Rev |
Log message |
Author |
Age |
Path |
106 |
Optimizing Adder.... at a very slow pace... Got to rush |
jguarin2002 |
4772d 05h |
/raytrac |
105 |
Control... |
jguarin2002 |
4783d 04h |
/raytrac |
104 |
Square Root and inversion support |
jguarin2002 |
4788d 14h |
/raytrac |
103 |
Square Root Ok |
jguarin2002 |
4788d 21h |
/raytrac |
102 |
FP additions done |
jguarin2002 |
4788d 21h |
/raytrac |
101 |
theres a new command to offset the first address in the file |
jguarin2002 |
4790d 11h |
/raytrac |
100 |
Additions Ended |
jguarin2002 |
4790d 11h |
/raytrac |
99 |
Square Root... comin sun.... |
jguarin2002 |
4790d 11h |
/raytrac |
98 |
Addition Mistery.. Solved |
jguarin2002 |
4792d 03h |
/raytrac |
97 |
2 floating points operands multiplication done and optimized, correction make on result sign when substracting |
jguarin2002 |
4794d 05h |
/raytrac |
96 |
2 floating points operands multiplication done and optimized |
jguarin2002 |
4795d 15h |
/raytrac |
95 |
2 floating points operands addition done and optimized |
jguarin2002 |
4795d 15h |
/raytrac |
94 |
Optimization on the functionality of the mantissa multiplier... the results are much more precise than last revision... |
jguarin2002 |
4797d 11h |
/raytrac |
93 |
New Fpbranch Directory Distribution |
jguarin2002 |
4797d 21h |
/raytrac |
92 |
Optimizations made on ema2 and ema3: First One, ema3 and ema2 both now has a 4 pipeline, this equalization has saved lots of registers. Second One: Shifters to denormalize mantissas before addition are implemented with multipliers and inverting the bits direction in the bit vectors, resulting in a saving of resources of 332->253 LEs in ema2 (23% saving) and 608->408 LEs in ema3svn status (32% saving) |
jguarin2002 |
4800d 09h |
/raytrac |
91 |
Optimizations made on ema2 and ema3: First One, ema3 and ema2 both now has a 4 pipeline, this equalization has saved lots of registers. Second One: Shifters to denormalize mantissas before addition are implemented with multipliers and inverting the bits direction in the bit vectors, resulting in a saving of resources of 332->253 LEs in ema2 (23% saving) and 608->408 LEs in ema3svn status (32% saving) |
jguarin2002 |
4800d 09h |
/raytrac |
90 |
Minor corrections to ema2 zero support |
jguarin2002 |
4800d 16h |
/raytrac |
89 |
floating point mult ready |
jguarin2002 |
4800d 17h |
/raytrac |
88 |
Exponent Managment Addition: supports now Zero |
jguarin2002 |
4800d 18h |
/raytrac |
87 |
For the sake of easyness, mmp is called now mul2 |
jguarin2002 |
4802d 18h |
/raytrac |
86 |
For the sake of easyness, mmp is called now mul2 |
jguarin2002 |
4802d 18h |
/raytrac |
85 |
get doesnt fit anymore... |
jguarin2002 |
4803d 10h |
/raytrac |
84 |
Floating Point Adders: (ema3,add3) is an a+b+c adder and (ema2,add2) is an a+b adder. Ema's are blocks to normalize the exponents and shift the mantissas as required, add's are blocks that make the mantissa addition and normalize the result again into a 32 bit IEEE float format |
jguarin2002 |
4803d 10h |
/raytrac |
83 |
Only to have it as an example of what to do to sign a mantissa |
jguarin2002 |
4803d 10h |
/raytrac |
82 |
FPBRANCH releaseeeesvn add fpbranch/get.vhd fpbranch/sm.vhd fpbranch/slr.vhd fpbranch/mmp.vhd svn add fpbranch/get.vhd fpbranch/sm.vhd fpbranch/slr.vhd fpbranch/mmp.vhd svn add fpbranch/get.vhd fpbranch/sm.vhd fpbranch/slr.vhd fpbranch/mmp.vhd |
jguarin2002 |
4806d 05h |
/raytrac |
81 |
Almost There |
jguarin2002 |
4810d 03h |
/raytrac |
80 |
not a revision |
jguarin2002 |
4810d 19h |
/raytrac |
79 |
I can't use a 20 bits mantissa.... eliminating optimization |
jguarin2002 |
4814d 13h |
/raytrac |
78 |
Metamorphosis |
jguarin2002 |
4822d 02h |
/raytrac |
77 |
Now support for addition and substraction of A(7,10) component vectors |
jguarin2002 |
4822d 15h |
/raytrac |