Re: [eigen] Eigen Benchmark (number of FLOPS)

[ Thread Index | Date Index | More Archives ]

but note that this benchmark is for sequential runs, the figure would
be quite different if you consider parallel runs.


On Mon, Dec 19, 2011 at 9:59 AM, Gael Guennebaud
<gael.guennebaud@xxxxxxxxx> wrote:
> don't worry, all the implementations included in this benchmark are
> comparable in terms of FLOPS in the sense that they all perform the
> same number of FLOPS for a given operation.
> gael
> On Mon, Dec 19, 2011 at 4:11 AM, Not Sure <kuerzn@xxxxxxxxxxxxxx> wrote:
>> Hi!
>> On the website you
>> show some pretty impressive comparison of Eigen's performance.
>> I have two consecutive questions:
>> 1. Do the different implementations/library of BLAS functions differ in
>> the number of floating point operations (FLOP) they need to complete?
>> 2. If yes: Did you calculate the FLOP per second using some reference
>> FLOP number or the number of actual performed FLOP of each
>> implementation/library?
>> Thank you for your answer,
>> Johannes

Mail converted by MHonArc 2.6.19+