From what I understand, all of the best players in the world display some form of extreme photographic memory capability. Even Fischer had this ability as recounted by anecdotes I read from those who witnessed it in action. I also read in one of my books that GMs memorize on average about 100,000 middlegame positional concepts to the point of them being able to play by intuition built from that memory. This is why they can play bullet/blitz games so well.
Another important aspect is age at when you being chess study. It's been known that the brain will actually adapt itself to chess thinking many times faster in a child than it will an adult. So what takes a child two years of study to gain in skill may take an adult ten years to accomplish.
All that aside, it also takes an extreme obsession with lots of hard work. I believe the formula for general accomplishment is something like:
Hard work/dedication + starting as adult over 20 = Expert or possibly FM many years later
Hard work/dedication + starting as a child about 6 or 7 = FM to IM (sometimes even GM)
Hard work/dedication + starting as a child about 6 or 7 + pre-displayed photographic memory = GM to Top GM
Carlsen displayed photographic memory well before he even started playing chess. His father recalls Magnus being able to name off every country and capital in the world and scarcely being older than a toddler at the time. It's my opinion that this is the root of the 'talant' tag given to the better chess players of the world.
I give up. It is a waste of time to discuss this issue here. The confusion regarding terminology is no mere issue of semantics. It is fundamental failure to be attentive to the science of how memory works.
More than a century ago, Alfred Binet tested the hypothesis that photographic memory served strong chess players. Professional psychologists have been engaged in more productive discussions ever since. But, when this concept comes up among chess players, we are back in 1890 with no idea what we are talking about.
http://chessskill.blogspot.com/2008/10/familiar-positions.html mentions a few relevant studies and assesses two positions separated by many years. The two positions have many common elements, but as photographs would hardly be recognizable as siblings.
Recalling countries and capitals is not photographic memory unless it is being "read" on a picture memory.
The way I understand photographic memory is that the person records the memory the same way as taking a picture whitout caring for datails. When the recall is needed the person will bring back the picture and then get the information needed.
In the example of the countries and capitals, If Carlsen saw a list of the same and when he recite them he recall the picture of the list and then read it, then this is photographic memory, but if he learned it in a different way then is something else.