[R] Use of processor by R 32bit on a 64bit machine

[Marc Schwartz]

I suspect that it was Intel's marketing department, after a few beers at the local bar...

;-)

Regards,

Marc

On Jun 29, 2010, at 9:09 AM, Joris Meys wrote:

> *slaps forehead*
> Thanks. So out it goes, that hyperthreading. Who invented
> hyperthreading on a quad-core anyway?
> 
> 
> Cheers
> Joris
> 
> 2010/6/29 Uwe Ligges <ligges at statistik.tu-dortmund.de>:
>> 
>> 
>> On 29.06.2010 15:30, Joris Meys wrote:
>>> 
>>> Dear all,
>>> 
>>> I've recently purchased a new 64bit system with an intel i7 quadcore
>>> processor. As I understood (maybe wrongly) that to date the 32bit
>>> version of R is more stable than the 64bit, I installed the 32bit
>>> version and am happily using it ever since. Now I'm running a whole
>>> lot of models, which goes smoothly, and I thought out of curiosity to
>>> check how much processor I'm using. I would have thought I used 25%
>>> (being one core), as on my old dual core R uses 50% of the total
>>> processor capacity. Funny, it turns out that R is currently using only
>>> 12-13% of my cpu, which is about half of what I expected.
>>> 
>> 
>> An Intel Core i7 Quadcore has 8 virtual cores since it supports
>> hyperthreading. R uses one of these virtual cores. Note that 2 virtual cores
>> won't be twice as fast since they are running on the same physical core.
>> Hence this is expected.
>> 
>> Uwe Ligges
>> 
>> 
>> 
>>> Did I miss something somewhere? Should I change some settings? I'm
>>> running on a Windows 7 enterprise. I looked around already, but I have
>>> the feeling I overlooked something.
>>> 
>>> Cheers
>>> Joris
>>> 
>>> sessionInfo()
>>> R version 2.10.1 (2009-12-14)
>>> i386-pc-mingw32
>>> 
>>> locale:
>>> [1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United
>>> States.1252    LC_MONETARY=English_United States.1252
>>> [4] LC_NUMERIC=C                           LC_TIME=English_United
>>> States.1252
>>> 
>>> attached base packages:
>>> [1] grDevices datasets  splines   graphics  stats     tcltk     utils
>>>    methods   base
>>> 
>>> other attached packages:
>>> [1] svSocket_0.9-48 TinnR_1.0.3     R2HTML_2.0.0    Hmisc_3.7-0
>>> survival_2.35-7
>>> 
>>> loaded via a namespace (and not attached):
>>> [1] cluster_1.12.3 grid_2.10.1    lattice_0.18-3 svMisc_0.9-57
>>>  tools_2.10.1
>>> 
>>> 
>> 
>