Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

From Wikipedia:

an n-simplex is an n-dimensional polytope which is the convex hull of its n + 1 vertices.

I was wondering if the definition is equivalent to say a simplex is synonym of a convex polytope?

Is simplex defined only for $\mathbb{R}^n$, not for other more general spaces?

Thanks and regards!

share|cite|improve this question
up vote 9 down vote accepted

A simplex (in $\mathbb{R}^{n}$) is a special case of a convex polytope, namely one with exactly $n+1$ extremal points (the convex hull of $n+1$ points in general position - that is to say they don't lie in a $(n-1)$-dimensional affine subspace). An example of a convex polytope which isn't a simplex is given by any one of the platonic solids except the tetrahedron.

Yes, there are more general notions of simplices. In a geodesic metric space, for example, it makes sense to speak of convex hulls. In hyperbolic $n$-space or on the $n$-dimensional sphere one can then define a hyperbolic or spherical simplex as the convex hull $(n+1)$ points that don't lie in a totally geodesic submanifold (and that are not too far apart from each other in the case of a sphere).

share|cite|improve this answer
Thanks! I remember that convex hull makes sense for a subset of a vector space with ordered base field. Isn't it? Is a geodesic metric space such vector space? If not, how does it make sense to speak of convex hulls there? – Tim May 1 '11 at 17:15
@Tim: In connection with vector spaces I've only seen convex hulls for real or complex vector spaces. A geodesic metric space is a metric space $(X,d)$ with the following property: For any two points $x,y$ there exists an isometry $\gamma: [0,d(x,y)] \to X$ such that $\gamma(0) = x$ and $\gamma(d(x,y)) = y$ (such an isometry is interpreted as a geodesic parameterized by arc length). A convex set $C$ is then a set such that for all $x,y \in C$ we have $\gamma([0,d(x,y)]) \subset C$ for all geodesics $\gamma$. The convex hull of $S$ is then the intersection of all convex sets containing $S$. – t.b. May 1 '11 at 17:22
Thanks! So are these two unrelated definitions of convex hull? – Tim May 1 '11 at 17:27
@Tim: No. For $\mathbb{R}^n$ they are equivalent. – t.b. May 1 '11 at 17:28

The definition is a little unclear. In order for the convex hull of $n+1$ vertices in $\mathbb{R}^n$ to actually be $n$-dimensional, the vertices must be in general position: that is, they should not lie in a proper affine subspace of $\mathbb{R}^n$. For example, a square is a convex polytope which is the convex hull of $4$ vertices, but it is not a $3$-simplex: a $3$-simplex is a tetrahedron.

I suppose the answer to the last question is "no": when people talk about geometric simplices they are definitely referring to the ones in $\mathbb{R}^n$, although some people talk about abstract simplices in the sense of abstract simplicial complexes.

share|cite|improve this answer
Thanks! (1) So can I say a simplex is defined as a convex polytope whose vertices do not liey in a proper affine subspace of $\mathbb{R}^n$? (2) is it defined for more general space than $\mathbb{R}^n$? – Tim May 1 '11 at 16:39
@Qiaochu: The definition isn't incomplete (though it could be a bit clearer); I think the fact that they are in general position is covered by the statement that the convex hull is in fact an $n$-dimensional polytope. – Zev Chonoles May 1 '11 at 16:39
1) You need the condition that there are $n+1$ vertices, but yes. 2) A more general space than $\mathbb{R}^n$ will generally not have a notion of "convex hull." – Qiaochu Yuan May 1 '11 at 16:40
@Zev: okay, that's fair. – Qiaochu Yuan May 1 '11 at 16:41
@Qiaochu: Here is a book that discusses the abstract notion of a convex hull.… – Jay May 1 '11 at 23:57

I often work over simplices in $\mathbb{F}_q^d$ or other discrete spaces. Here, when we say a $k$-simplex, we mean a set of $(k+1)$-points spanning a $k$-dimensional subspace. I.e., for us, a $k$-simplex is just the set of vertices.

share|cite|improve this answer
Thanks! What is $\mathbb{F}_q^d$? – Tim May 1 '11 at 17:25
Sorry. It is the $d$-dimensional vector space over the finite field with $q$ elements. As Qiaochu noted, there is no analog of "convex hull" in $\mathbb{F}_q^d$, which is why we simply refer to the simplex as the set of vertices. – JavaMan May 1 '11 at 17:28

No, they are not equivalent. For example, an $n$-cube is a convex polytope but is not a simplex.

share|cite|improve this answer
Thanks! Why is the example not? What is the right definition? – Tim May 1 '11 at 16:40
It has too many vertices. – Qiaochu Yuan May 1 '11 at 16:41
@Tim Notice that every vertex in a simplex shares an edge (a 1-simplex) with every other vertex. – yasmar May 1 '11 at 17:08

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.