I have a basic question on Voevodsky's stable homotopy category of spectra $\mathbf{SH}(S)$, where $S$ is a finite dimensional noetherian scheme.

Let $E$ be an $\Omega$-spectrum and $\varphi \colon F\to E$ be a morphism of spectra satisfying that for any smooth $S$-scheme $X$ the induced map $$ \mathrm{Hom}_{\mathbf{SH}(S)}(X,F)\longrightarrow \mathrm{Hom}_{\mathbf{SH}(S)}(X,E) $$ is an isomorphism of groups. The question is: Is $\varphi\colon F\to E$ an isomorphism of spectra?

  • 3
    $\begingroup$ If you mean induces a $\pi_{*,*}$ isomorphism then I believe this is correct, because the collection $\{ \Sigma^{p,q}X \mid X \in Sm/S,p,q \in \mathbb{Z} \}$ is a collection of compact generators of $SH(S)$. $\endgroup$ – Drew Heard Aug 7 '18 at 10:54

Here is a slightly more fleshed out version of the comment above.

First, the claim that the collection $\mathcal{C} = \{ \Sigma^{p,q} U \mid U \in Sm/S, p,q \in \mathbb{Z} \}$ is a collection of compact generators, is Theorem 9.1 of https://arxiv.org/pdf/math/0310190.pdf (there may be earlier references). In particular, the smallest localizing subcategory of $SH(S)$ generated by $\mathcal{C}$ is $SH(S)$ itself.

Now let $\phi \colon F \to E$ be as in your statement, with cofiber $Z$, and assume that the induced map $Hom_{SH(S)}(X,F) \to Hom_{SH(S)}(X,E)$ is a $\pi_{*,*}$ equivalence. It follows that $[X,Z] = 0$ for all $X \in \mathcal{C}$. Since $\mathcal{C}$ generates, this implies that $Z \simeq \ast$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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