> Big-file performance is optimized for UTF-8, although CP437 files > perform nearly as well. I'm able to navigate around a 4 GB test file in > either encoding with no issues. In your example, the "ish" in "textish" > may account for the sluggishness. Good point and very likely the case. The file, being something of an inode dump, contained long long stretches of binary. But still, I'm thrilled that it even worked at all!