Informix 5, greater thans, and cpu usage

I am trying to speed up a number of queries in SQL that seem to be the source of some very slow response times in our applications. The queries use a number of select statements of the form: SELECT rowid, cset, cacc, serialcolumn FROM gca WHERE ((CSET = "1") AND (CACC >= " ")) OR ((CSET > "1")) ORDER BY cset, cacc We have non-unique indexes on cset, cacc and so I beleive that this index already is in the correct order we are asking for. However, this type of query seems to take a long time (>20 seconds for 15,000 rows in gca) and I am seeing my cpu usage going up to 100% while this query runs, while memory usage and i/o are low. I have a suspicion that Informix is sorting the result set even though the index would return rows in the correct ordering, and that is why I am seeing high cpu usage. The output of explain says that it is using the index to satisfy all the parts of the query, but also that it is using a temp table to do the order by. This adds to my suspicion about a sort taking place. so, my questions are: Shouldn't Informix not need to do a sort, if the rows returned in index order are the correct order for the ORDER BY? If indeed Informix is performing this sort, would that peg my cpu like I am observing? Is there a way to rewrite or otherwise optimize this type of query? Thanks in advance, Gerry Tynen