Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 23 Oct 2011 13:59:58 -0700
From:      =?windows-1251?B?y/7h7uzo8CDD8Ojj7vDu4g==?= <nm.knife@gmail.com>
To:        FreeBSD <freebsd-questions@freebsd.org>
Subject:   Re: Recommended nVidia card for cuda/opencl on FreeBSD?
Message-ID:  <CAHi1Jscy8qt-V7AEqn6bNPP78JgxiZ3t32iyoSN0NXkO7UgwKw@mail.gmail.com>
In-Reply-To: <20111023162818.09cd24f8@scorpio>
References:  <4EA45EE8.4020409@antonioshome.net> <20111023162818.09cd24f8@scorpio>

next in thread | previous in thread | raw e-mail | index | archive | help
Do you want CUDA 1.0, 1.1, 1.2, 1.3, 2.0, 2.1 compatible? I have a 9800GT
(pretty cheap now-a-days + it runs modern games), which has the lowest CUDA
1.0.

Also, I am interested in how you will do the work. Currently, it's necessary
to run the CUDA SDK and Toolkit under Linux emulation or chroot, despite the
fact that the NVIDIA drivers for FreeBSD include CUDA support. According to
this,
http://blogs.freebsdish.org/jhb/2010/07/20/using-cuda-with-the-native-freebsdamd64-nvidia-driver/,
you still need to compile the CUDA apps under Linux, where the SDK is. Only
after that you can run the binaries on FreeBSD.


-- 
Lyubomir Grigorov (bgalakazam)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAHi1Jscy8qt-V7AEqn6bNPP78JgxiZ3t32iyoSN0NXkO7UgwKw>