On 9/6/06, eks dev <[EMAIL PROTECTED]> wrote:
still, DenseBitsMatcher (BitSetIterator warpped in Matcher) works faster than anything else for this case:int skip(Matcher m) throws IOException{ int doc=-1, ret = 0; while(m.skipTo(doc+1)){ doc = m.doc(); ret+=doc; } return ret; }
Huh... that doesn't make much sense. Maybe it's your pentium-M with it's shorter pipeline and less penalty for a pipeline flush on a branch prediction miss. You could try substituting BitUtil.ntz2 for ntz in OpenBitSet.nextSetBit... it uses a full binary search down to the byte level, and then an array lookup. The nice thing about being *open* is that you can write more than one type of iterator if you really want to:-) -Yonik --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
