Re: [sage-devel] Re: Runtimes of binary versus from-source

2012-07-10 Thread Jan Groenewald
Hi This difference in runtime is gone now that the binary has been built on the same toolchain as the from-source compile. The binary/PPA runtime is now 12.5 seconds on a laptop CPU (negligibly close to the 11.5 seconds on a desktop CPU). So the runtime slowness seem related to the Ubuntu 10.04 b

[sage-devel] Re: Runtimes of binary versus from-source

2012-07-06 Thread Volker Braun
The next version of ATLAS will have "generic" presets which will be used e.g. if throttling is enabled. This should greatly reduce the risk of mis-tuned ATLAS installs. On Friday, July 6, 2012 12:58:42 PM UTC+1, Dima Pasechnik wrote: > > What is tuned on one host can be mis-tuned on another. >

[sage-devel] Re: Runtimes of binary versus from-source

2012-07-06 Thread Dima Pasechnik
On Friday, 6 July 2012 18:18:54 UTC+8, Volker Braun wrote: > > A 20x speedup seems to be a bit too good to be true. It could be that the > binary has a botchered (mis-tuned) ATLAS install. What is tuned on one host can be mis-tuned on another. Nothing strange about this. Perhaps a binary ins

[sage-devel] Re: Runtimes of binary versus from-source

2012-07-06 Thread Volker Braun
A 20x speedup seems to be a bit too good to be true. It could be that the binary has a botchered (mis-tuned) ATLAS install. I guess its possible to have that kind of speedup, but only with very specific problems that immensely profit from SSE3/4/AVX. On Friday, July 6, 2012 9:50:37 AM UTC+1,