By: Michael S (already5chosen.delete@this.yahoo.com), August 12, 2014 2:22 am
Room: Moderated Discussions
Aaron Spink (aaronspink.delete@this.notearthlink.net) on August 12, 2014 2:11 am wrote:
> Michael S (already5chosen.delete@this.yahoo.com) on August 12, 2014 1:58 am wrote:
> > Intel used to not just run SPEC on competitor's machines in the lab, but also to
> > submit results to official SPEC database. Last time as recently as last month:
> > http://www.spec.org/cpu2006/results/res2014q3/cpu2006-20140715-30427.html
> >
> > Hopefully, they will do it to AMCC.
> >
>
> Way back when, I had access to Spec acceptance discussion summaries. First time a non-first party
> submitted a result there was a bit of a hullabaloo about it. People were saying things like it wasn't
> fair, etc. Someone then basically said, "well if they think it isn't fair, they can submit a result
> themselves". Pretty much all the disagreement left the conversation at that point.
>
Yes, that's exactly the idea.
Still, it's not going to happen until aarch64 appears in one the major Linux distribution.
Fedora is pretty close, but for purpose of official SPEC submissions (by competitor, no less!) Fedora is probably too experimental.
> Michael S (already5chosen.delete@this.yahoo.com) on August 12, 2014 1:58 am wrote:
> > Intel used to not just run SPEC on competitor's machines in the lab, but also to
> > submit results to official SPEC database. Last time as recently as last month:
> > http://www.spec.org/cpu2006/results/res2014q3/cpu2006-20140715-30427.html
> >
> > Hopefully, they will do it to AMCC.
> >
>
> Way back when, I had access to Spec acceptance discussion summaries. First time a non-first party
> submitted a result there was a bit of a hullabaloo about it. People were saying things like it wasn't
> fair, etc. Someone then basically said, "well if they think it isn't fair, they can submit a result
> themselves". Pretty much all the disagreement left the conversation at that point.
>
Yes, that's exactly the idea.
Still, it's not going to happen until aarch64 appears in one the major Linux distribution.
Fedora is pretty close, but for purpose of official SPEC submissions (by competitor, no less!) Fedora is probably too experimental.