Altera_Forum
Honored Contributor
Honored Contributorfloating point number
Hi All,
How can I convert a std_logic_vector number into a floating point number? What is the command-operator? what library to use? is it synthesizable ? ThanksForum Discussion
9 Replies
Altera_ForumIf you want a synthesisable fp number you will have to use slv and user the floating point cores provided by altera. But why do you want floating point? Usually fire FPGAs it is far more efficient to use fixed point.
- Altera_Forum
Hi,
Do you mean ALTFP_CONVER from the IP library? If yes, it creates this entity, which is different from documentation. for example: what is "distance" input? Doesn't have this input in documentation. ENTITY ALTFP_CONVERa_altbarrel_shift_fof IS PORT ( aclr : IN STD_LOGIC := '0'; clk_en : IN STD_LOGIC := '1'; clock : IN STD_LOGIC := '0'; data : IN STD_LOGIC_VECTOR (31 DOWNTO 0); distance : IN STD_LOGIC_VECTOR (4 DOWNTO 0); result : OUT STD_LOGIC_VECTOR (31 DOWNTO 0) ); END ALTFP_CONVERa_altbarrel_shift_fof; Thanks - Altera_Forum
That is a altbarrel_shift_fof core.Not fp convert.
- Altera_Forum
HI,
I am using version quartus 16.1.0 and this is the vhdl i am getting after generating ALTFP_CONVER. Maybe a quartus bug. anyway looks like it works. - Altera_Forum
Sorry, it doesn't work. I think there is a problem with this IP in quartus 16.1.0. it generates ALTFP_CONVERa_altbarrel_shift_fof instead of ALTFP_CONVER.
- Altera_Forum
If it is generating that odd looking IP, which looks like the ports for a barrel shifter, you should raise a support ticket with altera..
- Altera_Forum
hi
iam doing a project in floting point arithemetics so i want to separate the mantisa exponent and sign bit form an unsigned integer anybody have know how to convert and accces by vhdl code/command - Altera_Forum
If you want an unsigned integer, you will need to use the alt_fp_convert megafunction
- Altera_Forum
Of course you can do the conversion in your own code. Float to integer is relative simple arithmetic, just a variable shift.
