Vanilla pieces

Most chess or chess variant pieces are only defined by their way of moving and capturing, some other variant pieces tend to get special properties like freezing, immunity and transparency abilities.

Here I'm trying to come up with a definition of the first category. Somewhere I had seen that somebody used to call these pieces "vanilla pieces", as to distinguish them from special pieces. I can't find that page anymore but it seems to be a reasonable terminology.

Vanilla pieces are chess or chess variant pieces, that have no other property or ability than a way of moving, and capturing of enemy pieces.
A way of moving and capturing means a fixed set of vectors along which the piece can step, slide or leap from it's current square to it's new square.

If pieces conform to this definition, but their way of moving deviates from their way of capturing, I'll call them divergent vanilla pieces.

I'm asking myself if this definition is going to hold for long, or if I'm going to find cases where this definition does not coincide with what I intuitively see as vanilla.
Comments are welcome.

Examples
The FIDE King and pawn are not vanilla. Royalty, castle, en passant and promotion are special properties and abilities.
The FIDE Queen, Rook, Knight and Bishop are definitely vanilla pieces.
Compounds like R + N, B + N are vanilla. I think in general, any piece defined as the compound of two other vanilla pieces, is a vanilla piece.
The Bulldog Hunter is a divergent vanilla piece.
The Chu Shogi Lion is not vanilla, because of the odd trading rules. But I'm not sure if I want to call the Lion vanilla if we take away the trading rules.
The fairy pieces from Musketeer chess are all vanilla. But if I'm mistaken about this, I'm sure somebody will respond here.

W.r.t. divergent vanilla pieces, I realize that we can think of two extreme categories:
- Non-capturers: a piece can move but not capture.
- Non-movers: a piece can capture but not move.
By heart I don't know if non-capturers and non-movers are already part of an existing variant. But undoubtedly they classify as divergent vanilla pieces.

Interesting post. I use almost exclusively "vanilla" pieces in my games, most probably because I'm conscious that opening them to "non-vanilla" would drive me too far. I'm a bit afraid to look there, it's an infinite world, multi-dimensional. V.R.Parton was less shy than me and has explored this category a lot, you should look at it, it's interesting.

I have 2 questions: 1) do you consider xiangqi cannon a vanilla or not? Or janggi cannon? 2) why "vanilla" ? This escape by knowledge of English as a foreign language.

Finally, I hope this thread will not be hickjacked as too many on this site. The mods should do something...

Somebody was using this term somewhere else for 'normal chess pieces', but I forgot who and can't find anymore where exactly this was used.

When I was into sudoku designing, 'vanilla sudoku' meant traditional 9x9 sudoku as opposed to sudoku variants. At least for most people in that community.

Google translate confirms that 'vanilla' can be used in this way.

W.r.t. the Janggi canon, I read that they can move and capture diagonally in the palace. For me that is beyond vanilla.

Thanks for mentioning the Xiangqi Canon and I was already thinking about it. Intuitively I would call this a divergent vanilla piece. But I would like to rethink the wording of my definition, before I feel confident that it also covers the Xiangqi Canon.

Thank you Evert. By XQ Cannon or Janggi Cannon, I was rather thinking to pieces moving like that but on a non-marked board. If I rephrase my question, what about a Shako-cannon, and what about a janggi-cannon that would move on a regular board, ie, needing to jump a screen both to capture and move?

Thank you

I'm going to give the same answer as I gave for the Xiangqi Canon.

I usually refer to pieces without side effects as 'regular chess pieces'.

But I would not hold it against a piece if it can promote.   And the definition assumes a board with a 'translation invariant' topology (with board edges implemented as inaccessible squares beyond them). With irregular boards (like Janggi) it becomes more tricky. But I would blame the diagonal Palace moves on the board topology, rather than the piece. So I would consider the Janggi Rook regular. For the Janggi Cannon there still is the problem that Cannon trading or jumping is not allowed.

I also have the inclination to only judge the 'pseudo-legal moves' of a piece. Royalty or anti-trading rules I see more as global game rules that cause some of the pseudo-legal moves to not be actually legal in a given position.

Ideally the legality of a move would only depend on the relative board step, and the occupant of the target square. But only leapers satisfy that. Moves of sliders and hoppers (as well as lame leapers) have a 'path', and the validity of the move can depend on the occupancy of all squares in that path, not just the final destination. It is a bit of a matter of taste whether the conditions on this occupancy should be 'color blind' (which is obviously not the case for the destination itself), or 'type blind'.

W.r.t. promotion, almost all Shogi variant pieces promote, and I would not like to say that hardly any of the Shogi variant pieces is regular/vanilla.

Jean Louiscazaux, The use of the word "vanilla" to describe game concepts has always annoyed me, too, since it has no relation to the flavor vanilla, the most expensive spice in the word. 

I think that the etymology of this use of vanilla recently arose as a mondegreen or corruption of the word "banal" or "banales", which means unoriginal, commonplace, or boring. In correct English, the author would probably be describing basic pieces as "banales pieces", which would sound like "vanilla" to someone doesn't know the more obscure word. 

To make matters worse, unlike the French, English speaking countries make no effort to preserve the integrity of the language. Each year, more internet slang and incorrect usages find their way into official dictionaries. It makes English all the more confounding to learn as a second language, and it is going to make it difficult for our students to understand material written even a few generations ago. 

I usually employ the term 'unorthodox piece' for a piece not occurring in FIDE Chess.

I think using the term 'vanilla' for 'ordinary' has its root in 'vanilla' being the most common flavor for icecream.

Perhaps it would be useful to distinguish 'regular chess piece' from 'regular shogi piece': the latter can promote, the former not.

Divergence is pretty uncommon. So capturing and moving the same way could be made part of the conditions for being called regular / vanilla.

Thanks for your answers. 

Just to say, I think safran is much more expensive than vanilla. Or, I'm very rich because I have a nice little stock of good vanilla from Reunion island in my kitchen. And few dust of one stick in my homemade marmalade is wonderful :=)

Indeed, "vanilla" is used in that way (plain, boring, inoffensive) due to it being a default flavor, that nobody bats an eye at. As result of mass-production of low-quality vanilla-flavored goods, much of which contains no actual vanilla, the flavor is not as highly regarded as it once was. 

Saffron is more expensive, by far.