Facebook Ads: How do you split test?

I just read "Facebook Advertising" by Perry Marshall. It's not an epiphany, but it's a good read that lays out the basics and some experiences with FB ads; also some good advice on images and text.

The book says that if you want to split test which one of your ads performs best, you can only do that in the power editor and by putting ery single one of the ads in its own ad set (there are 3 levels in the power editor: campaigns, sets and ads).

If you use the ad manager or just create several ads within one ad set, they won't run long enough to objectively determine a winner - FB will soon favour one ad over the others (the best one, but without enough impressions for objective evaluation).

Is that true? Any experiences?

Also, does the ad manager in general work as well as the power editor? The book claims that if you use the ad manager, FB might automatically determine stellar prices for PPC or OPPC, etc...true?

What are your experiences with the ad manager vs. the power editor?

Thanks a lot in advance! FB advertising seems to be an impenetrable jungle...