Re: Complexity comparison: logarithm versus multiplication versus summation

Helmut Sennewald wrote: 
my guess for the complexity ratio is

20:1:1      logarithm:multiply:add

Complexity mainly means execution time or number of cycles.

That's one way to measure complexity. Another would be to look at die size for a multiplier, an adder, and uh... a logarithmer.

Multipliers are huge compared to adders. I dunno /how/ much larger they are, but 20:1 is prolly not /too/ bad a guess. Logs are usually implemented as range reduction followed by a lookup table seeding a polynomial, and the poly is built from multipliers and adders.

And now it's time to retell the old joke:
On the eighth day, God was walking through the Garden of Eden when he saw two snakes trying to chop down a tree.
God: "What are you doing?"
Snake1: "Well, You're the One who told us to go forth and multiply."
God: "What's that got to do with cutting down a tree?"
Snake2: "We're adders! We can't multiply without logs!"
<rim shot>

--
Jim Thomas            Principal Applications Engineer  Bittware, Inc
jthomas@xxxxxxxxxxxx  http://www.bittware.com    (603) 226-0404 x536
When you know how things work, the world is one big sandbox.  - Avins
.

Relevant Pages
Loading