Thanks a lot. This was useful, we managed to same version of 3.2 on
linux and G5. Intel building would require a bit of work. We will just
stick with one platform and maybe do more extensive cross platform
validation to establish variance between FSL versions.
-Arthur
On Jun 18, 2008, at 6:45 PM, Steve Smith wrote:
> Hi - 3.2 is very old - much older than the existence of intel Macs.
> Any version of 3.2 that you have already got does include all the
> source code so you could copy it and try to fix the build scripts
> and Makefiles to build on Intel, but you will probably find a lot of
> compiler incompatibilities and it will probably be a pain for you to
> get working - it's probably a lot easier to just upgrade your FSL
> version and re-run your analyses, or else exclude the Intel Macs
> from your pipeline for the older analyses.
>
> Cheers, Steve.
>
>
> On 19 Jun 2008, at 04:17, Arthur Mikhno wrote:
>
>> Hi Steve,
>>
>> I am having the same problem as Bryon did a few months ago. We have
>> a pipeline consisting of a Linux, Mac G5 and Intel macs. The Linux
>> and Mac G5 are running FSL 3.2. However, the Intel machines are
>> newer and 3.2 was not available. We want to eventually upgrade all
>> of the them to the latest version but for our current studies we
>> need to have consistency. Is there FSL 3.2 for intel mac's
>> available? or not where could I download the source? and would it
>> compile?
>>
>> Thanks,
>> Arthur
>>
>> On Jul 18, 2007, at 4:27 AM, Steve Smith wrote:
>>
>>> Hi,
>>>
>>> Indeed it's possible that the core algorithms for those programs
>>> didn't change too much between 3.2 and 3.3. Likewise for 4, except
>>> for FDT which will have a major upgrade to allow crossing fibre
>>> modelling for diffusion data.
>>>
>>> However, if you really need the old sources you could email me
>>> directly and I can give you a link to them - though if you
>>> currently have 3.2 then that will include all the sources, even if
>>> it was a binary download! Though you may have problems compiling
>>> very old code with very new gcc, wherever you get 3.2 sources from.
>>>
>>> Cheers, Steve.
>>>
>>>
>>>
>>> On 18 Jul 2007, at 00:08, Bryon Mueller wrote:
>>>
>>>> Hi,
>>>>
>>>> We have processed several hundred subjects using a pipeline that
>>>> uses various FSL3.2 routines (BET, FDT, FLIRT, FAST, AVWMATHS).
>>>> Our computer system will change linux flavors shortly, and we
>>>> don't have the source code for FSL3.2 (just the RH9 binary). The
>>>> only source code I can find on the FSL download page is for
>>>> 3.3.11. Looking over the "What's New in this release" page, it
>>>> seems like BET, FDT, FLIRT, FAST, and the various AVW programs
>>>> have changed little from FSL3.2 to FSL3.3.11.
>>>>
>>>> I am wondering if you expect that we would get the same results
>>>> when data processed under the same pipeline running FC5 and
>>>> FSL3.2 vs CentOS4 and FSL3.3.11? Same question, But with the soon
>>>> to be released FSL4 version? Can one still obtain the FSL3.2
>>>> source code?
>>>>
>>>> Thanks,
>>>>
>>>> Bryon
>>>
>>>
>>> ---------------------------------------------------------------------------
>>> Stephen M. Smith, Professor of Biomedical Engineering
>>> Associate Director, Oxford University FMRIB Centre
>>>
>>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
>>> +44 (0) 1865 222726 (fax 222717)
>>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
>>> ---------------------------------------------------------------------------
>>
>
>
> ---------------------------------------------------------------------------
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director, Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
> +44 (0) 1865 222726 (fax 222717)
> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
> ---------------------------------------------------------------------------
|