Someone who knows more about photography than me recently mentioned that one should use standard stop based ISO values (100, 200, 400, 800, etc.) He indicated that using a value in between is "not as good." I have never heard of this before and, other than complicating the math by going between stops, it doesn't make sense to me. Is this another gap in my knowledge or is there some logic to it.
It's a characteristic that may apply to some Canon cameras, but not to Nikons.
This statement is generally true but according to my research, the D700 is an exception. See D700 Read Noise Results from Optical Black Data Collection for some details. It may be that more recent Nikon DSLRs will follow the D700 although I know the D7000 behaves more like previous Nikons.
If you made it through then definitely kudos to you!
When that thread was active it seems everyone assumed business as usual for Nikon, in other words, one amplification before the Analog to Digital Converter (ADC). But my data shows that the D700 does employ a two stage amplification, like most Canons. And that comes with a slight read noise penalty at the intermediate ISO values.