Different group counts after trim.seqs w/1 vs. 2 processors

mothur bugs
1

I am working with a dataset with a .fasta and .qual extracted from the original .sff file. I just upgraded to a new computer and have the latest version of mothur for 64bit Windows.

When I run trim.seqs with 1 processor, I get significantly more sequences in all groups than when I run it with 2. Is this a mothur error?

Commands are:
trim.seqs(fasta=file.fasta, oligos=file.txt, qfile=file.qual, maxambig=0, maxhomop=8, flip=T, bdiffs=0, pdiffs=2, qwindowaverage=35, qwindowsize=50, minlength=250, processors=1)
vs.
trim.seqs(fasta=file.fasta, oligos=file.txt, qfile=file.qual, maxambig=0, maxhomop=8, flip=T, bdiffs=0, pdiffs=2, qwindowaverage=35, qwindowsize=50, minlength=250, processors=2)

Example group counts:
A with 1 (2393) vs 2 (1413) processors
B with 1 (625) vs 2 (345) processors


Thanks in advance all! ~Kim
2

Thanks for bringing this to our attention. I have fixed the issue and released a new Windows version this afternoon.