RE: [sv-ec]E-mail Vote: Closes 12am PST October 26th 2007

From: Steven Sharp <sharp_at_.....>
Date: Mon Oct 22 2007 - 17:31:00 PDT
And my math errors on 2055 continue :-)

floor(value/((2**M)/N)) has to be written as (value*N)/(2**M) using integer
operations, so that takes a multiply and a divide.

And it does not put all the extras in the first bins.  It spreads the
extras more evenly across the bins, only biasing toward the first bins
when it has to.

So, does anyone have an efficient way of computing the bin index that
puts the extras in the first bins or last bins, given the value and N
and M?  It seems to me that it is important that whatever rule is
specified should be something reasonably efficient to compute.

Steven Sharp
sharp@cadence.com


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Mon Oct 22 17:31:33 2007

This archive was generated by hypermail 2.1.8 : Mon Oct 22 2007 - 17:31:46 PDT