This post may sound a bit harsh, please take it as suggestion and not criticism
Developing an engine is great, if your goal is learning or experimentation. You will have to deal with lots of different systems and need to design a scalable architecture as well as good APIs. You will encounter problem after problem, and most of them have been solved one way or another, and the Internet is at your fingertips for solutions.
There are however several downsides to developing your own engine, be it for a game or as a project for others.
- Resources. Developing an engine is -- depending on the scope and the abstraction level of underlying technologies -- a massive endeavour. It often takes several manyears to get to a point, where the engine is feature-rich enough to support different games. But eventually, this highly depends on your scope.
- Don't reinvent the wheel. Popular engines have often been around for many years, and likely provide many of the features that you are looking for. A lot of them are battle-tested by users over time, and have received continuous improvements, bugfixes and optimizations. You may want to check out if what you try to achieve is not readily available.
- Consider your audience. You express that you're interested in the needs and expectations of potential users, and I think this a very good step to clarify before you start implementing. You may not get enough significant answers in the SFML forum alone, I'd suggest you search in separate forums and platforms for people wanting to build games and the problems they encounter, or why they're unhappy with existing engines. Providing features that others don't already have and finding your niche may prove difficult. It's much more realistic if you provide a few tailored standalone features rather than a all-fits-one solution. For example, I wrote a library Thor, which provides modular components that can be used independently.
- Commitment. There have been countless "engine" projects on this forum, and most of them died after a short time and were never widely used. To be taken seriously by the community, you must show that you're willing to bring a project to a stable point, maintain it, engage in community discussions. Ideally you also have something to show (i.e. code -- from this or an earlier project of yours).
If you, against all odds, still want to build an engine: start small, but good (robust code, careful design, reasonable use cases). Try to convince by providing innovative and distinct features, not the 21th animated sprite -- together with SFML, you have quite some headroom (check also the Wiki and GitHub).
Last but not least: instead of spending your entire time on an engine, you may want to invest that into a game and build great gameplay and content. In the end, people who play a game usually don't care if it's powered by engine X or Y. And games also appeal to people who are not SFML developers
Either way, good luck!
