Hey, I'm trying to generate a heat map of 30,000 fragments from probably 5-10 samples. Windows complains about memory shortage. Should I resort to Unix system? Also, if I only plot 1000 fragments out, they can finish it rather fast. 5000 would take more than 10 minutes. I don't know what to expect for 30,000... And on the side note, it seems like R only uses up to 50% of CPU while doing number crunching. Is there anyway to utilize 100% of CPU for R? I'm using R 2.6.2 on Windows XP SP2, average config. Thanks. UCLA Neurology Research Lab -- Regards, Anh Tran [[alternative HTML version deleted]]