| 1 |
Quoting Andrea Arteaga <andyspiros@×××××.com>: |
| 2 |
|
| 3 |
> For the interested on benchmarks, I wrote a proposal for writing a new |
| 4 |
> version of the BTL. You can read it on my blog: http://wp.me/pzWEm-99 |
| 5 |
> |
| 6 |
> The BTL is a benchmarking library written in C++ and maintained (now) |
| 7 |
> by the project Eigen. In my opinion it needs some changes, and the |
| 8 |
> article describes why and which ones. I'm waiting for feedback on the |
| 9 |
> proposal before coding anything. |
| 10 |
> |
| 11 |
There is a related area that would be of interest to me. It can be put under |
| 12 |
benchmarks as well. So far, you have been looking into performance, I |
| 13 |
am curious |
| 14 |
about conformance and possibly precision benchmarking. |
| 15 |
|
| 16 |
Does the library behave as it should. What is the accuracy of the result, do |
| 17 |
some version of the libraries give more precise results than another |
| 18 |
implementation. |
| 19 |
|
| 20 |
This is particularly important to me because of the discussion we |
| 21 |
generate here: |
| 22 |
https://github.com/cschwan/sage-on-gentoo/issues/3 |
| 23 |
and |
| 24 |
http://trac.sagemath.org/sage_trac/ticket/10508 |
| 25 |
where it is clear to me that it is due to something that gives a different |
| 26 |
result in cblas from ATLAS-3.9 and it would be good to test this formally. |
| 27 |
|
| 28 |
Francois |
Archives