pixel_rx1_reg*coefx4_reg.c1;
rx2_next <= pixel_rx2_reg*coefx4_reg.c2;
rx3_next <= pixel_rx3_reg*coefx4_reg.c3;
rx4_next <= pixel_rx4_reg*coefx4_reg.c4;

gx1_next <= pixel_gx1_reg*coefx4_reg.c1;
gx2_next <= pixel_gx2_reg*coefx4_reg.c2;
gx3_next <= pixel_gx3_reg*coefx4_reg.c3;
gx4_next <= pixel_gx4_reg*coefx4_reg.c4;

bx1_next <= pixel_bx1_reg*coefx4_reg.c1;
bx2_next <= pixel_bx2_reg*coefx4_reg.c2;
bx3_next <= pixel_bx3_reg*coefx4_reg.c3;
bx4_next <= pixel_bx4_reg*coefx4_reg.c4;
end process;


-- Pixel scaling computation - pipeline stage 6
-- Compute output
-- (p1*A1 + p2*A2 + p3*A3 + p4*A4)/65536
process(
rx1_reg,rx2_reg,rx3_reg,rx4_reg,
gx1_reg,gx2_reg,gx3_reg,gx4_reg,
bx1_reg,bx2_reg,bx3_reg,bx4_reg)
variable sum_r : signed(17 downto 0);
variable sum_g : signed(17 downto 0);
variable sum_b : signed(17 downto 0);
begin
sum_r := (rx1_reg+rx2_reg)+(rx3_reg+rx4_reg);
sum_g := (gx1_reg+gx2_reg)+(gx3_reg+gx4_reg);
sum_b := (bx1_reg+bx2_reg)+(bx3_reg+bx4_reg);

if (sum_r(17)='1') then
sum_r := (others=>'0');
elsif (sum_r(15) or sum_r(14))='1' then
sum_r := to_signed(16383,18);
end if;
if (sum_g(17)='1') then
sum_g := (others=>'0');
elsif (sum_g(15) or sum_g(14))='1' then
sum_g := to_signed(16383,18);
end if;
if (sum_b(17)='1') then
sum_b := (others=>'0');
elsif (sum_b(15) or sum_b(14))='1' then
sum_b := to_signed(16383,18);
end if;
r_next <= std_logic_vector(sum_r(13 downto 6));
g_next <= std_logic_vector(sum_g(13 downto 6));
b_next <= std_logic_vector(sum_b(13 downto 6));
end process;

-- outputnext_x_size
r<=r_reg;
g<=g_reg;
b<=b_reg;
hsync<=hsync_reg(6); -- TODO: why 6 and not 5?
vsync<=vsync_reg(6);
blank<=blank_reg(6);
next_x_size <= std_logic_vector(param_xaddrskip_reg);

END vhdl;

-- h_pixels_across <= 720 - 1;
-- h_sync_on <= 732 - 1;
-- h_sync_off <= 795 - 1;
-- h_end_count <= 864 - 1;
-- v_pixels_down <= 576 - 1;
-- v_sync_on <= 581 - 1;
-- v_sync_off <= 586 - 1;
-- v_end_count <= 625 - 1;

-- --PAL
-- h_pixels_across <= 1280 - 1;
-- h_sync_on <= 1720 - 1;
-- h_sync_off <= 1760 - 1;
-- h_end_count <= 1980 - 1;
-- v_pixels_down <= 720 - 1;
-- v_sync_on <= 725 - 1;
-- v_sync_off <= 730 - 1;
-- v_end_count <= 750 - 1;



(5-5/6)