Article: Parallelism at HotPar 2010
By: Mark Roulo (nothanks.delete@this.xxx.com), August 4, 2010 9:59 am
Room: Moderated Discussions
Dean Kent (dkent@realworldtech.com) on 8/4/10 wrote:
---------------------------
>David proposes a theory (or hypothesis) that 100x claims are crap. He supports
>this theory by using examples of the claim (or claims) he is aware of. Someone
>disputes his claim... but not by giving an example, but by stating that David's
>theory is crap unless he can show that virtually every published work is incorrect.
I think it is more that David proposes that most 100x claims are bogus and that the default assumption should thus be that unless the paper authors explain the source of the 100x claim we should assume that the 100x claim in *NOT* against well optimized, SSE'd, multi-threaded CPU code (ie, code designed to get the most out of the provided CPU silicon).
I think that AM's position is that each 100x claim should be given the benefit of the doubt and assumed valid until explicitly disproven.
-Mark Roulo
---------------------------
>David proposes a theory (or hypothesis) that 100x claims are crap. He supports
>this theory by using examples of the claim (or claims) he is aware of. Someone
>disputes his claim... but not by giving an example, but by stating that David's
>theory is crap unless he can show that virtually every published work is incorrect.
I think it is more that David proposes that most 100x claims are bogus and that the default assumption should thus be that unless the paper authors explain the source of the 100x claim we should assume that the 100x claim in *NOT* against well optimized, SSE'd, multi-threaded CPU code (ie, code designed to get the most out of the provided CPU silicon).
I think that AM's position is that each 100x claim should be given the benefit of the doubt and assumed valid until explicitly disproven.
-Mark Roulo