I think the inflation rate has been scenario adjusted already, but I will check again. But the GP thresholds are a good point. Do you have a rough estimate of the number of GPs that these civs would have generated at this point? For 600 AD as well.
I can't provide the best statistics, but 5-8 Great People between 600AD and 1700AD is pretty typical for the civs still alive in 1700AD (a dedicated player can spawn more than 10, with say, China, Byzantium, and Arabia). Most of the starting infrastructure of 1700AD civs seem to have this amount of GP's, through a combination of settled GPs and wonders placed that would usually be GE-rushed.
Meter reset is most noticeable with France and Japan because they essentially need 4-5 Great Artists for their culture UHVs, both of which check before 1700AD and are competed in the 1700AD scenario (although without the actual culture points reflected in the cities, which is a good thing). This means that in the 1700AD start they can pursue fundamentally different early GP strats than are required in 600AD; this feels really good when playing as 1700AD Japan though, as in a 600AD start you'll be running up at 20-30 turns/GP by the time you're able to generate things other than Great Artists.
Using the condition of 1700 Japan as a more thorough examination, the game assumes at least six GP spawned: 1 GM settled in Kyoto, 1 GE used to rush Itsukushima Shrine (there's no way Kagoshima is building that normally), and 4 Renaissance GAs to check the culture UHV. However, Japan will spawn a GP out of Kagoshima in 4 turns or so due to the meter reset + Itsukushima Shrine, and continue to do so at such a rate. That means Japan gets 6 free GPs, and one of those free GPs snowballs your reset GP engine.
Edit: My reply got folded into the quote.